Operations Management Theory and Practice by B. Mahadevan
Operations Management Theory and Practice by B. Mahadevan
third edition
B. Mahadevan
Delhi • Chennai
BRIEF CONTENTS
Preface
2 Operations Strategy
3 Sustainability in Operations
4 Project Management
6 Facilities Location
13 Lean Management
14 Demand Forecasting
15 Operations Planning
16 Resources Planning
18 Operations Scheduling
Preface
IDEAS AT WORK 1.1: NARAYANA HEALTH (NH): A JOURNEY THAT BEGAN WITH WINING THE HEART
OF INDIA
Intangibility
Heterogeneity
Perishability
Technological Developments
Environmental Issues
Summary
Review Questions
Net-Wise Exercises
Suggested Readings
2 Operations Strategy
IDEAS AT WORK 2.1: CAFE COFFEE DAY: A STRATEGY FOR AFFORDABLE LUXURY
IDEAS AT WORK 2.2: MEASURES FOR OPERATIONAL EXCELLENCE: THE CASE OF INDIGO AIRLINES
Product Portfolio
Process Design
Supply Chain
Technology
Capacity
Challenges in WCM
Summary
Review Questions
Problems
Net-Wise Exercises
Suggested Readings
3 Sustainability in Operations
IDEAS AT WORK 3.1: FIRST ENERGY: A SUSTAINABLE TECHNOLOGY TO ADDRESS COOKING NEEDS
IDEAS AT WORK 3.2: REUSE THROUGH RECYCLING: A CORE ELEMENT OF CREATING SUSTAINABLE
OPERATIONS
3.6 Remanufacturing
Mindset Inertia
Summary
Review Questions
Net-Wise Exercises
Suggested Readings
4 Project Management
IDEAS AT WORK 4.1: TERMINAL THREE AT INDIRA GANDHI INTERNATIONAL (IGI) AIRPORT, NEW
DELHI
Resource Levelling
IDEAS AT WORK 4.3: CRITICAL CHAIN: APPLICATION OF THE THEORY OF CONSTRAINTS TO PROJECT
MANAGEMENT
Summary
Review Questions
Problems
Mini Projects
Case Study
Suggested Readings
IDEAS AT WORK 5.1: MOTHER DAIRY: A CASE FOR SUPPLY CHAIN MANAGEMENT?
IDEAS AT WORK 5.2: SUPPLY CHAIN MANAGEMENT SOLUTION FOR INDIAN OIL CORPORATION
LIMITED
Post-process Indices
Process Indices
IDEAS AT WORK 5.3: PROJECT SHAKTI OF HUL: A UNIQUE DISTRIBUTION CHAIN FOR RURAL
MARKETS
Summary
Review Questions
Formula Review
Net-Wise Exercises
Mini Projects
Suggested Readings
6 Facilities Location
IDEAS AT WORK 6.1: AKSHAYA PATRA: THE NATION-WIDE MID-DAY MEAL SCHEME OF ISKCON
Regulatory Issues
Factor Advantages
Market-related Issues
Cost-related Issues
other Issues
IDEAS AT WORK 6.3: SPECIAL ECONOMIC ZONES: A POLICY ANGLE TO LOCATION PLANNING
Summary
Review Questions
Problems
Net-Wise Exercises
Mini Project
Case Study
Suggested Readings
Quick-response Requirements
Supplier Development
Vendor Rating
Cost
Strategic Restructuring
7.8 E-Procurement
Quasi-market Mechanisms
Summary
Review Questions
Case Study
Suggested Readings
IDEAS AT WORK 8.1: CAPACITY EXPANSION PLANS AT INDIAN OIL CORPORATION LIMITED (IOCL)
Waste Elimination
Subcontracting/outsourcing
Summary
Problems
Mini Project
Case Study
Suggested Readings
Volume
Variety
Flow
IDEAS AT WORK 9.2: THE POLYESTER FILAMENT YARN PLANT AT RELIANCE INDUSTRIES LIMITED
Intermittent Flow
Process Layout
Product Layout
Summary
Formula Review
Review Questions
Problems
Net-Wise Exercises
Mini Project
Case Study
Suggested Readings
IDEAS AT WORK 10.2: BLUE GINGER: THE VIETNAMESE RESTAURANT AT TAJ WEST END,
BANGALORE
IDEAS AT WORK 10.3: IMPROVING SERVICE QUALITY THROUGH BETTER COMMUNICATION: THE
CASE OF ARAVIND EYE HOSPITALS
Summary
Formula Review
Review Questions
Problems
Net-Wise Exercises
Mini Projects
Case Study
Suggested Readings
Summary
Review Questions
Problems
Suggested Readings
IDEAS AT WORK 11.2: CROSS-BADGING: A TOOL FOR NEW RAPID AND NEW PRODUCT
INTRODUCTION
IDEAS AT WORK 11.3: GETTING TOGETHER TO KNOW WHAT THE CUSTOMER WANTS
Value Engineering
IDEAS AT WORK 11.4: VARIETY REDUCTION EXERCISE: OPPORTUNITIES FOR COST CUTTING
11.6 Performance Measures for the Product Development Process
Summary
Review Questions
Net-Wise Exercises
Case Study
Suggested Readings
Joseph M. Juran
Philip B. Crosby
Karou Ishikawa
Shigeo Shingo
Genichi Taguchi
Conformance to Specifications
Employee Involvement
Histograms
Pareto Diagrams
CEDAC
Poka Yoke
IDEAS AT WORK 12.3: MATRIX DIAGRAM FOR ENHANCING THE COMPETITIVENESS OF A PRODUCT
Quality Costing
Quality Awards
Quality Certifications
IDEAS AT WORK 12.4: SRF’S TQM JOURNEY TOWARDS THE DEMING PRIZE
Summary
Review Questions
Problems
Net-Wise Exercises
Mini Projects
Case Study
Suggested Readings
13 Lean Management
Lot-size Reduction
Quality Circles
Summary
Review Questions
Problems
Mini Projects
Case Study
Suggested Readings
14 Demand Forecasting
Short-term Forecasting
Medium-term Forecasting
Long-term Forecasting
Sales-force Estimates
B2B Portals/Marketplaces
Subjective Knowledge
IDEAS AT WORK 14.2: A FORECASTING MODEL TO IMPROVE CREDIT CARD BUSINESS PERFORMANCE
AT COMMERCE BANK
Moving Averages
Econometric Models
Getting Started
Focus Forecasting
Summary
Formula Review
Review Questions
Problems
Net-Wise Exercises
Mini Project
Case Study
Suggested Readings
15 Operations Planning
Reservation of Capacity
Influencing Demand
Inventory-based Alternatives
Level Strategy
Chase Strategy
Mixed Strategy
Summary
Review Questions
Problems
Case Study
Suggested Readings
Summary
Review Questions
Problems
Suggested Readings
16 Resources Planning
Product Structure
IDEAS AT WORK 16.2: OPEN SOURCE ERP SOLUTION FOR VENKATESWARA HATCHERIES
IDEAS AT WORK 16.3: SERVICES RESOURCES PLANNING (SRP): AN ERP SOLUTION FROM RAMCO
Summary
Review Questions
Problems
Net-Wise Exercises
Mini Projects
Suggested Readings
Continuous Demand
Uncertainty of Demand
Seasonal Inventory
Decoupling Inventory
Cyclic Inventory
Pipeline Inventory
Safety Stock
Inventory-carrying Cost
Cost of ordering
Cost of Shortages
ABC Classification
Summary
Formula Review
Review Questions
Problems
Mini Projects
Case Study
Suggested Readings
18 Operations Scheduling
Planning-related Terms
Administration-related Terms
Scheduling Rules
Performance Criteria
Johnson’s Rule
Machine Redeployment
Measures of Performance
Summary
Formula Review
Review Questions
Problems
Mini Projects
Suggested Readings
Common Causes
Assignable Causes
Process Capability
Potential Capability
Actual Capability
Summary
Formula Review
Review Questions
Problems
Net-Wise Exercises
Mini Project
Suggested Readings
To
my parents, Balu and Saroja,
my wife Sujatha and
my daughter Dhrithi
PREFACE
OBJECTIVES
Operations Management: Theory and Practice is the successful culmination of evolved ideas
and clarity of thought arising out of teaching the subject at the Indian Institute of Management
Bangalore for nearly 22 years. This book is the outcome of continuous testing of alternative
ideas, concepts and pedagogical designs with MBA students, working executives from diverse
industries, and research scholars.
Although there are several books available on the subject written by international authors,
students find it difficult to relate to the examples used in them. The basic concept of this book is
to bridge this critical gap by preserving all the salient features one usually finds in international
textbooks, and at the same time, to enrich the book with contextually relevant examples.
Throughout my teaching career, I have discovered that contextual references can go a long way
in helping the students relate to the concepts discussed. Furthermore, such references can fuel
their imagination and improve their understanding of concepts. Therefore, the examples and the
Ideas at Work boxes in this book draw the students’ attention to the issues faced by Indian
organizations while applying the concepts discussed in the book. They also provide insights on
the variations adopted by such firms in the practical application of these concepts. Since several
firms and their product–service offerings are referenced throughout this book, I have provided an
index of companies at the end of the book.
Throughout the book, I have made an effort to provide a pleasant experience of going through
seemingly tough models and concepts in operations management. I hope the readers will enjoy
reading the book, and I look forward to receiving comments and suggestions from the students
and teachers using this book at [email protected].
ORGANIZATION
Operations Management: Theory and Practice (3e) comprises 19 chapters, which have been
divided into four sections.
Part I: Conceptualizing Sustainable Operations consists of four chapters that introduce how to
understand the operations function, the role of strategy in creating successful operations in firms,
the need for making it sustainable and the manner by which such firms eventually become a
physical reality to conduct their operations in the real life.
Part II: Operations and the Value Chain deals with the important issue of supply chain
management in three chapters. The topics covered in these chapters include supply chain
management, facilities, location and sourcing and supply management.
Part III: Designing Operations consists of six chapters that deal with various elements of the
design of an operations system. This includes process and capacity analysis, design of
manufacturing processes, design of service systems, product development and total quality
management.
Part IV: Planning and Control of Operations consists of six chapters in which alternative
approaches for the planning and control of operations are discussed.
PEDAGOGICAL FEATURES
To enable students to understand the concepts discussed in the chapters, the following
pedagogical features have been incorporated into each chapter:
Video Insights
These are provided in each chapter to help students understand the concepts better. Document containing the URLs can
be downloaded from the resource website.
This novel feature in each chapter directly relates chapter concepts to real-world practices in the Indian scenario and
enables students to develop an application-oriented approach.
Problems
The quantitative problems at the end of each chapter enable students to deepen their understanding of the concepts and
models of Operations Management.
Mini Projects
To help instructors guide students through an applied learning process, the chapters provide Mini Projects. These ideas
enable students to broaden their understanding of the subject through extended work.
Net-wise Exercises
Through Net-wise Exercises, students are exposed to online data so that they can access a wealth of information on
various additional topics related to the chapter and also apply the chapter concepts to this data.
Cases
End-of-chapter case studies provide insights into the operations management practices of organizations in the service
sector as well as the manufacturing sector.
Supplements
End-of-chapter supplements help students understand complex topics such as simulation and linear programming.
Both Instructor and student resources are available on the book’s companion Website,
www.pearsoned.co.in/bmahadevan
ACKNOWLEDGEMENTS
The role played by students, executives, research scholars and my colleagues at IIM Bangalore in
shaping my thinking is invaluable and beyond verbal quantification. Over the years, they have
constantly challenged my thinking and provided me with new perspectives on the subject. Many
industry participants and MBA students at IIM Bangalore have worked with me on several live
operations management problems during the past ten years. These problems invariably
culminated in some written project reports. I have drawn many examples from these reports
while discussing several concepts in the book. I am thankful to all of them for their contributions.
I am thankful to several Indian companies that have permitted me to use some of their material
in the book.
I greatly appreciate the assistance provided by Neeraj Bhalla, Associate Editor—Acquisitions
(B&E), Purushothaman, Production Editor— Higher Education, and his team at Pearson
Education for their professional approach and meticulousness that have significantly contributed
to improve the quality of this edition, and I am thankful to these people for their dedication and
support in this project.
As always, my wife Sujatha and my daughter Dhrithi provided me with moral support
throughout the development of this project.
B. Mahadevan
ABOUT THE AUTHOR
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
The success of Narayana Health (NH) is mainly due to the proper implementation of operations management in its day-to-day
activities. The growth of NH to 14 multi-speciality hospitals is mainly because of its planning, passion, and compassion; this is
what operations management is all about. Managing a workforce of 12,000 is mainly attributed mainly to strategic thinking and
implementation. The success rate is made possible by employing cost reduction in several methods. The main vision of the NH
group is not to lose a single patient in terms of want of money and conducting the surgeries in a cost-effective manner.
Source: Franck Boston. Shutterstock
Narayana Health (NH): A Journey That Began with Wining the Heart Of India
Indians are three times prone to heart diseases than Europeans. By 2013, while there is a need
for 2.5 million heart surgeries per year in India, we are doing only 90,000. Moreover, India
produces 80 cardiologists a year compared to 800 in the US. There is a huge shortage of
doctors and paramedical staff in the country. To complicate the matter even more, 70%
doctors are in urban areas, whereas 70% people live in rural areas. This appears to be a
serious problem of lack of adequate health-care for the citizens. However, a closer look at the
story of Narayana Health (NH) suggests that mass healthcare cover is not about money but
about planning, passion, and compassion. Operations Management can work at its best to
address such problems.
NH journey first started with a 300-bed hospital started by Dr. Devi Shetty in the outskirts
of Banga-lore city in 2001 that began providing heart surgeries to patients at significantly
low costs. Currently, the group has 14 hospitals in multiple locations with multiple
specialties that offer a total of 5,700 beds. Over the next 18 months, there are plans to add
new hospitals in Siliguri, Bhubaneswar, Mysore, Mumbai, and Delhi in addition to Cayman
Islands and Malaysia, which will add another 2,500 beds capacity to the NH network. There
are plans to expand this to about 30,000 beds by 2020. On account of this, the work force is
likely to grow from the current strength of 12,000 to nearly 40,000. Managing such a massive
expansion requires sound strategic thinking, creation of new facilities in a cost-effective
manner, and operating them on a day-to-day basis with better planning. All these issues
belong to the domain of Operations Management.
Many of the operational aspects of the NH network are governed by the vision that nobody
should go back from the hospital for want of money. A heart surgery package at NH could be
anywhere between ₹75,000 and ₹150,000 compared to a typical cost of ₹300,000 in other
hospitals. Achieving such dramatic cost reduction calls for employing several methods.
These include input cost reduction, benefiting from process innovations and scale economies,
and bringing innovative ideas in design of facilities and resource deployment. Further,
productivity improvements would be very crucial and an ongoing activity in such a system.
In NH, we find several examples illustrat ing these.
Due to volumes and supplier development efforts, NH is able to buy a pair of glove at
₹4.50 as against ₹9.50 by directly sourcing it from Malaysia. Many of the expensive
equipment such as MRI scan machine are installed by vendors on a pay-per-use model rather
than an outright purchase. This brings down the cost of an MRI scan and also increases its
utilization. Hospitals are located on the outskirts of the cities to bring down the capital costs.
The Mysore cardiac hospital was built using pre-fabricated materials and has no air-
conditioning, except in operation theatres, thereby bringing down the cost of the facility
dramatically low.
Another strategy to bring down the cost is to increase the volume of patient flow. To
achieve this, NH has diversified into multi-specialties (such as Eye care, Orthopaedics,
Neuroscience, and so on). Furthermore, NH recruits the best doctors and pays them as per the
best industry standards. However, they are convinced to work a little extra. For the doctors at
NH, 10 years ago, the plan was to do five surgeries a day with a 150-bed capacity, which
itself was felt high. However, currently with 5,700 beds, they perform 30 surgeries a day. As
a result, nearly 10–12% of all cardiac surgeries performed in India is done in NH network.
Narayana Health is an example of a service system. Like NH, every other service
organizations as well as manufacturing organizations face similar set of issues, although in
varying degrees. As illustrated in the case of NH, these issues can be addressed by applying
several tools and techniques, collectively known as Operations Management.
Source: Based on Babu, V. (2012), “Pulse on the future”, Business World, March 5, 2012, pp 40–45; Chaki, D. (2013), “Straight from the heart”,
Business India, Sep. 30–Oct. 13, 2013, pp 61–63.
Manufacturing, service, and agriculture are the major economic activities in any country. In
India, manufacturing and services together constitute nearly 75 per cent of the gross domestic
product (GDP). In recent years, growth in the GDP has been primarily due to the growth in these
sectors of the economy. Moreover, the share of the service sector in the GDP has grown steadily
from about 40 per cent to over 51 per cent between 1996 and 2006. In view of their contribution
to the country’s GDP, management of manufacturing and service operations are important
economic activities. Significant improvements in productivity and cost savings can be achieved
through operations management—a discipline that focuses on activities that relate to the
planning and control of operations in manufacturing and service organizations. Efficient
operations management can also have a positive impact on the overall health of the economy.
A manufacturing firm essentially engages in converting a variety of inputs into products that
are useful for individuals and organizations. For example, a manufacturer of machine tools
employs several production workers, buys raw material and components from various suppliers,
and manufactures machine tools in a factory using an extensive array of manufacturing facilities.
In this case, the factory encompasses a large number of interrelated conversion processes for the
transformation of raw material into the final product, that is, the machine tool.
A service organization, on the other hand, responds to the requirements of customers and
satisfies their needs through a service delivery process. Service organizations leave an
impression in the minds of their customers through their service delivery. Typical examples of
such organizations include management consultancies, automobile garages, hotels, hospitals and
banks. A service organization may not always make use of material inputs and may not always
produce products that are used by the customer. For instance, a law firm providing legal
consultancy to its clients may not provide material inputs to the system or produce material
output. Instead, the input and output are informational and experiential in nature. However, in the
case of a service system like an automobile garage, a restaurant or a health care system, there are
material inputs and material outputs (in the form of products consumed by the customers, as in
the case of a restaurant). Despite this difference, service systems also have a conversion process
that utilizes resources and delivers useful outputs from the system.
An operations system is defined as one in which several activities are performed to transform
a set of inputs into a useful output using a transformation process. These inputs and outputs can
be tangible, in the form of raw materials and physical products, or intangible, in the form of
information and experiences. Viewed in this manner, manufacturing and service systems could
be broadly classified as operations systems. Operations management is a systematic approach
to addressing issues in the transformation process that converts inputs into useful, revenue-
generating outputs. Four aspects of this definition merit closer attention:
Operations management is a systematic approach to addressing issues in the transformation process that
converts inputs into useful, revenue-generating outputs.
Operations management is a systematic approach. It involves understanding the nature of issues and problems to be
studied; establishing measures of performance; collecting relevant data; using scientific tools, techniques, and solution
methodologies for analysis; and developing effective as well as efficient solutions to the problem at hand. Therefore, for
successful operations management, the focus should be on developing a set of tools and techniques to analyse the
problems faced within an operations system.
Operations management involves addressing various issues that an organization faces. These issues vary markedly in
terms of the time frame, the nature of the problem, and the commitment of the required resources. Simple problems
include deciding how to re-route jobs when a machine breaks down on a shop floor, or how to handle a surge in demand
in a service system. On the other hand, decisions such as where to locate the plant, what capacity to build in the system,
and what types of products and services to offer to the customers require greater commitment of resources and time.
Operations management provides alternative methodologies to address such wide-ranging issues in an organization.
Transformation processes are central to operations systems. The transformation process ensures that inputs are
converted into useful outputs. Therefore, the focus of operations management is to address the design, planning, and
operational control of the transformation process.
The goal of operations management is to ensure that the organization is able to keep costs to a minimum and obtain
revenue in excess of costs through careful planning and control of operations. An appropriate performance evaluation
system is required for this. Therefore, operations management also involves the development of performance evaluation
systems and methods through which the operating system can make improvements to meet targeted performance
measures.
Since manufacturing and service organizations form the operations system, it helps to understand
the prevalent trends and the status of manufacturing and service organizations in India before we
dwell on the various elements of operations management.
The index of industrial production (IIP) is a measure of the growth in the manufacturing sector.
The Centre for Monitoring Indian Economy (CMIE) collects data on several macroeconomic
indicators, including the IIP. Table 1.1 shows some of the salient aspects of the manufacturing
and service sectors of the Indian economy. It is evident from Table 1.1 that the corporate sector
has been going through tough times. There has not been much growth in sales revenue. However,
there has been a steady increase in operating expenses and compensation to the employees. This
has resulted in a situation where the profit after tax (PAT) has been progressively and
significantly shrinking in the last three years. The service sector companies have been
experiencing this more significantly than the manufacturing sector companies as evident from
the table. On the whole, this points to the importance of deploying sound operations management
practices in both manufacturing and service sectors of the Indian economy.
Over the years, there has been an increase in raw-material consumption. This perhaps indicates
that manufacturing organizations are increasingly buying components and semi-processed items
from suppliers. Furthermore, there has been a gradual increase in the cost of material and labour,
which are primary inputs in the manufacturing system. Such an increase in the input cost is likely
to put greater pressure on firms to cut down waste and improve productivity to remain
competitive in the market. Therefore, the focus areas of operations management are likely to be
in the areas of better supplier management, elimination of waste from the system, and
improvement in overall productivity. Several sectors of the industry have been focusing on some
of these initiatives.
TABLE 1.1 Salient Aspects of the Corporate Sector of the Indian Economy
All numbers in the table represent growth % over the previous year.
Data compiled from Economic Outlook of Centre for Monitoring Indian Economy (CMIE) using their time series data available at
http://economicoutlook.cmie.com
Globally, India is emerging as an important manufacturing base and is competing closely with
China in attracting several multinational companies to set up their manufacturing plants. Several
studies point to emerging opportunities for Indian manufacturing to grow and attain a global
presence. India has a unique advantage in the form of abundant low-cost labour and technical
manpower. The Global Manufacturing Competitiveness Index 2013 released by Deloitte rated
India at number four, next to China, Germany, and the US among 38 countries. It predicted India
to reach the second position in the next 5 years. The report suggested that India’s rich talent pool
of scientists, researchers, and engineers as well as its large, well-educated English-speaking
workforce and democratic regime would make it an attractive destination for manufacturers.
Global manufacturing executives increasingly view India as a place where they can design,
develop, and manufacture innovative products for sale in local as well as in global markets.
Therefore, Indian manufacturing firms can, on the one hand, exploit the low-cost advantage as an
entry strategy to capture global markets and, on the other develop, unique capabilities to position
as an active participant in the global value chain.
The examples discussed here underscore the need for use of operations management to remain
competitive in business and tap the emerging opportunities in the global arena. Effective
operations management also requires a greater understanding of the various activities pursued
under its banner and the critical linkages between operations and other aspects of business.
1.3 SERVICES AS A PART OF OPERATIONS MANAGEMENT
The service sector encompasses a wide spectrum of activities in every country. The growth of the
service sector in India during 2002 to 2007 has been very significant. Table 1.2 provides some
details on the share of some sectors of services in the overall GDP. The central government
began taxing three services in 1994–95. This has grown steadily, and as of 2012–13, the number
of services taxed has gone up to 119. During this period, both the number of assessees and the
service tax revenue has been growing very rapidly. According to the Central Board of Excise and
Customs (CBEC), between 1994–95 and 2012–13 the tax revenue has grown from ₹4.1 billion to
₹1.325 trillion.1 These figures indicate the growing importance of services in the Indian
economy and the need to apply management practices to plan and control operations in the
service sector.
Although services and manufacturing are classified as separate sectors in a macroeconomic
sense, from the perspective of operations management, this separation is artificial. In operations
management, a “pure product” and a “pure service” are just two ends of the spectrum, and not
separate entities. In reality, a vast majority of operations share a continuum of services and
products. Therefore, most of the principles, tools and techniques of operations management
apply to both these sectors. This product–service continuum is illustrated in Figure 1.1.
TABLE 1.2 Service Sectors in India: Share of GDP in Percentage
# Advanced Estimate
* Compiled from Chapter 10 on Services Sector in Economic Survey 2012–13, government of India, Ministry of Finance, Economic Division. For more
details see http://indiabudget.nic.in
Intangibility
Fundamentally, services differ from manufacturing with respect to tangibility. Because services
are experiences rather than objects, they cannot be touched, tasted or felt as in the case of
objects. At the most, the recipients of services can form an opinion (based on some personal
assessment) about the quality of the service offered. This has important implications for defining
and assessing the quality of the service.
Fundamentally, services differ from manufacturing with respect to tangibility, heterogeneity, simultaneity, and
perishability.
On the other hand, in a product-oriented operation, the product is defined by certain attributes
and the customer faces less ambiguity with respect to the product, its attributes, and its
performance. This is because the customer can touch and feel the product and make his/her own
assessment of the product.
Heterogeneity
The second differentiating aspect of services is the high degree of heterogeneity associated with
them. Since the experiential component is dominant in a service, it is likely that no two services
are exactly alike. The differences are attributed to the differences in the service receivers
(customers), the service providers, and other parameters of the service delivery system.
Therefore, a dentist attending to two consecutive patients having identical ailments may provide
more or less the same type of service. Nevertheless, the two patients may have different
perceptions of the quality of the service and may have different satisfaction levels. Moreover, the
time spent by the dentist in both the cases could vary greatly. High heterogeneity results in high
variability in the operations system performance and the need to factor them into the planning
and control of operations.
Perishability
Services are perishable. This implies that they cannot be inventoried. Thirty minutes of a
doctor’s consulting expertise today cannot be stored for future use, reused, or returned in a future
period. Similarly, the treatment for an acute headache likely to be faced by a patient sometime in
the future cannot be stored in advance and reused at that time. The possibility of inventorying the
supply and using it at a later time is very common in a manufacturing system. The implication
for operations is that service systems require methods that work without inventories.
Detailed discussions with the officers revealed several opportunities for improving the
existing situation. The activities at the ARPO were labour-intensive, and the organizational
rules and procedures were highly centralized. Further, the discussions revealed that there was
scope for automation of some of the activities. In the present scenario, it took a lot of time to
access information on the application status, and Web-based services could reduce this time.
The uneven load on police stations and the need to modify and simplify certain rules were
some other issues that merited attention. Independent of the study, some administrative and
procedural changes were already underway at the ARPO. Attention to all the issues that were
highlighted during the study through better capacity management could help reduce the lead
time in processing passports.
The data collected during the study was subjected to detailed analysis, and alternative
methods were identified for the design and operational control of the operations at the
ARPO. These include the use of waiting line models to explore alternative ways of
redesigning the service delivery system; process redesigning to eliminate waste;
improvements in productivity and the discovery of hidden capacity in the system; the
identification of a new system for tracking operational performance, benchmarking and
improvement; and the use of workload analysis and capacity planning methodologies. We
shall discuss these aspects of operations management in the forthcoming chapters.
Source: Adapted from N. Ravichandran and D. Bahuguna, “Rule Bound Government Agency to Customer Centric Service Facility: Can Indian
Passport Offices Make The Leap?” IIMB Management Review 18, no. 1 (2006): 59–66.
Despite these differences between products and services, from an operations management
perspective, there are similarities between these two categories. Table 1.3 lists some of the
salient differences and similarities between a manufacturing and a service system. The
differences, as we have already discussed, are mainly due to the four distinguishing features of
services. The similarities, on the other hand, are due to the basic need to manage operations.
Design of operations requires decisions on capacities, products, and services to be offered.
Operational planning and control requires that suppliers are managed, operations properly
planned and scheduled, and demand matched with supply. These issues are common to both
manufacturing and services, though the tools and techniques used to handle these issues may be
specific to manufacturing and services.
TABLE 1.3 A Comparison of Manufacturing and Service Organizations
The last layer, that is, the supplier layer, is external to the organization. Consisting of a network
of suppliers, sub-contractors and other specialized service providers, this layer collectively
represents the inbound logistics of the organization.
The customer layer interacts with the marketing function, and the inputs from the marketing
function are crucial for planning operations. The supplier layer interacts with the materials or
procurement function of an organization. The procurement function, in turn, interacts with the
core operations layer and ensures timely availability of the various materials required. The
customer layer also provides critical inputs to the layer of innovation, which forms the basis for
the strategy formulation process. Based on the strategies made, several changes are made to the
core operations layer. Therefore, there is a critical linkage between the core operations layer and
the innovation layer.
The basic inputs in an operating system are labour, material, and capital.
The basic inputs in an operating system are labour, material and capital. The availability of
skilled labour is an essential requirement as it is not possible to operate the system satisfactorily
and deliver good-quality goods and services without adequate supply. The second input in
operations is material. All manufacturing organizations process raw materials and convert them
into useful products. Materials are required in several service systems as well. A healthcare
system such as a multi-specialty hospital requires medical consumables such as cotton,
disposable syringes, and other equipment. Similarly, a hotel requires a variety of raw material for
preparation of food. Therefore, material forms a critical input in operations. The third important
input in operations is capital. Capital is required to buy resources such as plants and machinery,
to pay labour, and for investing in materials that are being processed into finished goods.
FIGURE 1.5 Operations management: a systems perspective
Processing includes the various activities that an operating system undertakes to convert the
raw material into useful products for customers. The conversion process adds value to the
product and enables the organization to sell it in the market. In the case of service organizations,
the conversion process involves committing the various resources at its disposal to deliver useful
services to the customer. Irrespective of whether it is a manufacturing or a service organization,
there are a set of activities to be performed in the conversion process. These activities are
presented in the processing section of Figure 1.5.
Processing includes the various activities that an operating system undertakes to convert the raw material into
useful products for the customers.
Product and process designs are an integral part of the conversion process. Every operations
system must make important decisions with respect to the type of products and services to be
offered. Furthermore, once the right mix of products and services are identified, appropriate
processes for manufacturing and delivering them to the customers need to be identified. This
involves deciding on the type of technology to use, the type of machines to be used in the
conversion process, and the exact method of creating the products and services. These steps are
known as product and process design. For example, when the low-cost airline Spice Jet started
its operations, it had to make these choices. It chose to fly to only three locations connecting
western India to Delhi. It had to decide on the type of aircraft to use, the in-flight services to
offer, and various other aspects, such as ticketing and reservation, airport logistics, and passenger
interfaces at the airport and outside. In manufacturing, a similar exercise can be visualized for
Hyundai, when it started to manufacture passenger cars in India in 1998.
The best way to understand the need for operations management and the issues to be addressed in a manufacturing
system is to make a factory visit. A factory visit gives us a bird’s eye view of various aspects of managing the
operations. The Honda factory in the USA and Mahindra’s XUV 500 factory in India provide a good introduction to the
concepts of operations management in manufacturing organizations. To know more about this, find the video links
(Video Insights) in the Instructor Resources or Student Resources under section Downloadable Resources
(http://www.pearsoned.co.in/BMahadevan/) subsections Bonus Material.
Once the product and process designs are finalized, an operations system must focus on ensuring
that the demand for products and services is met. At this stage, the organization may also require
operations planning to ensure the availability of adequate material and capacity to meet the
targeted production and service delivery. This is known as operations planning. The operations
system also requires methods for real-time control of operations as there may be several
deviations from the plan. For example, there might be a breakdown of machinery or rejection of
a lot of material supplied by a vendor. Alternatively, some customer might change his/her order
quantity or cancel the order altogether. The operations control activity takes care of these
changing requirements in an operations system. In the airline example, the issues relate to the
pricing of tickets, capacity planning, and the handling of tickets cancellations and schedule
delays. A planning system will help the airline in route planning, rerouting aircraft, and short-
term capacity management.
Another important activity in the conversion process pertains to ensuring adequate supply of
materials for the operations system. This requires that the suppliers of various materials are
identified and relationships established with them. With such an arrangement, it will be possible
to place orders for material with the suppliers and receive them within the stipulated time. This
activity is labelled as purchasing and inventory control in Figure 1.5. From an operations
management perspective, the processing part of the system primarily focuses on identifying the
nature of resources required, planning for material and capacity, and ensuring that changes in
real time are addressed.
A planning system will help the airline in route planning, rerouting aircraft, and short-term capacity
management.
A fast-food restaurant may provide various types of food for breakfast, as per its process and product design. In
several manufacturing organizations, services are also offered in the form of after-sales support and warranty.
An important aspect of operations management is estimating demand. This is done through forecasting.
The output of an operations system consists of goods and services. An organization that
manufactures passenger cars will provide many variants of the passenger car. On the other hand,
a fast-food restaurant may provide various types of food for breakfast, as per its process and
product design. In several manufacturing organizations, services are also offered in the form of
after-sales support and warranty. In addition to these tangible outputs, it is inevitable that an
operations system will leave certain impressions and value judgments in the minds of its
customers and stakeholders. For instance, a manufacturer of copper products such as Hindustan
Copper will not only supply copper products to its customers, but also create an impression in
the mind of the customer. For instance, its customers may feel that though the quality of the
product is good, it is costly in comparison with its competition. People living in the vicinity of its
factory may form opinions about the company on matters related to environmental pollution. The
labour practices adopted by the company are likely to influence its prospective employees.
In recent times, organizations have begun to understand their role as a responsible entity in
society. Pollution of the atmosphere creates greater problems leading to global warming.
Industries such as leather processing and petrochemicals tend to generate large amounts of
effluents. Sustainability is another area of major concern. In a country like India, businesses and
society compete for limited natural resources such as water. In the state of Kerala, the local
society protested against Coca-Cola for drawing water from the locality for their bottling plant.
In such situations, the government and judiciary bodies such as the Supreme Court play an active
role and intervene to ensure that these issues are resolved. To avoid disrupting the balance in
society, an operations system must take such factors into consideration while configuring its
input–processing–output elements.
It is useful to understand how the operations system considers these factors and reacts to them.
In any system, the feedback loop serves the purpose of identifying the deviation paths and
highlighting the areas that need immediate correction. In an operations system, there are three
feedback loops available for making corrective measures. These include quality management,
maintenance management, and process improvement. These operations management activities
provide checkpoints to identify the areas requiring improvement and ensure that corrective
measures are indeed taken.
The demand for the goods and services offered by an organization cannot be entirely
controlled by the organization. At best, an organization can influence the demand to an extent.
From a systems perspective, the demand is an exogenous variable. However, planning for
production, capacity and material clearly depend on the demand that the operations system is
likely to face in the upcoming periods. Therefore, an important aspect of operations management
is to estimate demand. This is done through what is known as forecasting.
From the description of the various aspects of operations management in the systems
perspective, one can distil some of the important functions of operations management. These
functions are carried out irrespective of whether the organization is in manufacturing or services.
Table 1.4 lists these functions.
The important functions can be understood by: (a) categorizing them under design issues and
control issues, as shown in Table 1.4; and (b) classifying them as long-term issues and short-term
issues, based on the planning horizon.
Design issues in operations management establish the overall constraints under which the operations system
functions.
For example, once the capacity of the resources to be used in the system is decided, it sets limits
for the actual use of the system in operation. If Maruti Suzuki India Limited installed an engine
plant that has the capacity to produce 200,000 engines annually, then, in actual operation, it can
produce no more than 200,000 engines. Similarly, if a multi-specialty hospital is located in Pune
with a 200-bed facility, it is understood that it will largely serve residents in the vicinity of Pune
up to a capacity of 200 in-patients at any time. Setting policy guidelines and procedural details
for various aspects of the operations system is also a design activity. A typical example is the
setting up of a quality management system that establishes rules and policies with respect to
what is acceptable and what is not.
Operations management decisions listed in the right-hand column of Table 1.4 share a
different characteristic. They all address the vital issue of planning and operations control. Once
the design choices are exercised, operations management amounts to putting the available
resources to best use and handling various issues as and when they arise. The available capacity
could be better utilized through production planning and by scheduling operations so that idle
time is minimized. Furthermore, required capacity and material could be estimated and made
available through purchasing and scheduling procedures. All these constitute operational control
decisions in operations management.
Every issue addressed in design is inevitably addressed again in operational control. However,
the context differs in the two cases. Capacity planning as a design issue lends itself to a different
type of analysis compared to capacity planning in operational control. In the case of the latter,
the objective is to match the requirement to the available capacity, whereas in the case of the
former, the objective is to find out how much capacity is required for meeting targeted business
plans. Moreover, design issues often turn out to be strategic in nature, while operational control
issues are tactical in nature. Strategic decisions frequently involve large capital outlay and are
made with critical inputs from an operations strategy process. Such decisions are made by the top
management to improve the competitiveness of the organization. Operational control issues, on
the other hand, are tactical, repetitive and routine in nature. Lower-level operations managers
often make such decisions.
Challenges in operations management arise mostly from developments that create a need for
new, and, at times, more efficient system of operations management. The economic reforms in
the 1990s opened up the indian markets to overseas players, which posed new challenges for
domestic companies. With time, there have been more changes in the form of rising customer
expectations, technological developments, and growing awareness about environmental issues.
Each of these developments needs to be addressed with appropriate changes in the strategy,
design and implementation of operations management practices.
Competitive Pressure due to Economic Reforms
It is more than 20 years since the economic liberalization and globalization policies of the
Government of India came into force. During these years, the Indian corporate sector has learnt
what exactly the melting pot of global competition means. Due to integration of the Indian
economy with the global economy, the Indian companies are required to benchmark the
manufacturing sector against the best in the world to enhance their competitiveness. Companies
that are able to meet the stiff competition are able to discover the unprecedented opportunity for
growth as they discover new markets and customers in other parts of the world. Competitive
pressure manifests in terms of three major challenges:
Falling prices
There has been a continuous fall in prices of branded, engineered consumer products as well as
industrial products. For example, the on road price of a Hyundai Santro GLS model in 2000 was
₹440,000/- and the price of a Santro Xing GLS model in April 2014 was about ₹480,000/-. Over
the last 15 years, the annual growth rate in price of several white goods such as refrigerators,
washing machines, and ovens have been far less than inflation rate. Similarly, industrial buyers
constantly enforce annual cost reduction targets for their suppliers. Therefore, learning to
profitably operate in an era of falling prices is a challenge.
Build-to-order requirements
Customers increasingly demand several variations of products and services. They also demand
flexibility to change their requirements until the point of service and product delivery. Systems
that allow them to customize, configure, and visualize their own version of products and services
are more in demand than a standard off-the-shelf version with a few pre-determined variations.
In order to respond to these emerging requirements, a firm need to develop build-to-order
capabilities.
Growing Customer Expectations
Look at the number of tariff plans and options offered by mobile service providers such as Airtel
and Vodafone these days. They provide multiple postpaid and prepaid options that appear to
satisfy every customer category in the market. This is a good indication of the growing customer
expectations and the ability of organizations to respond to them. Such developments have been
greatly facilitated by the economic liberalization process that we discussed earlier. A decade ago,
customers would not have made any fuss about the services provided by the Department of Post
and Telegraphs, simply because they had no other choice. The situation was similar in most
sectors, whether a customer wanted to travel by air, get a phone connection, or buy a car. Today,
however, they have a variety of choices in of these situations. Many courier services can
substitute India Post. Several private airlines and telecom companies offer multiple versions of
services. Many different models of cars, starting from basic models such as the Maruti 800 and
the Omni to high-end cars such as the Toyota Camry are available today.
Exposure to such developments increases customer awareness, enabling them to demand
better-quality products and services. They prefer more variety, greater levels of customization,
and the ability to have it in discrete quantities and without waiting too much for the product or
service they want. This means that manufacturing organizations and service providers constantly
need to invest in understanding changing customer tastes and provide goods and services by
aligning various aspects of the operations. Organizations need to develop the capability to bring
out newer product and service versions that are not only better but also cost-effective. This
introduces new challenges for operations management.
Technological Developments
In the last few tears, technological developments have changed the way businesses in many
sectors operate. Take, for example, the banking sector in our country. With the advent of
automated teller machines (ATMs) and the Internet, the face of banking is undergoing a sea
change. Customers need not visit a bank branch to draw cash or to check their account balance.
These transactions can be carried out at an ATM. Drafts and cheques can be replaced with
electronic payment gateways and fund transfer mechanisms. Similar facilities are now available
in other sectors. For example, consider the process involved in buying a train ticket. By visiting a
Web site like http://www.irctc.co.in, a customer can accomplish all the required tasks pertaining
to ticket booking and cancellation at leisure.
The changes that manufacturing organizations are going through are no less than service
organizations. In this age of electronic markets, a manufacturing organization can procure goods
and services by organizing a reverse auction on the Internet. In a matter of 3 to 4 hours, the best
price for a component and the supplier willing to provide the component at a desired quality can
be located. Similarly, a team of design personnel from across different geographical locations
can participate in new-product development using technological tools.
From these examples, we can appreciate that today’s businesses are constantly challenged by
the rapid technological advancements. These advancements tend to have a dramatic impact on
the competitiveness of businesses. They force structural changes in a sector of the industry and
trigger several changes in the operational practices of firms.
Environmental Issues
Growing industrialization at one level indicates the economic progress of a nation. However, at
another level, it introduces new concerns and challenges in operations. In the recent past, there
has been growing concern regarding the impact of business activities on the environment. These
include concerns regarding the depletion of natural resources and the waste generated from
production systems and end-of-life products. Increasingly, firms are under pressure to take
responsibility of restoring, sustaining, and expanding the planet’s ecosystem instead of merely
exploiting it.
When the Government of India announced a scheme for special economic zones (SEZs), it
generated controversies and social concerns. In several such cases, local communities were
reluctant to part with fertile land, which was then used for industrial purposes. This has
implications for the location of manufacturing and service facilities.
Consider the case of passenger cars. In a country like India, 12 out of 1000 people have a
passenger car. In the United States, this figure is 512 per 1000 people. While this could be an
indicator for the passenger-car industry to set its growth projections, it also indicates the
challenges arising out of availability of petrol and diesel and increased emissions. Unless the
issue of emissions is addressed and alternative fuels and non-conventional technologies are
invented, the sector will not grow to meet its potential. Design of products and operational
practices must take these environmental concerns into consideration.
Growing urbanization creates societal problems arising out of scarcity of available resources
and generation of solid wastes. With high population density in cities, the consumption of energy
and water in countries like India is on the rise. Such a situation requires better practices and
newer methods of addressing these requirements using better operational practices. These days,
due to the increased use of electronic gadgets, the amount of e-waste being generated is also on
the increase. Similarly, the use of plastics and bio-non-degradable items also pose challenges to
businesses and society.
Businesses need to refine their operations management practices to be able to address these
requirements. Operations management practices must address environmental concerns in order to
ensure a sustainable world.
Each of the challenges identified in the previous discussion clearly points to new priorities for
operations management. Economic reforms and liberalization have established the importance of
the customer/market. Organizations need to relate the operations system to the customer/market.
Otherwise, they may find it difficult to survive in the long run. As customer aspirations grow,
organizations must find ways and means of dealing with the proliferation of variety. Operations
management practices need to be developed for this purpose. One way in which organizations
can meet growing customer aspirations is by getting into a continuous learning mode. In the
future, addressing emerging environmental concerns through better operations practices will
become a necessity. Some such priorities for present-day organizations are discussed in some
detail here:
Relate the operations system to customer/market requirements. Choices in manufacturing and services cannot be made
on the basis of internal conveniences, as was the practice in the pre-reforms era. All the choices made in an organization
demand that the customer is central to the design, whether it concerns performance measures, manufacturing and
service-delivery system design, supplier management, or product development. If the customer has difficulty using a
product or service, there should be methods for understanding the difficulty and rectifying the mistakes. This calls for
better quality management practices in every organization.
Acquire capabilities to tolerate product/service proliferation. Customers are likely to demand more choices and value-
offerings from manufacturing and service organizations. In order to remain competitive, every organization needs to
understand customer requirements and incorporate them in new product development initiatives. New procedures and
systems should link the customer layer to other layers of an organization (see Figure 1.4). The development of new
products and models often becomes necessary for organizations. However, large product/service variations often result
in a spectrum of offerings ranging from “low-volume, high-variety” to “high-volume, low-variety.” Operations
management practices need to address these requirements and enable managers to handle the complexities arising out of
variety.
Develop systems and procedures that promote learning. Continuous improvements in factors affecting competition,
such as quality, delivery, cost, and responsiveness, are very important for every organization. These improvements do
not take place in isolation or by mere accident. Instead, improvements happen out of continuous learning on the part of
the employees in an organization. A good operations management practice in an organization will help the organization
set up a continuous improvement methodology. New systems and procedures that help employees to learn need to be put
in place.
Develop green manufacturing practices. Green manufacturing practices are operational practices that ensure that
products are manufactured in a sustainable manner by conserving scarce natural resources and minimizing pollution and
other negative impacts on society and nature.
Environmental imbalances are of great concern as they threaten the society and its sustainability.
Non-governmental organizations (NGOs) and interest groups such as Greenpeace, international
bodies such as the World Trade Organization (WTO), and government and regulatory bodies
impose stringent regulations on organizations to reduce the adverse effects of manufacturing
activities on the environment. Policy makers have brought in new legislations that put regulatory
pressures on businesses as a means to address environmental concerns. Notable among these are
the EU directives on compulsory product-take-back at the end of life of a product, The
Netherlands National Environmental Plan, and the package recycling and product take-back laws
in Germany.2 In Germany, the Commercial and Industrial Waste Avoidance and Management
Act holds producers responsible for end-of-life disposition, recovery, and reuse of their
products.3 Therefore, it is inevitable for every organization to develop green manufacturing
practices.
In an era where human beings are facing serious environmental concerns like global warming
and depletion of natural resources, organizations should reduce the total environmental burden
by moving beyond mere “greening” and take ownership of the environmental impacts associated
with the total life cycle of the product. This essentially calls for fundamental changes in the
underlying product and process design, taking into account all possible effects a product might
have throughout its life on the environment, both within the firm and beyond its boundaries. The
use of clean technologies is another important requirement for maintaining a balanced
environment.
SUMMARY
Operations management is a systematic approach to addressing issues in the transformation process that converts inputs
into useful, revenue-building outputs.
Globally, India is emerging as an important manufacturing base and is closely competing with China in attracting
several multinational companies to set up their manufacturing plants in India. Several recent studies point to emerging
opportunities for Indian manufacturing to grow and attain a global presence.
From an operations management perspective, a “pure product” and a “pure service” are just two ends of a spectrum. In
reality, a vast majority of operations share the continuum of products and services.
Despite several important differences between products and services, from an operations management perspective,
there are several similarities between the two settings.
Every organization has five layers that make up its value chain. The physical arrangement of these five layers depicts
the structural aspects of an operations system.
The decision context in operations management can be broadly classified as design issues and operations control
issues. Moreover, there are long-term and short-term decisions involved in operations management.
Some of the challenges faced by operations firms include the need to address increased competition due to economic
liberalization and globalization, growing aspirations of customers, rapid technological advances, and emerging
environmental concerns.
REVIEW QUESTIONS
1. What is an operations system? give some examples of operations systems.
2. What do you understand by the term “operations management"?
3. What are the major operations management issues that manufacturing organizations face in India?
4. How would you define a service system? give some examples to support your definition.
5. How important is it to apply formal management principles to service operations?
6. “There is nothing like a service system or a manufacturing system. In reality, there is a continuum between these two
extremes.” Comment on this statement.
7. “Services are very different from manufacturing. Therefore, it is not appropriate to use the same set of principles for managing
operations in manufacturing and service organizations.” Do you agree with this statement? give reasons in support of your
argument.
8. What do you understand by the term systems perspective? Use the example of operations management to explain.
9. What is the role of operations in an organization? What are the other functions in an organization? Are these functions
independent of one another? give some examples to support your argument.
10. What are the various functions of operations and how are they linked to other parts of an organization?
11. Distinguish between the following terms:
12. Identify three emerging environmental concerns and their impact on operations management. What changes do you envisage
in operations management to address these concerns?
NET-WISE EXERCISES
1. Visit the following Web sites to understand the current trends and the growth in India’s manufacturing sector:
http://indiabudget.nic.in/budget2013–2014/es2012–13/echap-09.pdf
http://nmcc.nic.in/default.aspx (examine the contents under various heads)
http://nmcc.nic.in/pdf/strategy_paper_0306.pdf (go through the report on a strategy for the manufacturing sector)
a. What is your understanding of the status of the manufacturing sector in the country?
b. What are the key steps that the National Manufacturing Competitiveness Program proposes to improve the
competitiveness of manufac-turing?
2. Visit http://www.servicetax.gov.in/ to know more about the growth of service tax in India. Look up
http://fieo.org/view_section.php?id=0,23&lang=0 to know more about the status of the Indian service sector and its export
potential. Furthermore, there are several important links listed in this Web site. Look them up and prepare a report on the
status of the service sector and its export potential by answering the following questions:
a. Comment on the growth of the Indian services sector and its exports during the last ten years.
b. Is India a dominant player in global trade for services? Does it export more services? What types of services are
exported from India?
3. In the future, all manufacturing and service organizations would be required to adhere to sustainable practices. Therefore,
every operations manager must develop a good understanding of the issues and challenges that they are likely to face and pos-
sible actions that they can take to address these concerns. Visit the following URLs and study in detail how environmental
issues are being addressed by organizations:
http://www.ecocycle.org/
wwww8.hp.com/us/en/hp-information/global-citizenship/reporting.html
a. What are the implications of the mandatory requirement of “product take-back"? How will it affect the operations
management practices of manufacturing organizations? Can you identify two areas on which product take-back will
have significant impact and show how they will be affected?
b. How does Hewlett-Packard address environmental concerns? Identify three interesting aspects of their policy and show
how an operations manager could implement these policies in his/her organization.
SUGGESTED READINGS
Operations Strategy
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
In order to achieve operational excellence, Indigo airlines have deployed check-in staffs as baggage handlers. These practices
help in clearing the peak hour demand and saving time and cost for the company. The company’s strategy is to increase the
responsibilities of the staff and thrive as a successful operator in the aviation sector.
Source: https://book.goindigo.in/
Operations strategy is the process of making key operations decisions that are consistent with the overall
strategic objectives of the organization.
Every organization, whether in services or manufacturing, has a strategic intent. The strategic
intent could be to provide goods and services at a low cost, or to provide a highly customized
product or service. For example, the strategic intent of a law cost air line is to provide affordable
air travel. The strategic intent influences the overall corporate strategy of an organization, which,
in turn, shapes its operations strategy. Operations strategy is the process of making key
decisions regarding the operations function of an organization on the basis of inputs from its
overall corporate strategy. In this manner, it links the overall strategic intents and the corporate
strategy to specific aspects of the operations system. Strategic decisions include, for example, the
extent of capacity to be built into the system, the type of process and technology to be used, the
nature of products and services to be offered, and the type of supply chain to be configured. In
the process of making these decisions, it is important to identify key measures of operational
excellence. In this chapter, we shall look at all these aspects of operations strategy in detail.
Before we discuss the strategy-formulation process, it would help to understand the context and
relevance of operations strategy in any organization.
An analysis of the competitive scenario in our country in the last ten years reveals that it is now
imperative for organizations to have a sound operations strategy. Due to globalization of the
economy, Indian manufacturing and service firms have had to face competition from other parts
of the world. Indian organizations are now required to have a global outlook as opposed to the
traditional domestic outlook. They can no longer afford to ignore the need for customer
orientation and cost cutting, which have become crucial to the survival of an organization in
these competitive times. In the post-liberalization era, Indian firms are exposed to greater levels
of competition than ever before. Korean and Chinese goods flood the markets these days, and
this has profoundly influenced various sectors, ranging from industrial goods to consumer non-
durables. When markets get overcrowded with players, excess capacity builds into the system
and organizations face pricing and other pressures. In general, when the number of players
increases and a particular sector has excess capacity, competing organizations promise quick
delivery and greater options in products and services at a lower cost in order to win customer
orders. Indian organizations in several sectors are currently facing such a situation. On account
of this, a number of Indian firms have faced a significant fall in prices, a dramatic reduction in
delivery quotes, and the proliferation of variety. Organizations need to devise methods to remain
competitive in the market under these conditions.
A strategic planning exercise enables an organization to respond to the needs of the market in the most effective
manner by:
i. identifying products and services that are likely to succeed in the marketplace, and
ii. aligning the resources and activities in the organization to deliver these products and services.
Even when the annual inflation during the last ten years has averaged at about 5–6 per cent,
several component manufacturing firms have been asked to reduce the price of the components
that they supply on an annual basis. Triveni Engineering, a manufacturer of turbines, faced a 40
per cent reduction in the price of turbines in the “less than 3.5 million watts” category during the
period 2000–04. ABB India reported that the price of a 33-kV circuit breaker dropped from
₹275,000 in 1990 to ₹180,000 in 1999. In such a scenario, better cost management practices are
required in manufacturing and service organizations to handle the threat of competition.
Pricing pressures, lead-time reduction, and the proliferation of variety are some of the key challenges faced by
Indian organizations.
In addition to cost pressures, organizations have faced competitive threats in the lead time for
supply of goods. Lakshmi Machine Works (LMW) is a Coimbatore-based manufacturer of
machinery for the textile industry. Until 1997, it had been able to obtain customer orders and
supply machinery after a period of 48 to 60 months. In several cases, customers were willing to
pay in advance and wait for such a long time. However, with the entry of several Korean and
European manufacturers, the lead time for supply of machinery was pushed down to 10 months.
Therefore, LMW had to respond to this challenge in the market not only to remain in business
but also to be profitable.1 Similar examples of a demand for lead time reduction exist in several
other sectors of the industry. While it was customary to book a passenger car and wait for a few
months for delivery, today, a manufacturer of passenger cars cannot afford to make customers
wait that long. The development of superior capabilities to cut down lead time is an important
requirement today.
The third area of concern is the proliferation of variety. In response to mounting competition,
there has been a proliferation of variety in several sectors of the industry. These sectors include
white goods, electronic appliances, and industrial goods. Nowadays, it is interesting to note that
even simple commodities such as wheat flour and salt, which were never considered to be
branded items, have more than a dozen varieties to choose from. Clearly, organizations need to
develop a high capability for managing variety.
Returning to the example of passenger cars, the industry was delicenced in 1993. Until that
point, customers had a very limited choice of features, colours, shapes, and sizes. All this
changed with the liberalization of the sector. Today, passenger-car manufacturers are forced to
offer a wide range of choices to customers.
One can draw several key inferences from the developments in the area of operations
management in the last decade. Some of these are as follows:
Competitive dynamics will change on account of socio-economic changes and the policy changes of the government.
The expectations of customers will also change along with these.
Organizations need a structured approach to scan the market and distil the changing needs at the marketplace. They also
need a mechanism to chalk out a plan for responding to these changes in the most effective way.
With the changes in the marketplace, the competitive priorities for an organization must also change. Organizations
need to tune their operations to match with the competitive priorities.
The above-mentioned processes are expected to repeat several times in the future, and organizations must be in a
position to respond to changes whenever the need arises.
As per our definition of corporate and operations strategy, all these functions are achieved
through a strategy-formulation exercise. Therefore, it is important for organizations to develop
the capability to devise strategies for operations and to revisit the strategy-formulation exercise
as and when there is need for it. We shall now turn our attention towards the strategy-
formulation process.
Order-qualifying attributes are the attributes that customers expect in the product or service they consider for
buying.
There are other attributes that have the potential to sufficiently motivate the customer to buy
the product. Such attributes are known as order-winning attributes. The perception of the
customer is that the presence of these attributes indicates to him or her that the product/service
under consideration surpasses the basic requirements and, therefore, has the potential to provide
better value for money. The presence of order-winning attributes in a product/service helps the
customer differentiate it from the competitors’ offerings. It also favourably influences the
customers’ buying decision with respect to the product/service. The more the number of such
attributes that a customer perceives in a product/service, the greater is the chance that the
customer may buy the product/service.
Order-winning attributes are attributes that have the potential to sufficiently motivate the customer to buy the
product.
What constitutes order-winning and order-qualifying attributes might change from time to
time. During the early and mid-1980s, superior quality was an order-winning attribute for several
multinational organizations. Therefore, they invested much of their strategic planning and
implementation efforts towards developing quality-management systems. However, in the 1990s,
quality became an order-qualifying attribute as customers began to expect high levels of quality
in products/services. Moreover, several competitors were able to match customer expectations
with respect to quality. The next order-winning attribute was high delivery reliability.
Organizations developed just-in-time practices to addresses the issue of delivery. Other order-
winning attributes that have evolved over time include efficient consumer response, speed,
variety, and convenience.
Step 3: Identify Strategic Options for Sustaining Competitive Advantage
After an analysis of the competitive dynamics, an organization will be in a position to identify
the order-winning and order-qualifying attributes for the products/services that it offers. At the
end of the exercise, the order-winning and order-qualifying attributes provide the organization
with a set of options for sustaining competitive advantage. Competition analysis may indicate
that the organization will do better by providing high-quality goods, or that a wider variety of
options will bring success. It could also indicate convenience in usage in some cases. Such an
exercise often points to more than one possible attribute for order winning. These attributes also
help an organization develop appropriate operational measures of excellence.
Operational-excellence measures provide the critical linkage between order-winning and order-qualifying
attributes identified through the strategic planning exercise and the choices made in the operations.
Notes: *Single supplier; **25 Suppliers were supplying components to the seat-building department.
Source: J. P. Womack, D. T. Jones, and D. Roos, The Machine That Changed the World (New York: Rawson Associates, 1990).
The example we just discussed points to some interesting aspects of performance measures.
First, it shows that performance measures serve a very useful purpose in benchmarking with the
competitors. Second, the operational measures pertain to specific aspects of operations
management. For example, measures such as return to normal quality after new-model
introduction and average development time reflect the flexibility and delivery aspects of an
organization. Measures such as number of suggestions (for improvement) pertain to quality.
Clearly, there is a generic set of options that an organization would pursue as its competitive
priority. For every such option, several operational measures could be used to assess the
performance of the operations and to ensure that they are consistent with the organization’s
strategic objectives.
Four generic options are generally found to be useful in any operations-strategy exercise.
These include quality, cost, delivery, and flexibility. In a strategic planning exercise,
organizations identify one or more of these as their competitive priorities. If, for instance, the
strategic exercise of an organization indicates the need to provide faster and more reliable
delivery of products and services to customers, the operational measures of excellence must be
able to capture the performance of the operations system with respect to reliability and speed of
response. Therefore, measures of excellence should be appropriate and case specific, pertaining
to lead time, schedule adherence, and on-time delivery. Table 2.2 presents a sample list of
operational measures. These measures are not exhaustive or prescriptive. The list merely
indicates some possible suggestions, as found in practice. In reality, organizations come up with
their own set of measures consistent with their strategic goals.
Four generic options are generally found to be useful for any operations-strategy exercise: quality, cost, delivery,
and flexibility.
Five sets of measures are given in Table 2.2. The first four—quality, cost, delivery, and
flexibility—pertain to measures for assessing each generic option. All these measures are
operational in nature and often do not translate into financial numbers found in annual reports.
On the contrary, these measures capture the performance of an operations system in each of these
categories. For example, the value of import substitution and target-cost reduction exercises
directly measure the operational aspects of cost reduction.
While the first four measures directly pertain to the four generic categories of competitive
priorities for any organization, the last set consists of measures that indirectly affect any of the
four measures. These measures are used by several organizations in improving their competitive
advantage. They draw attention towards areas requiring improvement, which would have
otherwise been ignored. Take, for example, the measure of indirect labour to direct labour ratio.
As the ratio increases, it points to more indirect workers in the organization. More indirect
workers may often point to high levels of supervision and more effort in managerial planning
and operational control. Such an increase in activity invariably increases the lead time of several
processes. Further, it also adds to the overall cost. The number of certified deliveries indicates
how many vendors have been certified While the first four measures directly pertain to the four
generic categories of competitive priorities for any organization, the last set consists of measures
that indirectly affect any of the four measures. These measures are used by several organizations
in improving their competitive advantage. They draw attention towards areas requiring
improvement, which would have otherwise been ignored. Take, for example, the measure of
indirect labour to direct labour ratio. As the ratio increases, it points to more indirect workers in
the organization. More indirect workers may often point to high levels of supervision and more
effort in managerial planning and operational control. Such an increase in activity invariably
increases the lead time of several processes. Further, it also adds to the overall cost. The number
of certified deliveries indicates how many vendors have been certified to supply without inward
goods inspection procedures. As this number increases, the procurement costs will come down
and the quality will also improve, because the suppliers would have passed through stringent
quality requirements before qualifying.
TABLE 2.2 A Sample List of Operational Measures
Direct Measures
Quality Cost
First-pass yield Average days of inventory (number of inventory turns)
Quality costs Manufacturing cost as percentage of sales
Defects per million opportunities Procurement costs, total cost of ownership
Number of suggestions per employee Value of import substitution, cost reduction
Process capability indices Target cost reduction efforts
Delivery Flexibility
Lead time for order fulfillment and service delivery Number of models introduced
Procurement and manufacturing lead time New-product development time
On-time delivery for supplies Breadth and depth of the product and service offerings
Schedule adherence Process flexibility
Indirect Measures
Indirect-labour to direct-labour ratio Number of suggestions per employee
Ratio of lead time to work content Non-value-added content in processes
Process-rate to sales-rate ratio Number of certified deliveries
Average training time per employee Delivery quote for customized products and services
Translating the corporate strategy into the operations strategy essentially involves making
choices in the design and operational control of the operations system. The choices are made in
such a fashion that the options are consistent with the overall strategic objectives of the
organization. Figure 2.2 depicts the options available for strategic decisions in operations
management. The product portfolio addresses the core issues of the type of products to offer
through the operating system as well as the breadth of each offering. Similarly, technology,
capacity, supply chain, and process options enable an organization to make appropriate choices
in the respective domains.
Product Portfolio
One of the strategic choices that an organization can make is with respect to its product portfolio.
By product portfolio, we mean decisions regarding the products that the organization wants to
produce, the number of variations in each product line, and the extent of customization that it can
offer to its customers. In reality, organizations operating in the same sector and offering similar
products/services differ widely with respect to the range of choices in their offerings to
customers. This is a reflection of their overall strategic objective. A few examples will help us
understand this concept better.
FIGURE 2.2 Options for strategic decisions in operations
Let us consider air travel from Bangalore to Delhi. IndiGo and Jet Airways offer similar
services for passengers, but differ vastly in terms of the services offered. Jet Airways is a full-
service airline that provides business- and economy-class travel facilities. In addition, it offers
free air miles accrued through a frequent-flier scheme. IndiGo, on the other hand, offers just
economy-class tickets. There are wide differences in the in-flight services offered by the two
airlines. While Jet Airways begins a journey by serving a glass of soft drink to its customers and
follows it with appetizers, a meal, coffee, tea, and mineral water (in unlimited quantities), in the
case of IndiGo, nothing is available free of cost. A limited choice of snacks and meals is
available against payment. At the outset, it may appear as though Jet Airways’ offer is more
attractive. However, a closer analysis will show that the two airlines pursue different strategic
objectives and, therefore, have different competitive priorities and customer segments to cater to.
IndiGo’s priority is one of low operational costs while that of Jet Airways is of high-quality
service.
A similar example of how the product portfolio varies with the overall strategic objective can
be seen in the varied range of restaurants in metropolitan cities such as Mumbai. Restaurants
ranging from small self-service eating joints to multi-cuisine restaurants with a host of well-
trained waiters and huge dining spaces span the entire range of product/service port-folio
options. These options are related to the overall strategic intent of each organization.
Several such examples exist in the manufacturing industry as well. Consider Dell and Lenovo,
two well-known computer manufacturers. The overall strategic objective of Dell appears to be
one of providing highly differentiated products, whereas Lenovo appears to emphasize robust
and reliable computing power. Therefore, the product portfolio choices for these two
manufacturers are different.
The product portfolio of an organization refers to decisions on what products the organization wants to
produce, the number of variations in each product line, and the extent of customization it can offer to its
customers.
These examples demonstrate the role of a product portfolio in fulfilling the desired strategic
objectives of an organization. If the overall strategic objective is to provide a highly
differentiated set of products and services to the customer, it is possible to achieve this by
offering product/service customization. On the other hand, if the overall objective is one of cost
leadership, then the product portfolio tends to be narrow.
For an organization desiring to provide a wide product portfolio to its customers, flexibility
will be the most important competitive priority. Several aspects of operations management will
be configured to meet this objective. Operational control systems and performance measures will
be chosen in such a manner that they emphasize flexibility over other criteria such as cost and
delivery. The organization should develop an efficient and responsive new- product development
process that can respond to the changing requirements at the marketplace with newer versions of
the product.
Process Design
Appropriate choice of processes is another option through which an organization can translate its
strategic objectives to operations. Process design refers to the overall configuration of the
operations system in such a way that the various activities are consistent with the process choice.
Viewed alternatively, process choices influence the nature of material and information flow in an
organization. At a broad level, three types of flow happen on account of process choices:
continuous streamlined flow, intermittent or batch flow, and jumbled flow. The choice of process
technology will be consistent with that of the product portfolio decisions that an organization
makes.
Process choices lead to three types of flows: continuous streamlined flow, intermittent or batch flow, and
jumbled flow.
Consider a volume manufacturer such as Hero MotorCorp. Ltd. (HML) and a variety
manufacturer such as Bharat Heavy Electricals Limited (BHEL). The process choices of these
two manufacturers are vastly different and are in line with their overall objectives. The overall
objective of HML can be stated as providing a variety of high-performance motorbikes at
reasonable prices. BHEL’s overall objective is to fulfil complex and huge turnkey engineering
contracts in the power sector on time. A manufacturer emphasizing high production volumes,
fewer varieties, and low cost (HML in our example) will make process choices that are
appropriate for this requirement. In operations management, a mass or a batch production system
is appropriate for such manufacturers. In the case of BHEL, on the other hand, the process choice
will be influenced by the need to have a wide variety and a large production volume of a single
or a few units. Project shops and functional layouts are some appropriate processes for BHEL.
Similar examples could be found in services as well. A banking service for a high networth
individual (Such as SBI Kohinoor Barjara Branch) will have a process design which is different
from the regular branches of State Bank of India3.
An organization working towards the objective of providing a wide range of products to its
customers will have to make a process choice that is more of the batch/intermittent flow type.
The production planning and control systems, scheduling, and the procurement practices will all
be influenced by the flow system an organization chooses to adopt. Some of these issues are
discussed in greater detail in Chapter 4.
Supply Chain
A manufacturer of goods relies on other business organizations for raw materials and
components required in the manufacturing process. It also depends on other business entities to
supply its finished goods to the end customers through a market. The supply chain is the
network of entities supplying components, raw materials, and a host of services to the
organization as well as those distributing the finished goods and services to customers through
alternative channels. Organizations need to make appropriate supply-chain choices consistent
with their strategic objectives.
The supply chain is the network of entities supplying components, raw materials, and a host of services to an
organization as well as those distributing the finished goods and services of the organization to its customers
through alternative channels.
Recent studies have shown that fundamentally two types of supply chains can be configured—
an efficient supply chain and a responsive supply chain. As the name suggests, an efficient
supply chain is designed with the main objective of pursuing efficiency goals with respect to
supply-chain operations. Essentially, this translates into cost optimization and maximum
utilization of the resources employed in supply-chain operations. In the case of a responsive
supply chain, the key objective is to develop a capability to respond fast to market requirements.
To design an appropriate supply chain, it is imperative to have a good understanding of the
product profile for which the supply chain is configured. The product profile is the result of the
product–portfolio decisions made as part of the operations strategy.
Let us consider the design of a supply chain for the distribution of Annapurna atta (flour) and
salt. Salt and flour are functional products. Functional products are characterized by stable
demand patterns, little or no change in consumption patterns, and a reasonable level of
predictability of demand. Moreover, customers are not likely to pay a significantly higher price
because of some special aspect in the distribution of the product or the service. There are no
fashion and fad elements in these products, based on which a price premium could be extracted
from customers. Under these conditions, cost minimization and efficiency goals will be the
strategic requirement, and the supply chain should be configured accordingly.
The strategic requirements for some products are in sharp contrast to those for functional
products. For instance, consider products such as Tanishq watches and jewellery offered by
Titan. These products are trendy, expensive, and sensitive to fashion fads. There is also a need
for creating greater variety and more designs to satisfy diverse tastes and preferences. As the
demand for such goods is difficult to predict, the organization has very little lead time to react to
changing patterns and customer preferences in the market. This category of products is known as
innovative products. The supply chain for such products should be designed with the basic
objective of providing a high level of responsiveness. The two examples discussed here
demonstrate the need for configuring the outbound supply chain (dealers, distributors, retailers,
etc.), in the case of consumer goods, in tandem with the choice of the product portfolio and the
overall strategic objectives of the firm.
In the case of industrial goods, too, the design of supply chains should be consistent with the
strategic objectives of the organization. This is true for the design of inbound supply chains (raw
material and component suppliers) as well. We shall now look at two more examples to
understand the design of supply chains in the case of industrial goods. The first is the
requirement of raw materials such as steel plates and rods for the manufacture of components by
a manufacturer of gearboxes for automobiles. In this case, the product falls neatly into the
description of a functional product. Therefore, the organization needs to configure its supply
chain for the raw materials on the basis of efficiency. Let us consider another product—a digital
control system for remotely monitoring various parameters and thereby controlling the
operations of a steel-making machine. Due to several new developments in electronics and
processor speed and capabilities, the product is likely to be subjected to frequent changes in
design and enhancements. It should have a supply chain that is able to bring in new
enhancements in the control system as and when they happen in the market and help in making
product upgrades more frequently. In this case, the organization is handling an innovative
product and should be able to configure a responsive supply chain for the control system.
Efficient supply chains are designed with the central objective of overall supply-chain cost
minimization and better asset utilization. Organizations need to incorporate several features in
their supply chain to achieve this objective. The design of a responsive supply chain, on the other
hand, begins with the basic premise that uncertainty in demand and large forecast errors are often
the reality. Therefore, the supply chain requires certain strategies to address these issues. Further,
developing systems to improve responsiveness is another objective of the design. Our intention
here is to merely show how the design of the supply chain will be influenced by the overall
strategic objectives of the organization. In Chapter 5, we shall return to the question of supply-
chain design and provide some more details on the different methods organizations adopt to
design efficient and responsive supply chains.
An efficient supply chain is designed with the main objective of overall supply-chain cost minimization and
better asset utilization.
Technology
Technological advancements in recent years have offered new opportunities for creating
competitive advantage. Recall how Asian Paints first utilized technological advancements for
mixing basic pigments to distribute paints in a large variety of colours and a large assortment of
sizes. Organizations making a strategic choice to operate in the mid-volume, mid-variety
products and services could utilize new technology such as flexible manufacturing systems and
computer-integrated manufacturing to improve productivity and to respond faster to the market.
Technology can provide unique advantages to organizations in providing better products and
services to customers. Prominent among these are the following:
Increased resource utilization: Computer-based operation of systems has enabled high resource utilization in systems.
Scheduling flexibility: Technology often enhances flexibility in scheduling various customer orders as it enables
organizations to react to changes swiftly. Therefore, organizations can use technology to provide better customer
service.
Ease of changes: Changes in design and process plans can be easily accommodated through the use of technology.
Further, complex operations can be handled without much difficulty and the desired quality levels can be achieved.
Ease of expansion: Technology also provides volume flexibility to the organization and makes it much easier to expand
in response to a growing market.
Reduced lead time: Advanced technological features make it possible to cut down lead time.
Lower in-process inventory: Several of the benefits discussed here directly translate to lower work-in-process inventory
and reduced cost.
It is evident that by using new technology options for processes, organizations can react faster to
customer needs, manage a wide portfolio of product offerings, and yet maintain high levels of
productivity. Therefore, process technology is considered to be a potential enabler of operations
strategy in organizations. In some cases, the use of radically new technologies results in high
depreciation and operating costs. However, if the strategic intent is to provide high levels of
responsiveness to customers, it may be possible to recover these costs by passing them on to the
customer by way of a premium for the additional value provided in comparison to the
competition.
Capacity
Capacity refers to the maximum number of units of goods that can be produced per unit of time
in a manufacturing system. In a service system, it refers to the maximum number of service
offerings that can be made per of unit time. For example, when a two-wheeler manufacturer
reports that it has an installed capacity of 20,000 vehicles per day, it merely indicates the
maximum number of two-wheelers it can produce in a day. Similarly, if an inter-city bus
operator says that its capacity is 500 passengers per day, it points to the limit of its service.
Capacity decisions in operations have an important strategic dimension as they directly and
significantly influence the cost of goods and services offered in three ways. First, there is an
accrued cost advantage due to economies of scale. Economy of scale is an economic principle
that illustrates the relationship between the cost of the goods produced by a system and the
appropriate levels of capacity built into the system. According to this principle, as a firm grows
and production units increase, it will have a better chance to decrease its costs. A firm can thus
benefit from lower total costs by operating at a predetermined level.
In a manufacturing system, capacity is the maximum number of units of goods that can be produced per unit of
time. In a service system, the term capacity refers to the maximum number of service offerings that can be made
per unit of time.
The second aspect of cost advantage comes from cost accounting principles pertaining to
marginal costs and contribution. The cost of goods and services produced is a function of the
fixed and variable costs of the operating system. Every organization has three major types of
investments in fixed costs. They are: (1) asset costs arising out of investment in machinery and
buildings; (2) manpower costs of the supervisory and managerial workforce; and (3) facilities
maintenance costs to sustain several support facilities. By increasing the productive capacity (and
consequently the sales), it is possible to spread these fixed costs over a larger base of activities
and thereby achieve lower costs and greater contribution.
The third benefit arising from large capacity investments is that it provides additional cost
advantages in procuring other factors of production. Large capacity implies a large production
base and large amounts of procurement of key raw materials and other resources. Large volume
provides bargaining power in the market. Furthermore, the firm will be in a position to influence
key trends in the industry and to channelize the development efforts of its suppliers and
collaborators to its overall benefit in the long run. Let us consider an example to understand how
this is possible. Reliance Industries Limited is one of the largest petrochemical manufacturers in
the world. Many of the finished products of Reliance, such as petrol and LPG, are stored in tanks
and transported through tankers to various distribution points. A large number of tankers is
required at the Jamnagar facility and the Hazira facility near Surat. Typically, the daily
requirements are in the range of 3000 trucks. The large scale of operation enables Reliance to
negotiate better rates for trucking services.
It is obvious that as every individual unit sells at a price p and has incurred a variable cost of v,
the difference between the two is the excess over the variable cost that could cover the fixed
costs. This quantity is known as the contribution margin (c), and can be represented by the
following equation:
c=p−v
The break-even point is the point at which the contribution margin is able to cover the total
fixed costs. Therefore, at the break-even point, the organization will have a no-profit, no-loss
situation, but would have covered all the fixed costs invested in the system.Break-even sales
refers to the number of units to be sold at the break-even point. The break-even point and break-
even sales can be expressed as follows:
The break-even point is the point at which the contribution margin is able to cover the total fixed costs.
EXAMPLE 2.1
A company manufactures ball-point pens that it is able to sell at ₹15 per piece. The variable
cost of the pen is ₹10 per unit. If the company has made a total investment in fixed costs to
the tune of ₹30,000, what is the break-even sales for the pen? Graphically represent the cost
structure and the break-even point.
Solution
F = Fixed costs of production = ₹30,000
v = Variable cost of production of one unit of the pen = ₹10
p = Selling price of one unit of the pen = ₹15
Therefore, contribution of one unit of the pen towards the fixed costs (c) can be calculated as
follows:
c = p − v = ₹15 − ₹10 − ₹5
As seen in Figure 2.3, the point of intersection of the total cost and the revenue curve is the
break-even point, and the corresponding sales are the break-even sales.
FIGURE 2.3 Graphical representation of the break-even point
From our discussions on the various options available for operations strategy, it is evident that
the cost and flexibility dimensions compete with each other. Flexibility would mean more options
in product variety and process design, which may not allow the organization to exploit the
benefits of volume. Therefore, capacity costs may have to be spread over smaller volumes of
several varieties. Further, the organization may need a responsive supply chain. Forecasting
errors could be high, resulting in non-moving inventory and wastages. All these choices tend to
increase the cost of the products. In contrast, providing less flexibility would mean fewer product
varieties. Organizations pursuing such an objective will be able to pursue cost-reduction goals
very well through the various strategic options available for configuring their operations. Fewer
varieties of products will enable the organization to build large capacities, have a streamlined
flow process, and configure an efficient supply chain.
Figure 2.4 shows the relationship between flexibility and cost dimensions in operations
strategy. As shown in the figure as flexibility increases, the cost is also likely to increase. There
are two options left to an organization to handle situations arising out of increased flexibility.
One is to pass on the costs to the customers as a premium. Generally, as flexibility increases, the
perceived value of the product also increases and customers do not complain about paying a
premium for flexibility. The other option is to develop procedures in the operating system to
overcome such trade-off obstacles in comparison with those of their competitors. World-class
manufacturing concepts enable organizations to pursue this option.
FIGURE 2.4 Flexibility versus cost: competing dimensions
Management practices in manufacturing have undergone significant evolution during the course
of the twentieth century. The culmination of these is the guiding principles of excellence in
manufacturing, known as world-class manufacturing (WCM).4 We have already seen in the
previous sections how quality, cost, delivery, and flexibility are important components of the
operational excellence of any organization. There is one crucial difference between world-class
manufacturing organizations and others. The latter category of organizations constantly faces
trade-off obstacles in meeting the desired levels in the competitive priorities. For example, a firm
may perform very well in quality parameters but not so well when it comes to the other three
parameters. For such firms, providing high-quality goods and services would mean increased
cost due to tighter and closer quality control and management and perhaps an increase in the lead
time, thereby resulting in poor performance in delivery. In some other cases, organizations might
have overcome the cost-versus-quality trade-offs, but not the other two, that is, flexibility and
delivery.
A world-class manufacturing organization excels in all the four generic competitive priorities— quality, cost,
delivery, and flexibility—by overcoming all tradeoff obstacles.
World-class manufacturing organizations, in contrast, will be able to perform very well in all
the four parameters at the same time and meet the desired goals in each of these. Overcoming
trade-off obstacles requires substantial investment in systems and procedures that enable
organizations to build this capability. Such investments are made in almost all functions of
operations management and considerable time is required to stabilize the system.
Here, we shall discuss these principles briefly. These concepts are dealt with in detail in other
chapters.5
The core philosophy of JIT systems is to provide an organizational framework to continuously reveal
opportunities for the elimination of non-value-added activities.
Total quality management (TQM) philosophy helps an organization in organizing the entire workforce into
small improvement groups and in creating a mindset for continuous improvement.
Employee involvement
In order to put the three systems discussed so far in place, it is necessary to break away from the
rigid classification of labour that is characteristic of traditional manufacturers. In organizations
that use JIT, TQM, and TPM, production workers may perform some maintenance tasks.
Alternatively, another group of employees may work on some improvement projects. The group
may have to be allowed to perform their own quality checks, and a certain degree of autonomy
may have to be granted for the scheduling and planning of their production tasks. All these
changes may appear to be unconventional for a traditional manufacturing system. Functional
managers and production supervisors may have a feeling of losing control. Personnel managers
may not like these new arrangements as they may have to negotiate with the workers’ union to
change some of the work practices. It may pose problems for the cost accountant because he/she
may not know how to allocate the labour cost between the direct and the indirect components
that a worker performs in the new set-up. Alternatively, the cost accountant may wonder how to
define direct labour. Despite these hiccups, the experience of world-class manufacturing
organizations suggests that the net effect of such changes is a substantial gain for the
organization. These new changes in working patterns are defined as employee involvement.
Simplicity
The last feature that is peculiar to WCM organizations is the simplicity with which they function.
The best examples of simplicity can be seen in the Toyota production system. The introduction
of JIT, TQM, TPM, the kanban system, fail-proofing systems (such as poka yoke and andon
lights), and employee involvement have all contributed to the ideal of simplicity. As business
grows, and as more varieties of products are introduced into the system, the complexity of the
manufacturing, management, and control systems do not increase in proportionate terms. This
helps WCM organizations to overcome trade-off obstacles. Moreover, this feature also indicates
how well systems and planning methodologies have been established that are robust enough to
aid the increase in the scale and complexity of activities in an organization. In ordinary
organizations, systems, and methodologies are not tuned to scaling up operations in the future.
Therefore, increasing variety, for example, may result in a hike in cost due to the increased
complexity of managing operations.
A simple demonstration of the various benefits of simplicity can be seen in the use of kanban
cards, a method of production scheduling typical to JIT systems. Kanban cards are used in
several world-class manufacturing organizations. In very simple terms, the use of kanban
resembles a supermarket operation. In a supermarket, as a customer takes away his/her item of
choice from the display shelf, the line staff will merely replenish the stock to desired levels at
periodic times. As the line staff replenishes the stock, the inventory of stock at the stores
diminishes. The supplier of the item periodically monitors the stocklevel at the stores and
responds to the depleting stock with his/her routine of replenishment. Proceeding in this manner,
a simple event of sale of finished goods to a customer may trigger a chain reaction in the system
and induce all the upstream components of the supply chain to react to the change.
Instead of using a complex planning process and paperwork to communicate orders across
various people in the system, an alternative system like kanban (which is highly visible and
instantaneous in response) can be used. In the former method of planning and control, a
centralized planning department computes the requirement at various stages in the system and
subsequently communicates the decisions arrived at through the planning process to all affected
entities. In the case of kanban, on the other hand, this is left to the affected parties themselves.
The entire system operates using a decentralized planning framework, whereby it is able to
respond to changes more effectively.
Challenges in WCM
Transforming into a world-class manufacturing organization has not been easy for most
organizations even though the principles of WCM are very simple and intuitively appealing.
Several reasons could be attributed to this. The most significant among them is the fact that
organizations have a mindset inertia that generally resists changes. Only during periods of crisis
that threaten the very survival of the organization are changes accepted without much resistance.
Besides crisis situations, the only other factor that can galvanize change is the active role of top
management in spear-heading the change process. The second reason is the quantum of effort
required to establish the required systems and planning methodologies in a world-class
manufacturing organization. Transformation into a truly world-class manufacturing organization
may call for sustained efforts over 8–10 years.6 Often, organizations lack such levels of stamina
and therefore lose the tempo midway. If these two issues are addressed properly, transformation
will definitely be possible.
In recent years, several developments that affect operations management practices have taken
place in the marketplace. We shall now take a look at how some of these developments have
influenced operations and what are their implications for operations strategy.
Ability to provide goods and services at a fraction of the cost at which they are available in developed
countries
large installed base of manpower with technical knowledge, manufacturing know-how, and experience in
manufacturing management
Despite these new challenges, Indian manufacturing firms have greater chances for market
expansion on account of the liberalized economy because of two important factors. The first is
the overall attractiveness of Indian organizations due to cost factor advantages. Indian
manufacturing firms can provide goods and services at a fraction of the cost at which they are
currently available in the developed countries. This is because of the relatively low cost of
labour. The second advantage for India is the large installed base of manpower with technical
knowledge, manufacturing know-how, and experience in manufacturing management. Therefore,
Indian firms need to understand these issues and incorporate them in their operations strategy.
Cost reduction opportunities and other strategic options need to be pursued to compete on the
basis of operations. To compete in global markets, Indian organizations need to develop the
following capabilities:
Identify a segment and create global scale.
Develop and hone process skills and create operational excellence.
Move from mere manufacturing of components to design and development through greater investment in research and
development.
The Outsourcing Wave
Globally, firms are looking towards India for their software development and maintenance needs.
Based on the successful experience of outsourcing software jobs to India, firms in developed
countries are increasingly shifting a variety of other jobs here. This is what we refer to as the
outsourcing wave. India’s share in worldwide service exports was 0.6% in 1990, 1% in 2000,
and 3.3% during 2011. Between 2007–08 and 2011–12, the annual growth rate in export of
computer software and services including ITES was 18.27% in rupee terms. The key motivation
for a firm to outsource some of its processes stems from three factors: cost, capacity, and core
competency. A vast majority of examples from the BPO industry point to significant cost
benefits accruing to the outsourcing firm on account of factor cost differentials. By investing in
specialized processes and processing only specific requirements, the outsourcing service
provider is able to deliver the same service at a much lower cost.
Business process outsourcing (BPO) is an arrangement by which some of the non-core business processes of
an organization are carried out by a third party on behalf of the organization.
The second reason for outsourcing is the benefit of quick expansion of capacity without
actually investing one’s own funds. Third is that the organizations are increasingly showing a
preference towards investing their funds, management time, and other resources in the core
aspects of the business. This is primarily due to the fact that value addition is mainly dependant
on the core competency of a business. Since the primary consideration for a BPO is cost,
operations strategy for a BPO firm must emphasize cost leadership. Otherwise, the BPO
activities may be shifted to a competitor. Another critical performance measure is quality. Since
an organization often outsources the entire range of operations in a business process to a third
party, they need to develop stringent quality norms. Therefore, such organizations need strategic
plans to ensure adherence to strict quality norms. In several other cases, in addition to the cost
and quality requirements, stringent delivery requirements may also have to be met as the
outsourced processes may be in the intermediate stages of the value-creation process.
The key motivation for a firm to outsource some of its processes stems from three factors: cost, capacity, and
core competency.
Collaborative commerce refers to the use of the Internet or electronic markets to enable trading partners to
transact efficiently.
SUMMARY
A strategic planning exercise enables an organization to respond to market needs in the most effective manner by
aligning its resources and activities to deliver products and services that are likely to succeed in the market.
Operations strategy is the process of making appropriate decisions in the operations function on the basis of inputs
from the corporate strategy.
A strategy-formulation exercise enables an organization to identify order winners and order qualifiers. These attributes
change from time to time.
Four generic performance measures are generally found to be useful in any operations-strategy exercise—quality, cost,
delivery, and flexibility.
Translating corporate strategy to operations strategy requires making appropriate choices with respect to product
portfolio, processes, technology, capacity, and the supply chain.
World-class manufacturing organizations feature five basic elements of operational excellence. These include just-in-
time (JIT) manufacturing, total quality management (TQM), total productive maintenance (TPM), employee
involvement, and simplicity.
The dismantling of trade barriers demands that Indian manufacturing and service organizations equip themselves with
the required operations-management practices to tap global trading opportunities.
Current trends such as business process outsourcing (BPO) and collaborative commerce through the Internet point to
new requirements in operations management.
REVIEW QUESTIONS
1. What does the term operations strategy mean? How is it different from corporate strategy?
2. Why do firms engage in operations-strategy exercises? How often do they perform such exercises?
3. What are order-winning and order-qualifying attributes? Give three examples of each in the service and manufacturing
industries.
4. How do firms identify order-winning and order-qualifying attributes?
5. Briefly describe the strategy-formulation process.
6. What is the use of measures for operational excellence? Explain with the help of a few measures.
7. What is the linkage between product choices and process choices in an organization?
8. An organization is currently manufacturing basic cooking utensils for household use. Although it has been in operation for the
last ten years, of late there has been a margin squeeze due to a fall in its sales volume. The organization is trying to decide if it
should provide a wider range of product choices to its customers in order to improve its margins. What are your suggestions to
the organization for resolving this confusion? Would your answer have been different if they were manufacturers of high-end
premium cookware?
9. What does world-class manufacturing mean? What are the characteristic features of a world-class manufacturing
organization?
10. Why do some firms such as Arvind Mills invest in huge capacity in their operations? Will they benefit from this?
11. How can an organization configure its supply chain using an operations-strategy process?
12. What is the role of technology in operations strategy?
13. What do you understand by the term flexibility–cost trade-off in operations strategy?
14. How does a world-class manufacturing organization differ from other organizations?
15. How do economic liberalization and globalization measures influence operations management practices? Give some specific
examples to argue your case.
16. What are the main factors promoting business process outsourcing?
17. How can the Internet affect the practice of operations management? Does it have any implications for operations strategy?
PROBLEMS
1. Quick Photo Solutions is in the business of processing photographic films. The annual fixed cost of equipment incurred by
Quick Photo Solutions is ₹600,000. The demand for film processing is 50 rolls per day, and they work for 250 days a year.
The variable cost of processing a film (including labour, power, and chemicals) is ₹90 per roll. Quick Photo Solutions is
presently charging a processing fee of ₹150 per roll.
2. Raja runs a reprographic service in a university campus. His investments in photocopying machines and physical
infrastructure in terms of office space, constituting the fixed costs, amount to ₹250,000. The variable costs incurred by Raja
are as follows: cost of one sheet of paper = 5 paise; cost of power = 12 paise per page; labour cost per page = 23 paise. Raja
charges 60 paise per page for photocopying.
3. A manufacturer of condiments has to decide between Machines A, B, and C, three alternative machines available for final
packaging. While the initial fixed costs are high for Machines B and Ccompared to Machine A, their operating costs are
lower. The relevant cost data for each of the machines is given in Table 2.3.
The manufacturer is not sure which machine he should choose. What should he do?
(Hint: Identify the volume of production at which use of these machines is justifiable)
NET-WISE EXERCISES
1. Visit the following links:
Spice Jet: http://www.spicejet.com
Jet Airways: http://www.jetairways.com/IN/
On these Web sites, there are several links you can click on. Click on Products and Services, Plan Your Travel, and
About Us. After visiting both the sites, write a report to address the following questions:
2. BMW is one of the leading car manufacturers in the luxury car segment. It is known for its premium standards and quality.
Visit the Company, Research & Development, and Production links on the company Web site,
www.bmwgroup.com/e/o_owww.bmwgroup.com/forschung_entwicklung/forschung_entwicklung_2011.html. There are
several sub-links under these heads. Click on them to understand various aspects of BMW’s operations.
Maruti manufactures India’s largest-selling car. Visit Maruti’s Web page, http://www.marutisuzuki.com/, and click on
About Us and Our Cars.
After visiting both the Web sites, prepare a report to answer the following questions:
3. Download the report on Dupont’s approach to Operational Excellence available at the following URL:
http://www2.dupont.com/Consulting_Services/pt_BR/assets/downloads/OP%20E_White%20Paper_FINAL_12%2014%2005.pdf
After going through the report, prepare a write-up by answering the following questions:
a. Why should organizations address the issue of operational excellence? What benefits can they get out of this exercise?
b. What are the salient features of Dupont’s approach to operational excellence?
c. What is the significance of operational risk management? How is it addressed?
CASE:GIGNER HOTEL
Roots Corporation Limited is a fully owned subsidiary of the Indian Hotels Company Limited (IHCL), which is a part of
the Tata Group. With more than 90 properties, IHCL is India’s largest hotel chain. It has been in the hospitality sector for
over a century. Roots Corporation Limited operates a group of hotels under the brand name Ginger Hotels. The first hotel
was launched in Whitefield, Bangalore, in June 2004. Today, Ginger hotels are located in 27 cities in India.
At a glance, a Ginger hotel will appear to be very similar to any other hotel. A Ginger hotel offers all the facilities that a
normal hotel would offer. These include check-in facilities; rooms with a TV, fridge, and a tea/coffee maker; room services
such as laundry; restaurants; digital safes; Wi-Fi connections; meeting rooms, a business centre, gymnasium, car rental
service, doctor on call, and currency exchange. However, the similarity ends at this level.
A Ginger hotel distinguishes itself in several ways in the manner these services are offered. Unlike other hotels, Ginger
hotels offer a limited à la carte menu in the restaurant at a nominal price. In case a guest does not like what is being offered,
it is possible to call up nearby restaurants, place an order, and collect the food from the Give n’ Take Counter in the hotel.
The rooms are compact and well maintained, and are available at a price that is much lower than the price charged by other
hotels for a similar service.
“Please help yourselves” is a line that can be seen on most of the brochures and booklets in a Ginger hotel, and it aptly
reflects its most distinguishing feature It is not uncommon for guests to use the self-service check-in kiosk, identify their
room, and carry their luggage to the room. As soon as a guest enters a Ginger hotel, he/she will come across several
operations with a self-service facility. Some elements of self-service are described here:
Self-Service Check-in: Upon arrival, guests can check into the hotel without any assistance from the reception counter.
This is possible because Ginger hotels have self-check-in kiosks.
Give n’ Take™ Counter: Ginger hotels have a “Give n’ Take” counter that the guest can use to deliver used clothes for
laundry in the morning and to collect washed clothes after 7.30 p.m. the same day.
Smart Get Set: There is an ironing room in every floor in Ginger hotels. Guests can use the room for pressing their
clothes. Further, there are water dispensers on each floor, from which guests can fill their bottles.
Smart Knick Knacks: Ginger has installed vending machines for hot and cold beverages and packed snacks. These
vending machines can be accessed round the clock, irrespective of whether the restaurant is working or not.
Smart Mart: There are vending machines that supply other things such as toiletries, combs, toothpaste, hygiene
products, and mosquito repellants.
The company summarizes “the Ginger experience” as one providing intelligent, well-thought-out facilities and services at
great value and with no frills attached.
SUGGESTED READINGS
R. Dhawan, G. Swaroop, and A. Zainulbhai,. “Fulfilling the Promise of India’s Manufacturing Sector”, McKinsey
Quarterly, 2012, pp. 1–10.
H. Gebauer, G-J. Ren, A. Valtakoski, and J. Reynoso, “Service-driven Manufacturing: Provision, Evolution and
Financial Impact of Services in Industrial Firms”, Journal of Service Management, 23(1), 2012, pp. 120–136.
R. H. Hayes and G. P. Pisano, “Manufacturing Strategy: At the Intersection of Two Paradigm Shifts,” Production and
Operations Management 5, no. 1 (Spring 1996): 25–41.
T. Hill, Manufacturing Strategy: Text and Cases, 3rd edition (Burr Ridge, IL: Irwin/McGraw-Hill, 2000).
D. L. Kellogg and W. Nie, “A Framework for Strategic Service Management,” Journal of Operations Management 13
(December 1995):323–337.
T. Ohno, Toyota Production System: Beyond Large-Scale Production (Portland, OR: Productivity Press, 1992).
G. P. Pisano and W. C. Shih, “Does America really need manufacturing?”, Harvard Business Review, 90(3), 2012, pp.
94–102.
R. J. Schonberger, World-Class Manufacturing: The Lessons of Simplicity Applied (New York: Free Press, 1986).
R. J. Schonberger, Building a Chain of Customers: Linking Business Functions to Create the World Class Company
(New York: Free Press, 1996).
R. J. Schonberger, World-Class Manufacturing: The Next Decade (New York: Free Press, 1996).
R. J. Schonberger, Let’s Fix It: Overcoming the Crisis in Manufacturing (New York: Free Press, 2001).
J. Shah and L. S. Murty, “Compassionate, High Quality Health Care at Low Cost: The Aravind Model,” IIMB
Management Review 16, no. 3 (September 2004): 31–43.
W. Skinner, “Manufacturing Strategy on the ‘S’ Curve,” Production and Operations Management 5, no. 1 (1996): 3–
14.
J. P. Womack and D. T. Jones, Lean Thinking (London: Touchstone Books, 1996).
CHAPTER 3
Sustainability in Operations
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Over the recent past, there has been growing concern on the impact of business on environment. These arise on
account of depletion of natural resources, waste generated from production and service systems and at the end-
of-life of products.
Over the recent past, there has been growing concern on the impact of business on
environment. These arise on account of depletion of natural resources, waste generated from
production and service systems, and at the end-of-life of products. Increasingly, firms are under
pressure to take responsibility for restoring, sustaining, and expanding the planet’s ecosystem
instead of merely exploiting them. In this chapter, we shall see various facets of sustainability
and their relevance to the manner in which firms organize their operations.
3.1 SUSTAINABILITY: A KEY BUSINESS REQUIREMENT
In our quest to lead a comfortable life, mankind continuously engages in innovation and new
product and technology development. On the other hand, business organizations are constantly
engaged in increasing their profitability and market shares. In this process, one of the main
agendas that organizations seem to be pursuing is to create a consumption-oriented society.
Innovation initiatives coupled with marketing and advertising activities in an organization
focuses on fuelling the imaginations of the customers and fulfilling their new found wants. On
the other hand, design and new product development activities are focusing on making existing
product lines obsolete and bringing new generations of products and services. As consumption
increases, production and distribution activities also increase, creating fresh business
opportunities. This approach to business has resulted in certain outcomes:
1. In this approach to business growth and profitability, we seem to have forgotten for the moment that resources required
for sustaining the current levels of consumption may be fast depleting.
2. The increased level of industrial activity generates enormous amounts of toxic industrial wastes that threaten to harm the
atmosphere.
3. The increased level of consumption leads to greater levels of disposal of solid wastes. Modern societies (especially in
big cities) are finding it increasingly difficult to handle the wastes.
4. Increased consumption of natural resources also lead to more atmospheric pollution that it may soon threaten our health.
There are enough indications of these trajectories in our development path. Studies indicate that
the US has 5% of the world’s population but consumes 30% of the world’s resources and creates
30% of the world’s waste. On an average, an American consumes 17 times more than Mexican
counterpart and 100 times more than Ethiopian. If everybody consumed at the US rate, then we
would need 3–5 planets to sustain our current levels of consumption. On account of these, in the
past three decades alone, one-third of the planet’s natural resources base has been consumed.
Moreover, the CO2 emissions have grown on an average by 2% per year, leading to a 40%
increase in emissions during 1990–2007. One illustrative example is the case of bottled water. In
2004, nearly 28 billion plastic bottles were produced. Nearly 85% of them landed in the garbage.
Furthermore, in order to manufacture these bottles, 17 million barrels of oil was used and about
2.5 million tonnes of CO2 emitted from the manufacturing system. Depletion of the ozone layer
in the atmosphere, faster melting of the polar icebergs, and frequent devastations caused by the
nature are some of the larger pointers to the need to address the issue of sustainability.
Governments across the world have recognized the importance of sustainability. In the last 10
years, we saw a multifaceted approach to tackle the problem of environmental protection and
sustenance. Policy makers have brought in new legislations that put regulatory pressures on
businesses as a means to tackle the problem. With increase in economic and social costs of
disposal, increasing protests from environmental protection groups against dumping wastes in
third world countries and reducing spaces for landfill, legislations have also evolved from the
earlier forms of pollution-level control to the recent forms such as ‘compulsory take-back of
products at the end-of-life’. Notable among them are the EU directives on compulsory product
take-back at the end-of-life, The Netherlands National Environmental Plan and the package
recycling and product take-back laws in Germany.
The legislations based on ‘extended producer responsibility’ require business organizations to take responsibility
for the reverse supply chain. This has introduced more complexities in operations management.
This required the business organizations to take responsibility for the reverse supply chain
(used, thrown away products by the customers reaching back the manufacturing facility). These
legislations are based on ‘extended producer responsibility’—a policy tool to minimize
environmental waste generated from businesses. This has introduced more complexities in
operations management (OM) for the firms. This also increases the cost of the operations. Scare
resources will increase the cost of various inputs for the business organizations. Increasing
difficulties in handling disposals will increase the supply chain costs for an organization.
Therefore, the economic viability of organizations may be threatened if the issue of sustainability
is not factored into organizations.
A widely adopted definition of sustainability is ‘Development that meets the needs of the present without
compromising the ability of future generations to meet their needs’.
Economic sustainability refers to the ability of business organizations to deliver products and
services that fetches revenues in excess of the cost. At the plant level, it is operationalized as
production or manufacturing costs. OM practices are at the core of economic sustainability.
Since organizations use a variety of resources and engage in design, production, distribution,
recycling, and recovery of used products and services, they will need to fine tune their practices
to address this. Productivity enhancing initiatives, choice of design, material, and technology that
can reduce the cost are aspects of economic sustainability.
Environmental sustainability refers to the ability of organizations to deliver products and
services without compromising the balance between natural systems and living beings. Only
with such a balance, natural resources will be available over a longer period of time. Merely
pursuing an agenda of economic sustainability may not be feasible for several reasons. The
scarcity of available resources and handling disposal of end-of-life products and services will
pose greater threats even to economic sustainability. Moreover, firms will have to respond to
governmental regulations. Environmental sustainability is often related to waste reduction,
pollution reduction, energy efficiency, emissions reduction, a decrease in the consumption of
hazardous or harmful or toxic materials, a decrease in the frequency of environmental accidents,
and so on.
Therefore, OM practices will have to change to address these aspects. They may have to
account for the energy and other resources they use and the resulting ecological footprint they
leave behind. Supply Chain structure may also need some change to respond to these
requirements. The new product design philosophy may also require some change to address
these.
Social sustainability will relate to the choices made by business organizations that will enable
the current and future generations to live a healthy life. What it would mean is that companies
need to operate in a prudent and responsible manner and take care of employee health and safety,
and the quality of life of the external community. As operations is one of the areas employing the
most personnel and having the highest footprint and impact on the external community, it can
have a significant effect on sustainability’s social dimension.
The notion of sustainability is better understood from the transition required in the mindset of
business organizations. Figure 3.1 illustrates this idea pictorially. Sustainability is not achieved
by independently pursuing the three dimensions. This is because there are interdependencies
between the three, which finally drives the sustainability. Therefore, the current understanding of
the notion of sustainability is the intersection of economic, social, and environmental
dimensions. However, this understanding may not fully capture essence of sustainability. In
order to be truly sustainable, the interactions between the three must undergo a change. The
natural systems truly provide the overarching framework in which organizations can design
economic systems to address the demands of the social systems, thereby creating sustainable
entities for the future. OM gives new opportunities to significantly contribute to sustainability.
Sustainable OM is defined as the set of planning and execution methodologies that allow a
company to structure its business processes to achieve sustainable performance.
FIGURE 3.1 Alternative perspectives on sustainability
Incorporating sustainability ideas in operations require that several OM practices are modified.
For instance, there may be certain changes required in the area of supply chain design. While
new products may make use of the traditional supply chains, organizations may have to
configure a new supply chain for taking back used products. Similarly, new product development
and product design principles may require alternative perspectives. In the area of manufacturing
and process planning, technology choices must support clean technologies. Plant layout designs
could be done in such a fashion that renewable energy sources are deployed to the fullest extent
possible.
Incorporating sustainability ideas in operations require that several operations management practices are
modified.
In order to make these choices in OM, organizations need a basic framework for developing
specific options in each area of operations. Figure 3.2 provides a generic framework for
developing sustainable OM practices.
Avoidance is the first strategy in developing sustainable operations. What it means is that
every effort is required to avoid use of resources while delivering products and services. This
requires that the new product development process makes important changes in the design of the
product or service offered so that resources are not used in the first place. For instance, the
design of a new service may ensure that paper work is minimized by capturing data directly in
digital form. Use of electronic tickets for travel in Indian Railways is an example that promotes
avoidance. Similarly, in the design of a public library, innovative design of building, windows,
and reliefs can result in minimum use of lights during day time as natural daylight can be
harvested very well.
When Lucas TVS supplied starter motors to the nearby Hyundai plant, the earlier practice of
individually packing each motor in a plastic cover and a carton and then packing dozens of such
cartons in bigger cartons was dispensed with. Instead, trollies with hooks that can hold 50–100
starter motors were specially designed. The finished starter motors were simply hanged on the
trolley and wheeled into the Hyundai factory. This completely avoided the use of plastic and
cartons, which were strewn and thrown away at the Hyundai factory. A strategy of avoidance
puts an organization on a sound footing on matters pertaining to sustainability.
Reduction in usage is the next step where avoidance of resource usage is not possible.
Implementing this strategy in operations requires closer coordination and better planning among
different entities in the supply chain. For instance, logistics operations can be redesigned in such
a manner that fewer trips can be made between supply points and demand points by pooling
requirements either by origin or destination. The Japanese practice of frequent deliveries,
popularly known as milk runs will be evaluated in the larger scheme of clean technology
solutions for the society. Implementation of sensor-based mechanisms to cut-off the use of
resources automatically when not needed, can greatly help in the strategy of reduction. Value
engineering efforts in design will have the dimension of reduction of critical resources for
manufacture, maintenance, and use of scarce resources.
Since most part of services is about handling information, service design will emphasize use of digital resources
than physical resources.
Switching to renewable sources is another strategy for sustainability. Design of factory layouts
are undergoing change in line with the principle of sustainability. Passive cooling systems and
innovative air circulation design in the shop floor, use of special material for the roof to harvest
natural lighting for the factory, exploiting the vast area of roofing and terrace to harvest solar
energy through heating and photovoltaic systems, and greater investments in water harvesting
methods will be some of the features pertaining to the design of factories in the context of
sustainability. Since most part of services is about handling information, service design will
emphasize the use of digital resources than physical resources. Music, travel, entertainment,
publishing and education, banking and financial services, and several other sectors will need to
use electronic market infrastructure to replace non-renewable resources in the chain with digital
resources that are relatively more sustainable.
Extended use of products and services will be a major departure in the design philosophy
while new generation of products and services are developed. The current dominant paradigm in
business is to promote throw-away culture and use it as a basis for running the consumption
engine. Designing for planned obsolescence and perceived obsolescence assumes that the world
has unlimited resources, and it will be possible to keep tapping them endlessly. As more and
more resources are consumed, it will create newer challenges arising out of depletion of
resources and problem of disposal of wastes. Therefore, a logical direction for sustainable
operations is to extend the use of resources. In the future, it may prove to be a cost-effective
strategy also.
Reuse of old products is another dimension in creating sustainable operations. As consumption
levels increase, it will be impossible to dispose of used and unwanted items. Therefore, steps will
be required to take-back used material and discover economic value out of them. Increasingly,
governmental regulations are forcing business organizations to take this approach to manage
operations. Major changes may be required in several areas of OM to practice this strategy.
Reverse supply chains will be as important as forward supply chains. Inventory planning and
control practices will have to be modified as there will be two sources of supply for several
components (used as well as new). Remanufacturing, refurbishing, and recycling will be
important elements in an operations system. Disassembly of used products will be as important
as assembly of new products in a manufacturing system. One possibility to address avoidance
and reduction of use of resources is to extend the notion of total quality management to the
domain of pollution. What it may mean is that organizations will invest more on pollution
prevention than pollution control. Pollution control is ex-post. After creating waste and polluting
the environment, it needs to be cleaned through a set of activities called pollution control. On the
other hand, in pollution prevention, by a proper set of choices in design, process technology and
manufacturing methods efforts are made to reduce waste and energy use. Therefore,
implementing a sustainable OM system may involve newer systems, skills, alternative
philosophy, and approach to several OM practices. There will be benefits for the organizations,
customers, and society as a whole in the long run.
One possibility to address avoidance and reduction of use of resources is to extend the notion of total quality
management to the domain of pollution.
Traditionally business organizations have been concerned only about the forward logistics.
Forward logistics begins from the raw material supplier and ends when the finished product or
service is delivered to the ultimate customer. In between, it may involve multiple stages of
component manufacturing, assembly, packaging, logistics, and distribution through a chain of
trading partners. In order to address the issue of sustainability, it is important to address the
reverse logistics as well. As the name implies reverse logistics begins at the customer end and
eventually ends at the original equipment manufacturer (OEM). If a reverse logistics chain exists,
then instead of simply discarding an equipment after useful life, the customer may return it back
to the producer. It typically involves several entities, options, and decision points. Figure 3.3
represents a simplified framework for reverse logistics. We can use a three-dimensional
classification to describe various aspects of reverse logistics: networks, decision options, and
entities.
We can use a three-dimensional classification to describe various aspects of reverse logistics: networks, decision
options, and entities.
In an organization promoting sustainability as a key value, the design efforts must focus on minimizing not only
pollution from manufacturing but also all environmental impacts associated with the full lifecycle of a product.
Design for environment (DFE) is a structured set of methodologies that enable an organization
create products and services that are easier to recover, reuse, or recycle. The DFE approach
forces the designer to examine the effects of a product and service on the environment and
provide for necessary changes in the design. Such an approach will balance the manufacturing
considerations with ease of use, post-use disposal, ease of remanufacturing, and consumption of
precious resources during the entire life time of the product from concept stage to final disposal.
This is typically referred to as cradle-to-grave analysis. Taking such approach would mean for
example that the single use camera will be suitably redesigned such that it is possible to
remanufacture the returned cameras after use by the customers.
Since the focus is on the entire lifecycle of a product, a structured methodology to assess the
environmental impact over the lifecycle may be required. For instance, Nokia uses an externally
audited lifecycle assessment methodology to calculate the environmental impact of their products
and processes. The assessments will span the entire mobile device lifecycle, from raw material
acquisition to the end of the product’s life. They also assess energy efficiency, sustainable use of
materials, and small packaging. It helps them identify and focus on the areas where we biggest
reductions are possible. As a result of these Nokia has been active in introducing new, more
sustainable materials such as bio plastics, bio paints, and recycled metals into their devices. The
Nokia 700 uses bio-materials, recycled plastics, and recycled metals. This materials innovation
enabled them to reduce their dependency on fossil raw materials and the need for virgin metals,
use less energy in raw material acquisition, and introduce more sustainable industry practices.
Remanufacturing is the process of restoring end-of-life products (cores), components, modules, and parts to an
almost new condition in a manufacturing environment.
At the end of the assessment, the participating organizations will be given a feedback on the
strengths and opportunities for improvement. Comparative analysis with other companies in
the country will also be shared.
There are three levels of recognition based on the score obtained by an organization.
Organizations getting 900 and over are classified as leaders and those between 800 and 900
as challengers. The organizations that score between 700 and 800 are designated as believers.
All companies that qualify in any of the three bands will be recognized with their respective
‘Certificate of Merit’.
Source: Based on http://www.frost.com/prod/servlet/summits-details-summary.pag?as=attend&eventid=283496670. Last accessed on August 5, 2014.
A periodic review inventory control system works on the basis of reviewing the inventory in
fixed intervals, known as review period (R). During every such review, decisions are made with
respect to the quantity of material to be ordered so that a firm incurs minimum cost of the overall
inventory plan. Similar to the periodic review inventory control for manufacturing inventory (see
Chapter 17, Section 17.6 for details), there will be an order-up-to level, S, which will eventually
influence the inventory ordering decision.
The remanufacturing production process could be controlled by a periodic review, push policy
that operates as follows. Every R periods, all returned inventory is pushed into the
remanufacturing facility. Let this quantity be denoted by Qr. Since returns are uncertain, this
quantity will vary from one time period to another. IR denotes the inventory position of the
finished goods available for market consumption. Inventory position is the inventory on hand,
less backorders, plus any outstanding (manufacturing or remanufacturing) orders. If, after
releasing the remanufacturing batch, IR is less than the manufacturing order-up-to level S, then an
order Qm, is placed in the manufacturing facility to bring inventory position up to S. Therefore,
Qm = max (0, S − IR). Figure 3.5 illustrates the working of the system. The total relevant cost
(TC) for the plan is the sum of inventory carrying costs for returned products and finished goods
and backorder costs. The decision variables S and R determine the overall performance of the
system.
Establishing parameters R and S are critical for effective functioning of the system. Choosing
R is relatively easier of the two. R can be chosen on the basis of setup times and batch sizes using
an economic lot sizing algorithm. Alternatively, other managerial considerations such as
preferred reviewing cycles could be the basis for fixing R. However, there are no straightforward
methods to fix S. This is because both the demand for the finished goods in the market as well as
the quantum of returned goods are stochastic. Therefore, use of heuristic methods and simulation
can help identify a near optimal value for S. One can use the traditional models and make simple
intuitive adjustments to them in order to suit the current requirement. Such attempts often
provide promising results and enable practitioners fix the parameter S easily.2
The remanufacturing production process could be controlled by a periodic review, push policy of inventory
control.
Despite the mounting importance of sustainability, organizations are still making scanty efforts
to inculcate a culture of sustainability. Strategy making process and the operational choices in
functional areas of business are still not emphasizing sustainability. There are several reasons
attributed to these.
Mindset Inertia
Organizations are used certain ways of working. The new changes demanded by sustainability
norms force them to change their mindset with respect to many activities that they perform. For
example, the design philosophy must undergo major changes as we saw already. Management
control systems must promote use of better materials and recycling and remanufacturing
activities. These changes are less probably to happen immediately as there is no perceived value
in these in a direct fashion. Operations managers and design engineers will see these as needless
new complications in their ways of working.
The new changes demanded by sustainability norms will force operations managers and design engineers to
change their mindset with respect to many activities that they perform.
Creating a sustainable enterprise requires substantial new investments. A traditional pay-back and quick ROI
methods of investment justification will fail miserably and block introduction of the proposed changes.
SUMMARY
The economic viability of organizations may be threatened if the issue of sustainability is not factored into
organizations.
Sustainability pertains to development that meets the needs of the present without compromising the ability of
future generations to meet their needs. From a business perspective, the operational definition of sustainability will
encompass three dimensions: social, environmental, and economic.
Implementing a sustainable operations management (OM) system may involve newer systems, skills, alternative
philosophy, and approach to several OM practices in an organization.
In an organization promoting sustainability as a key value, the design efforts must focus on minimizing not only
pollution from manufacturing but also all environmental impacts associated with the full lifecycle of a product.
Design for environment is a structured set of methodologies that enable an organization create products and
services that are easier to recover, reuse, or recycle.
Reverse logistics begins at the customer end and eventually ends at the original equipment manufacturer. It consists
of a product take-back network and a product recovery network. There are several entities involved in reverse logistics.
Remanufacturing is the process of restoring end-of-life products (cores), components, modules, and parts to an
almost new condition in a manufacturing environment. For a remanufacturing operation to function effectively, an
organization must be concerned about the reverse flow of returned products from end users.
The remanufacturing production process could be controlled by a periodic review, push policy of inventory control.
Despite the mounting importance of sustainability, organizations are still making scanty efforts to inculcate a
culture of sustainability. There are several reasons attributed to these.
REVIEW QUESTIONS
1. What do you understand by the term sustainability? What are the dimensions of sustainability that an organization must focus
on?
2. Do you agree with the statement, ‘Organizations may greatly benefit from creating sustainable operations’? Substantiate your
argument with examples.
3. What do you mean by sustainable operations?
4. Identify three areas of operation management that will undergo significant changes in an organization promoting
sustainability.
5. Why should organization perform a ‘cradle-to-grave analysis’ while designing new products and services? 6. How does a
product take-back network promote sustainability in operations?
7. Distinguish between the following terms:
8. What are the impediments an organization faces in its journey of creating sustainable operations? How can these be
addressed?
NET-WISE EXERCISES
The Nokia People & Planet Report covers the key ethical, socioeconomic, and environmental
issues relevant to Nokia’s business and stakeholders. It covers Nokia’s approach to sustainability
and the focus areas, and reports their progress against the set targets. Visit the URL
http://company.nokia.com/en/about-us/people-planet/sustainability-reports to access the
sustainability reports. Read the Nokia People & Planet Report 2013 to understand various steps
taken by Nokia with respect to sustainability.
ITC believes that a company’s performance must be measured by its Triple Bottom Line
contribution to building economic, social, and environmental capitals. Adopting this approach,
the company has unleashed multiple drivers of growth in India, ensuring long-term sustainability
and competitiveness of its businesses. It is a voluntary disclosure initiative on the part of ITC
Ltd. Read the ITC Sustainability Report 2013 by visiting the URL
http://www.itcportal.com/sustainability/sustainability-report-2013/sustainability-report-
2013.pdf.
Based on your understanding of the issues discussed in these reports answer the following
questions:
a. How does Nokia and ITC approach the issue of sustainability? Compare and contrast their approaches to the issue of
sustainability.
b. Identify the major operations management practices that have been affected by sustainability at Nokia and ITC.
c. Comment on the measures of performance chosen by Nokia and ITC to measure their progress on matters pertaining to
sustainability.
SUGGESTED READINGS
Project Management
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
How does the planning and control of projects differ from other operating systems?
What are the various phases in project management? What are the organizational mechanisms employed for
effective planning and control of operations in a project?
What are the elements of a network analysis of projects? What are the alternative methods by which a project can
be graphically depicted for the purpose of analysis?
What do you understand by the critical path method (CPM) and the programme evaluation and review technique
(PERT)? How can various aspects of CPM and PERT be used for decision making in project planning and control?
There are several occasions when organizations perform large-scale activities in a non-repetitive manner. These include
building a manufacturing facility, designing new products, constructing bridges and roads, and designing new services. Project
management helps in the planning and control of these activities by providing the required tools and techniques.
ideas at Work 4.1
Building such a mammoth airport requires a different set of tools and techniques adopted
by a typical manufacturing or a service company engaged in known set of activities which
are typically repetitive in nature. Further, it calls for greater degree of coordination among
multiple stake holders spanning over several years. The work will be characterized by large
size and complexity leading to unprecedented levels of uncertainty. The collective set of
tools used to handle such situations is called project management. We shall see more details
in this chapter.
Source: http://www.newdelhiairport.in/master-plan.aspx, http://www.newdelhiairport.in/our-company.aspx
Project management offers alternative tools and techniques to address the planning and control issues
pertaining to large-scale activities performed in a non-repetitive manner.
There are several occasions when organizations perform large-scale activities in a non-repetitive
manner. These include building a manufacturing and service facilities, designing new products,
the construction of bridges and roads, and designing new services. These activities differ greatly
from the normal manufacturing and service operations. Typically, these are one-off activities
done at less frequent intervals. Several sets of activities need to be performed in a particular
order to accomplish the task. Moreover, the complex resource and time requirements for
performing these activities, as well as the possibility of simultaneous occurrence of events,
requires one to deploy systematic methods for planning and control. Moreover, the planning and
control of these activities requires a different set of tools. Project management offers alternative
tools and techniques to handle the planning and control of the activities described above and we
focus on this issue in this chapter.
Before we discuss the available tools and techniques for project management, it is important to
understand how project organizations differ from other organizations. Such an understanding is
important, as it provides some guidelines for managers to effectively plan, monitor, and control
these organizations. We identify four attributes to characterize a project organization.
1. Different from mainstream activities: Projects involve activities that are clearly different from mainstream activities.
Take, for example, the designing of a new product. The type of activities involved in designing a new product is
different from other regular activities that an organization may perform, such as planning for production, purchase of
material, or production scheduling. As we will see in Chapter 11, new-product development involves understanding
customer needs, market offerings, the technology available at the time of design, and the methods to be utilized for
designing and manufacturing the product. Normally, organizations put together a team of resource persons to
accomplish this task to meet time and cost targets. Clearly, these activities are different from the mainstream and tend to
be non-repetitive in nature.
2. No expertise currently available: In several organizations, when certain types of expertise are not available at a time, a
committed set of resources and people are required to acquire the new expertise in a finite time. Achieving this target
has several features common to the earlier example of new-product development. It calls for planning activities that are
different from the mainstream. Think about Chandrayaan, the lunar mission of the Indian Space Research Organisation
(ISRO), which was launched in October 2008. When the project started, ISRO would have had very little expertise for
Chandrayaan. Similarly, think about how ISRO develops its next generation of satellites and satellite launchers through
a systematic process of acquiring the required technology, talent, and resources. The manner in which they do this is
different from the normal production methods employed in any manufacturing organization. In such a situation, the
organizational forms, the performance standards, and the planning and control tools required are likely to be different
from the normal production-oriented tools. Having a project-oriented organization fulfils the requirements for achieving
this objective.
In several other situations, although a new product or service is available, the areas of competency are not yet
demonstrated. In such cases, sustained dedication of resources and a time-bound approach to acquire the competency is
required. For example, in an organization that has recently purchased a new gas turbine for in-house generation of
power, installation and maintenance of gas turbine is done using alternative planning and control tools until sufficient
experience is gained in handling the equipment. Normal methods of performance evaluation and control are not
applicable to this situation until the personnel acquire sufficient competency.
3. The product or service offered is large scale: When the product or service offered is of very large scale, it introduces
problems that are normally not encountered in regular operating systems. Therefore, the tools already available for
operations management are not appropriate for the planning and control of such systems. Very large systems introduce
the following additional problems that require alternative methods for planning and control:
a. Going from a subsystem to a total system: Consider a situation in which the quantum of effort required to
complete the task is large, as in the case of building a multi-storied commercial complex of over 1,000,000
square feet. Such a case involves multiple entities (architect, engineers, government regulatory bodies,
construction workers, the local community and society, and so on). It also involves multiple stages of the
process (designing of the building, the foundation, the superstructure, the electrical systems, the heating and
air-conditioning system, the carpentry, the plumbing, fabrication activities, acoustics, landscaping, horticulture,
and so on). Managing such a large set of activities requires different methods of planning and control. It is
useful to break down the set of activities into several subsystems and provide methods of linking one
subsystem with the other.
b. Long lead times: The characteristics we just discussed invariably also indicate long time periods for
completion. For example, in the case of developing the next generation of satellites by ISRO, the total duration
will be anywhere between five and eight years. Similarly, the time required from developing the initial concept
to commercial production of a passenger car could be in the order of two to three years. Managing activities
over such a long span of time is not possible with the traditional tools available for production planning and
control. Therefore, alternative tools and techniques are required. Moreover, it is also desirable to break down
the entire set of activities and classify them under shorter time periods for the purpose of planning and control.
4. High degree of customization: Several manufacturing and service systems may offer a smaller number of products and
services. Yet, there could be considerable amount of customization involved in the process. The Nuclear Engineering
Division at Bharat Heavy Electricals Limited (BHEL), Tiruchchirappalli, has been manufacturing several components
for nuclear reactors for the last two decades. Yet, each product that it delivers to a customer could vary substantially
from the earlier orders. The manufacture of power plants and large power transmission systems are similar examples.
Planning and control tools used for repetitive manufacturing are not appropriate for these activities.
There are several common features in these situations. First, the degree of uncertainty in all these
cases is likely to be significant. Planning tools that incorporate this uncertainty are greatly
required to manage such situations. Second, the number of entities involved in accomplishing the
various tasks is many. Furthermore, these entities have numerous complex interactions among
them. The performance standards of one entity or deviations from the specifications at one part
of the system have a significant impact on the other. If any set of activities involved in an
organization has any of the above attributes, then such activities could be planned and controlled
using tools and techniques collectively known as project management.
Project activities are likely to be unique and may involve high levels of subcontracting, breaking down the
system into several subsystems, and managing these and their interfaces.
Figure 4.1 brings out the salient features of project activities. Project activities are likely to be
unique and may involve high levels of subcontracting, breaking down the system into several
subsystems, and managing these and their interfaces. Due to this, the risk and uncertainty of task
completion are likely to be high and may call for appropriate organizational mechanisms to
mitigate these. Appropriate organizational forms that promote teamwork—not only among the
various functional areas within the project organization but also between other organizations and
participating subcontractors—will be an important requirement.
FIGURE 4.1 Salient features of project activities
Managing any project involves four key phases (see Figure 4.2). In the first phase, activities
focus on conceptual details pertaining to the project. The broad objectives of the proposed
project are first set. After that, based on a conceptual design and an initial project appraisal, the
feasibility of the project is established. This is followed by a budgeting exercise. Targets for the
time and the cost of the project are also established during this phase.
Managing any project involves four key phases: conceptual detailing, planning, implementation, and post-
project appraisal.
Let us consider the case of setting up a Mass Rapid Transit System (MRTS) for the city of
Bangalore as an example. In the conceptual phase, the project promoter—in this case Bangalore
Metro Rail Corporation Limited (BMRCL)—will have to arrive at the objectives of the project.
The objective could be to provide alternative channels for the public to travel in the city or an
affordable mode of transport. On the other hand, the objectives could also be reducing the
pollution levels in the city or de-congesting the main business areas in the city by January 2016.
Depending on the objective, various elements of the design should be worked out. For instance,
meeting the objective of providing an affordable mode of transport may influence the choice of
technology, the time frame for completion, major elements of design, and the budgeting
differently, as compared to meeting the objective of de-congesting the city by January 2016.
Planning constitutes the second phase in project management. In this phase, detailed planning
is carried out and milestones for completion, review, and control of the project are established.
Since projects are often huge and get completed over several years, it is useful to break it down
into smaller systems. Furthermore, to have tighter control and monitoring, a hierarchy of levels
needs to be established in the project. For each of these levels, targets for completion time, cost,
and an organization structure could be established. At the detailed planning level, activities that
need to be performed along with their interaction with other activities would have been
identified. Using this information, a network or a graphical representation of the project could be
made and used for the purpose of planning and control.
In our example, the total distance to be covered by the MRTS would be split into three or four
stages of completion. For each stage, the activities would be grouped on some basis. Acquiring
land for the project would have been one group, civil engineering activities another, electro-
mechanical could be the third, track-laying and signalling could be another, and so on. Within
each group there could be further subdivisions: For example, under civil engineering,
constructing stations and amenities could be one subgroup and building bridges could be another.
Once such groups are identified, the resources, time, and cost involved for each activity could be
estimated. Finally, all these could be represented in the form of a graph and used for planning,
analysis, monitoring, and control.
FIGURE 4.2 The four phases of project management
In the implementation phase, the focus shifts to the execution of the plans. Managing the
resources and the interactions among a large number of contracting organizations is one main
activity. Measuring the performance of the project and its progress and reacting to real time
issues such as time and cost overruns are done during this phase. More resources are sometimes
employed to hasten the project and restore schedule adherence. Since the degree of uncertainty in
projects is likely to be high, certain corrective measures need to be taken to bring the project
back on track. A good performance-monitoring and review system is required to take several
decisions during the implementation phase.
Returning to our example, in this phase, cost and time targets established for each subsystem
and activity of the MRTS project would be constantly monitored. Completion of bridges might
have been scheduled for September 30, 2014. If this activity gets delayed, then the project review
and monitoring system would bring it to the attention of the decision makers. Alternatives will be
evaluated to bring the project back on track. These may, for instance, include employing more
people at the site, resorting to alternative manpower planning, revising cost estimates and in
some cases, the completion date itself.
Projects often tend to be one-off activities. Nevertheless, several elements of planning and
control employed in a project could be utilized in future projects. Therefore, post-project
appraisal is the logical end and is important in any project. The key learnings from a project need
to be documented, analysed and archived for future reference. Such a practice pushes an
organization up in the learning curve, no matter how unique the executed projects are.
While the previous section details the various steps required for project management, it does not
show how exactly an organization could go about the task. A good framework for project
management must answer some key questions that a manager faces while planning and executing
projects (see Figure 4.3):
How should the large set of activities be systematically broken down and used as the basis for planning and control?
How should the manager allocate the resources for the project (that is, who should be responsible for execution and
monitoring of various activities in a project)?
What are the targeted costs for completion of each element of the project?
When should each of these elements be scheduled for execution?
How well has the project been progressing?
Work breakdown structure (WBS) is an organized methodology used to split an overall project into non-
overlapping components and identify deliverables for each component.
A good framework for project management will provide the project manager with tools and
techniques for answering these questions.
We shall look at these questions in some detail and understand the methods employed for
answering them.
FIGURE 4.4 A partial list of a work breakdown structure for a construction project
WBS serves the important task of providing a clarified view of the activities to be performed
in a project and provides the reference for estimating activity duration, network representation of
the project, planning, monitoring, and control. It also helps the manager to understand the
interaction among the activities and the need to manage the interface points as the project
progresses. Furthermore, it provides a basis for creating an appropriate project organization and
establishing milestones and control points for project evaluation. Figure 4.4 shows a partial list
of a work breakdown structure for a three-storey commercial building construction project.
An organization breakdown structure (OBS) identifies an appropriate organizational structure to execute the
tasks listed under each work package.
At the highest level, the OBS essentially addresses an appropriate form of organizational
structure for a project. A matrix type of organization is often found to be appropriate for project
organizations at this level. Figure 4.5 depicts a typical matrix organization structure. In a matrix
organization structure, a project manager has a dual role.3 With respect to project completion and
targets, a project manager has a subordinate relationship with the VP (Projects). On the other
hand, on functional matters of how the project needs to be executed, the project manager has a
direct reporting relationship with the chief of the respective functional area. Such an organization
structure enables the optimal utilization of specialized resources in multiple projects. It also
facilitates managing the interface areas identified through a WBS.
FIGURE 4.5 Matrix structure for a project organization
Cost breakdown structure (CBS) is a methodology to link the individual elements in a WBS to a dimension of
cost.
FIGURE 4.6 WBS–OBS–CBS: A three-dimensional perspective Adapted from: F. Harrison and D. Lock, Advanced Project
Management, (Gower Publishing Company, 2004).
4.4 TOOLS AND TECHNIQUES FOR PROJECT MANAGEMENT
We shall now take a look at the various tools and techniques that a manager can use in project
management.4 In the rest of the chapter, we shall focus on the following important areas of
application:
a. Developing a network representation of a project
b. Analysing project networks
c. Addressing time and resource constraints in projects
d. Handling uncertainty in project completion
The network representation provides a visual aid to a manager in understanding various activities involved and
their interactions with other activities.
Activity precedence
The elementary unit of analysis in a project network is an activity. An activity constitutes an
amount of work done towards the completion of the project. Each activity consumes a set of
resources and takes a finite amount of time for completion. Moreover, every activity may have
certain technological or logical constraints that link it with its preceding and succeeding set of
activities. Clearly, activities constitute the basic building blocks of any project network. A simple
example will clarify the above characteristics of an activity.
Each of these can be considered as an activity. It is easy to recognize that each of these activities
consumes some resources and time for completion. There are also some constraints in the order
in which the activities could be performed. For example, without understanding the market
needs, it is not possible to develop the conceptual design. Likewise, without the conceptual
design, it is not possible to develop the detailed design. On the other hand, some activities can be
performed simultaneously. While planning for production after developing the detailed design,
the organization could also initiate the process of creating the required marketing infrastructure.
In other situations, two or more activities should be completed before an activity can be taken up.
In our example, unless production is planned and the required marketing infrastructure is created,
it may not be possible to make the product reach the target market ultimately. Based on an
understanding of the problem and the constraints, one can develop a table depicting the
predecessors and successors for each activity. Table 4.2 has this information for our example.
Two conventions exist in constructing a network: representing the activity on arc (AOA) and representing the
activity on node (AON).
In AOA, as the name implies, activities are represented on the arc. Since a pair of nodes
connects each arc, the start node for the arc represents the beginning of the activity and the end
node, the end of the activity. Therefore, each node represents a milestone in the completion of
the project. A pair of successive nodes can be connected by only one unidirectional arc.
Therefore, sometimes it is necessary to have dummy activities in a network. Dummy activities
are fictitious entities having zero duration. They do not affect the project completion schedules or
violate precedence and succession constraints.
In contrast, in an AON, each node represents the activity itself. The arc from one node to the
other represents the precedence relationship between the activities. An AON network requires
one “start” node and an “end” node. Although an AON is much simpler to construct, AOA was
the first to be developed and widely used until recently.5 We shall illustrate these concepts with
Examples 4.1 and 4.2.
The path that has the maximum duration is the critical path.
EXAMPLE 4.1
Solution
Figure 4.8 shows the AOA network. Activity B has A as the predecessor and C and D as
successors. Similarly, Activity F has D as predecessor and H as successor. These
representations are fairly straightforward. Let us consider Activity E, which has two
predecessors, C and D. Since two arcs cannot connect Nodes 3 and 4, we direct Activity D to
another node, Node 5, and connect Nodes 4 and 5 using a dummy activity (shown by dotted
lines). This is the only way to satisfy the precedence relationship.
As explained earlier, the AON network is simpler to construct. The AON network always
starts with a “start” node and ends with an “end” node. All intermediate nodes are
constructed keeping in mind the predecessor and successor relationships among the activities.
The AON network is given in Figure 4.9.
FIGURE 4.8 The AOA network
EXAMPLE 4.2
Cool Homes (P) Ltd specializes in constructing quality homes in the city of Bangalore. Cool
Homes is known for timely completion of its projects. Consider one of the projects of Cool
Homes. Table 4.4 lists the activities, their precedence relationship with other activities and
their duration. Construct a network for the project, clearly depicting the activities and their
inter-relationships.
Solution6
Figure 4.10 has the AOA representation for the problem. In the network representation, each
activity is followed by a number that represents the duration of the activity in weeks. There
are parallel activities in the project.
TABLE 4.4 The Activities, their Precedence Relationship with Other Activities and their Duration
Note: In the figure, the numbers following each activity represent the duration of the activity
For instance, wall plastering/laying floors (Activity G) and obtaining power and water
connections for the site (Activity I) could proceed simultaneously. On the other hand, some
activities can be started only when more than one activity is completed. Final testing and
handover (Activity J) can be done only when power and water connections are obtained
(Activity I) and sanitary and electrical fittings are fixed in the house (Activity H).
These interactions and simultaneous occurrences throw up both opportunities and challenges
for planning the project completion, as we shall see later.
Slack is the difference between the late start and early start schedules of an activity.
This methodology of computing the critical path, the early and late schedules for activities,
and the slack and using this information for addressing resource- and cost-based issues is known
as the critical path method (CPM). The formulae for computing the various schedules, slack,
and project completion are as follows:
Early finish time (EF) for Activity i = EF(i) = ES(i) + t(i), where t(i) is the activity duration.
Late start time (LS) for Activity i: LS(i) = LF(i) − t(i), where t(i) is the activity duration.
Total slack (TS) for Activity i: TS(i) = LF(i) − EF(i), or LS(i) − ES(i)
The critical path method (CPM) is a methodology of computing the critical path, the early and late schedules
for activities, and the slack and using this information for addressing resource- and cost-based issues.
In Figure 4.11, we have the network for Example 4.2 with early and late schedules for the
activities. The critical path is also shown in the network with a bold line. The accompanying
Table 4.5 has the early start/early finish and the late start/late finish schedules for all activities.
FIGURE 4.11 The network for Example 4.2 with early and late schedules
The basic analysis of projects gives the necessary information to a manager to understand the
likely schedules for completion of projects and the options available for the manager to shift
some of the activities without compromising on the project completion. We shall see how
managers could use this information to fine-tune their planning while reacting to changes that
were originally not anticipated. For instance, resource availability could pose some additional
constraints in starting certain activities at a particular time. On the other hand, there could be
delays in some of the activities, jeopardizing the project completion and introducing more
activities into the critical path. We shall address these two common concerns and introduce some
tools that are available to address these.
Resource Levelling
Earlier, we saw that information on slack could help managers in addressing resource issues in
an organization. In order to understand this, we need to know the type and the amount of
resources that each activity requires. Once this information is included in our analysis, we begin
to see additional information about the project. This includes the mismatches between
availability and requirement, the peaks and valleys in resource consumption, and the need to look
for better schedules that demand resources evenly. Example 4.3 will help us understand these
issues better.
EXAMPLE 4.3
Let us consider Example 4.2. Assume that Table 4.6 shows the manpower required for each
activity of the Cool Homes project. Based on this additional information, analyse the
following:
a. What impact will an early start schedule-based project have on resource requirements?
b. Is there a basis on which the resource requirement could be levelled? If so, what are the changes required in the
schedules of activities in the project?
Solution
In the absence of any additional information, every project manager would like to schedule
the project either with early start schedules or with late start schedules. Such an approach is
intuitively appealing and simple to set up. However, it can place significant demands on
resources. The accompanying network representation (Figure 4.12) has the early start
schedule for the project. The activities in the critical path are laid out at the centre. The
numbers indicate the weeks during which the activity ends and the dotted line indicates the
slack available.
During Weeks 9–12, both Activities B and C are scheduled, thereby demanding a total
manpower of 18 (15 + 3). During Weeks 21–28, Activity E is scheduled. While it progresses,
two other activities are scheduled during that time (Activity D during Weeks 21–22 and
Activity F during Weeks 23–24). Similarly, during Weeks 29–30, two activities are
scheduled (Activities G and H), whereas during Week 31 only Activity G continues. During
Weeks 32–33, only Activity H is scheduled. Based on this, the week-wise manpower
requirement can be computed. The resultant manpower loading pattern for the early start
schedule is plotted in Figure 4.13.
As is evident from the figure, there is considerable amount of unevenness in the use of
manpower. This may call for frequent hiring and firing of workers or frequent shifting of the
workers across other construction sites. Such frequent changes could potentially lead to poor
morale of the workers, eventually reflecting on the quality. Project management
organizations actively pursue alternative schedules to minimize such effects in projects.
TABLE 4.6 Manpower Required for the Cool Homes Project
FIGURE 4.12 Early start schedule for the Cool Homes project
FIGURE 4.13 Manpower loading pattern for early start schedule
It follows from the above that if the project duration is reduced by one unit time by crashing an
activity, there will be a reduction in the indirect cost to the extent of FC and increase in the
project cost to the extent given by Eq. 4.1 for the activity. Using this information for each
activity, it is possible to analyse the cost–time trade-off in project management. Example 4.4
illustrates these issues.
FIGURE 4.17 Cost–time trade-off in project management
EXAMPLE 4.4
Let us return to Example 4.2 and suppose additional information is available to the project
manager of Cool Homes regarding crashing of some activities in the project. Analyse the
time–cost trade-off involved in crashing the project duration and identify the optimal crash
length for the project. The relevant information is available in Table 4.7:
The indirect cost for the project is estimated to be ₹5000 per week.
Solution
The Cost of the Project With No Crashing
The critical path for the problem does not change. Therefore, the project duration is 35 weeks
(refer to Example 4.2 for details). Since the indirect cost for the project is ₹5,000 per week,
the total indirect cost for the project is ₹175,000. Also, from the information available in
Table 4.7, we can add the normal cost of all activities to get the direct cost for the project
when there is no crashing. The direct costs for the project is ₹108,000. Therefore, the total
cost of the project with no crashing is ₹283,000.
TABLE 4.7 Cost and Time Data for the Cool Homes Project With Estimates for Crashing
Cost Per Week of Crashing the Activities
From the data in Table 4.7, we see that three activities cannot be crashed. Therefore, our
focus is only on the remaining activity. We first compute the cost of crashing per week using
the data in Table 4.7 and Eq. 4.1. Let us consider Activity A. The normal cost is ₹12,000 and
the crash cost is ₹20,000. Since the activity could be crashed by two weeks, the cost per
week of crashing is ₹4,000. Table 4.8 has the computation for cost per week of crashing for
all the activities.
We repeat this procedure until no more activities exist for crashing the project. In our
example, although Activity F can be crashed, it does not have an impact on the project
duration, as it is not on the critical path. Therefore, we stop the procedure.
TABLE 4.9 The Cost–Time Computation Involved in Crashing the Project
Table 4.9 has the detailed computation for crashing. We see from the table that the
optimum duration for crashing is 32 weeks. By crashing the project beyond this, the cost of
the project increases. Similarly, if the duration of the project is longer than 32 weeks, the cost
of the project increases. The accompanying plot of these results graphically portrays the
cost–time tradeoff involved in crashing the project (see Figure 4.18).
FIGURE 4.18 Graphical depiction of cost–time trade-off
For reasons mentioned in the beginning of the chapter, the degree of uncertainty is likely to be
high in several projects. Therefore, project managers need additional tools for managing
uncertainty in projects. Uncertainty stems mainly from the non-repetitive nature of the activities
and the inability of managers to accurately estimate time and resource requirements for
completion of activities. In order to analyse the network when uncertainty in time estimates exist,
a framework known as the programme evaluation and review technique (PERT) was
developed during the Second World War. PERT assumes that the activities that constitute a
project and their relationships to other activities are known with certainty. However, there is
considerable difficulty in estimating the time required for an activity. PERT uses several
important assumptions in analysing projects with uncertainty. It is important to know them
before we proceed with the details of the technique.
The programme evaluation and review technique (PERT) enables managers to analyse the network when
uncertainty in time estimates exist.
1. PERT relies on the subjective expertise of the managers actively involved in such projects to estimate the duration of
activities, about which considerable uncertainty exists. Three time estimates are made for each activity:
a. Most likely estimate of activity duration: tm
b. Most optimistic estimate of activity duration: t0
c. Most pessimistic estimate of activity duration: tp
2. The activity duration is assumed to follow a beta distribution. Unlike a normal distribution, beta distribution has a
skewed right tail. The choice of beta distribution seems reasonable because unique and large-scale projects may have
occasionally large activity duration due to uncertainty. Figure 4.19 is a graphical representation of a beta distribution
with the three time estimates.
3. The expected time for each activity and the variance of the activity duration can be approximated using the following
equations:
4. Since mean is an unbiased estimator, the summation of the expected activity durations (te’s) in a path is the expected
duration of the project along the path. It is assumed that the expected duration of alternative paths follow a normal
distribution.11 For a similar reason, the variance of the duration of the project along a path is the sum of the variance of
the individual activities in the path.
5. For example, let us consider one path, Path p, in a network consisting of four activities, a, b, c and d. Then, by our
definition,
The expected project duration Tc is the critical path for the network and is given by:
6. Since the expected duration of the paths in a network follow normal distribution, it is possible to estimate the probability
of completion of the project for a given Path p, within a targeted due date D, using the standard normal variate given by:
Since Tc is the critical path for the project, the probability of completion of the project by a due
date is obtained by replacing the subscript p with c in Eq. 4.7.
Several of these assumptions are considered reasonable for practical applications. The
inaccuracies in these assumptions far outweigh the additional insights that the analysis provides
to a manager in handling the uncertainty. We shall use an example to demonstrate these ideas.
EXAMPLE 4.5
Let us return to Example 4.2. Suppose Cool Homes have no prior experience in the kind of
construction project that they have currently undertaken. Consequently, there is considerable
uncertainty in estimating the activity duration. Assume that the senior management at Cool
Homes has collectively brainstormed and arrived at the following estimates for time duration
(see Table 4.10).
Use the PERT technique to compute and analyse the following:
a. What is the expected duration for each activity in the network?
b. What is the variance of the duration for each activity in the network?
c. Compute the expected duration for all the five paths in the network.
d. What is the probability of completing the project by the end of Week 45?
e. What are the key inferences that one can draw from these results?
Solution
TABLE 4.11 The Expected Duration and Variance for Each Activity
TABLE 4.12 Computation of the Expected Duration and Variance for Each Path
D = 45
Tc = 40
From a standard normal table, the probability is obtained to be 0.879.
One can draw several key inferences from the above computations. The introduction of
uncertainty into a project management analysis framework tends to increase the expected
duration of the project. Moreover, one can no longer talk in terms of a definite completion
date but rather the probability of completing by a certain time. In our example, although the
expected project duration is 40 weeks, there is only a 50 per cent chance of completing the
project by that date. Even after 45 weeks, the probability of completion by that time is only
88 per cent. Although, not directly evident from our example, it appears that if the near
critical path had a much higher variance, then the probability of completion of this path could
be lower than the “official” critical path. These issues point to the fact that it is important to
understand that the distribution of completion times and probabilities is more meaningful
than a single completion date, as in the case of the deterministic case.
FIGURE 4.20 A graphical representation of the normal variate
Simulation of Project Networks
The above example clearly brings out issues pertaining to handling uncertainty in projects. The
focus in handling uncertainty in projects shifts from uniquely computing the project completion
and critical path to that of identifying tendencies for alternative paths and activities to influence
project completion. Therefore, a better understanding of the issues related to project management
comes from obtaining a distribution of critical paths, activity slacks and project completion
times.
One way to obtain this information is to perform a simulation of the network. Based on an
appropriate distribution for the duration of each activity, one can sample values for all activity
durations for one run of the simulation. Using the sample, one can estimate the activity slacks,
the probability of project completion by a desired due date, and identify the critical path.12 By
repeating this sampling and computation procedure it is possible to fit a distribution of key
information, including activity slacks, project completion probability, and duration. It is also
possible to establish a criticality index for each activity to indicate how often the activity fell in
the critical path. For example, let us consider a simulation run of 10,000. If Activity A was on
the critical path 7,500 times and Activity B was in the critical path 1,500 times, then the
criticality index for Activity A is 0.75 and that for Activity B is 0.15. Clearly, a particular
activity such as A, which has a higher criticality index, may need special managerial attention.
In PERT, a distribution of completion times and probabilities is more meaningful than a single completion date
as in the case of CPM.
A project plan is created in much the same fashion as with critical paths. The plan is worked
backwards from a completion date, with each task starting as late as possible. Two durations
are entered for each task: a “best guess” (typically with a 50 per cent probability duration, as
in the case of the CPM), and a “safe” duration (with a higher probability of completion,
depending on the amount of risk that the organization can accept). Resources are then
assigned to each task, and the plan is resource levelled using the 50 per cent estimates. The
longest sequence of resource-levelled tasks that lead from the beginning of the project to the
end is then identified as the critical chain.
Recognizing that tasks are more likely to take more time due to random events,
idiosyncrasies and statistical fluctuation “buffers” are used to establish dates for deliverables
and for monitoring project schedules and financial performance. The “extra” duration of each
task on the critical chain—the difference between the “safe” durations and the best guess—is
the buffer at the end of the project. In the same way, buffers are computed at the end of each
sequence of tasks that feed into the critical chain.*
Note: *For more details on critical chains, see http://www.critical-chain.co.uk/
SUMMARY
The planning and control of operations in project organizations differ from other operating systems on account of high
levels of uniqueness, risk, and uncertainty in the former.
Project management typically involves four phases. These include conception, planning, implementation and control,
and post-project appraisal.
A structured approach to project management involves a three-dimensional framework. The three dimensions are the
work breakdown structure, the organizational breakdown structure, and the cost breakdown structure.
Projects can be graphically portrayed using activity on node (AON) or activity on arc (AOA). Such a representation
helps the project manager perform certain analyses of projects.
The critical path method (CPM) helps understand several important aspects of controlling projects such as the likely
duration of the project, activities that could be scheduled with some flexibility, opportunities for resource levelling, and
identify activities that need to be crashed to reduce the project duration.
PERT enables a project manager analyse projects that have high degree uncertainty with respect to activity durations. It
provides a structured approach to estimate the probability of completion of the project by a target time.
Variance of activity
REVIEW QUESTIONS
1. Is managing a project system any different from other manufacturing and service systems? Why?
2. Recall our discussions in Chapter 4 about volume– variety–flow. Use these parameters and show how project organizations
can be described in these terms.
3. Give three examples each from manufacturing and service systems that call for project management techniques for planning
and control.
4. What is meant by work breakdown structure? Why do you need this in project management?
5. Consider the design and development of a new executive health check-up for preventive healthcare for senior executives.
Develop a two-level work breakdown structure for this project.
6. Why should the project manager compute the slack of activities in a project?
7. Will there be paths in a project network with zero slack for all the activities in that path? If so, what can one infer from that?
8. What are the key differences in using CPM and PERT in a project? List the key implications for the project manager.
9. Give three examples of situations that may require use of PERT for project management.
10. Projects involve direct as well as indirect costs and project managers need to use this information in project management.
Comment on this statement.
11. What is criticality index? How can project managers use this information?
PROBLEMS
1. A project consists of ten activities. Table 4.13 depicts the precedence relationships between these activities. Develop a
network representation of the project.
2. Consider a project consisting of nine activities. Table 4.14 gives the precedence relationship among the activities. Draw an
AOA and an AON network representing the project. What are your key observations of this exercise?
3. An organization is considering the launch of a new product in the national market. This project consists of ten major activities.
The precedence relationship and the estimated duration of each of the activity are given Table 4.15.
4. A project consists of seven activities. Table 4.16 depicts the precedence relationships between these activities. Develop a
network representation of the project and answer the questions given below:
5. A project management organization is preparing a network for one of the projects that it has recently bid for. The client
demands that the project needs to be completed and delivered at its site by Week 25. The organization has identified the list of
activities to be performed and estimated the duration of each of the activities. The details are given in Table 4.17. Use this
information to draw a network and help the organization answer the following questions:
a. Will the organization be able to meet with the customers’ deadline of 25 weeks?
b. If the organization cannot, identify the set of activities that need to be considered for a possible reduction in duration.
c. If the cost of reduction of duration is ₹10,000 per week and the organization will have to pay a penalty of ₹7,000 per
week of delay in completion in addition to paying a fixed amount of ₹20,000, what will your recommendation be to the
organization with respect to reducing the duration?
6. Consider a project having seven activities and the precedence relationship illustrated in Table 4.18.
a. Draw a network of the project and identify the critical path for the project.
b. Calculate the slack for the activities in the network.
c. Suppose the project manager is likely to face some problems arising out of resource constraints. In your opinion, which
of the activities could the manager consider for rescheduling to optimize the resource utilization?
7. In the metropolitan city of Bangalore, the Bangalore Development Authority is undertaking several construction projects.
Many of these projects involve construction of huge multi-level flyovers and the organizations involved in the work do not
have adequate prior experience of constructing similar flyovers. Moreover, the process is riddled with considerable
uncertainty due to political and public interference. Consequently, although estimation of activities is possible, estimation of
the duration of the project is difficult. Table 4.19 gives the list of activities, and three possible time estimates for the activities.
Use this information to analyse the problem of uncertainty by answering the following questions:
8. A manufacturing organization has received a turn-key project for manufacture of custom-made equipment for one of its
clients. The project involves several activities and Table 4.20 has the details on the activities, duration and normal and
crashing costs for the activities. Since it is a turnkey project, the organization incurs a cost of ₹6,000 per week towards special
supervision of the project. This cost is incurred for the length of the project duration. The project manager would like to know
if there is some merit in crashing the project duration. Evaluate the cost–time trade-offs involved in the project and suitably
advise the project manager.
a. Prepare a manpower loading chart on the basis of early start schedules for the project.
b. Are you satisfied with the loading pattern? If not, devise an alternative schedule for the manpower loading.
c. The organization maintains a constant workforce of 70 workers and hires and lays off workers depending on the
requirement on a week-by-week basis. If the hiring charges are ₹2,500 per worker hired and the layoff charges are
₹4,000 per worker laid off, evaluate the impact of the two schedules that you have proposed for manpower loading.
10. A new construction work has 10 work packages designated as A to J. Table 4.22 has pertinent details for the project.
a. Make a network representation of the project and identify alternative paths in the network
b. What is the project duration?
c. The project management team realized a major omission in one work package. The omitted work package (designated
as K) has ‘C’ as the predecessor and ‘H’ as the successor. The duration of this work package is 4 months. Will there be
a change in the project duration after inclusion of this work package into the project?
d. Identify the work packages with non-zero slack.
11. Consider the Activity on Arc (AOA) representation of a project given here:
12. Table 4.23 here has the details pertaining to a new product launch programme of a company. Since the product line is new,
they have no prior experience.
TABLE 4.23 Data for Problem 12
a. Represent the data in a network and identify the critical path for the project ignoring the uncertainty in the project.
(Hint: Use most likely estimates for activity duration).
b. Based on the data available compute the expected project duration for the product launch.
c. What is the additional project duration attributed to the uncertainty of project execution?
d. Compute the expected project duration for a probability of completion of the following:
i. 50%
ii. 75%
iii. 90%
iv. 95%
13. An engineering consulting firm has recently bagged an order to provide turnkey consulting services for preparing a detailed
project report (DPR) for off-shore drilling of wells for ONGC. The firm has no direct experience in this kind of project
although they have executed several similar projects on shore. The firm is in the planning phase and is expected to quote a
time frame for preparation of the DPR. It is keen to answer several questions. The DPR preparation project consists of 10
activities and the time estimates of these activities are given in Table 4.24:
The firm requires answers to the following questions to help it plan the execution of the project better:
a. Collect relevant information and submit a report outlining the key features offered by the three vendors, their relative
strengths and weaknesses.
b. Explain how the key concepts covered in this chapter have been implemented in the software.
c. Are there any noticeable differences between the theory and practice that you have observed? If so, what are the
possible reasons for this deviation?
2. Read the book The Critical Chain by Eli Goldratt (The complete bibliographical details for the book are available in the
Suggested Readings section of the chapter) and prepare a 10-page report on the book. While preparing the report answer the
following questions:
a. In the opinion of Eli Goldratt, what are the key problems that a project organization faces?
b. What is the relevance of the theory of constraints in project management?
c. Do you see any advantages in applying the theory-of-constraints framework to project management? Are there any
potential problems?
d. In the context of project management, what new things have you learnt from this book?
CASE STUDY
After the customer order is received by marketing, it is reviewed by them and then circulated to
all the concerned departments. The engineering department then prepares the electrical designs,
which sent for customer approval through the marketing department. After the customer
approves the drawings, they are sent back to the engineering department. However, if the
customer makes any comments, then it will take a minimum of three days for processing in
between the marketing and engineering departments to address the comments. If the customer
asks for technical discussions, the engineering department shall attend to it.
After the drawings are sent to the marketing department, for customer approval, the
engineering department then prepares manufacturing drawings and informs the materials
department regarding
a. the material requirement for making prototypes,
b. details of materials with long lead times, and
c. tools and dies required for manufacturing.
On receiving the approved drawings from the customer, the engineering department then issues
manufacturing drawings to the production planning and control (PPC) department for the
distribution of the same.
The planning department then issues the drawings to the concerned departments. For all
revisions to the drawings and for the distribution of BOMs (Bill of materials) the process is the
same. For every revision, it takes two days time to process it through the PPC department. The
materials department then takes action for the procurement of the materials.
After the jobs have passed the internal testing, information is given to the marketing
department to organize for customer inspection. Sometimes, advance notice is given to the
marketing department regarding the probable dates of completion of jobs so that customer
inspection could be planned immediately after production. Sometimes, the customer may have
to wait for the final inspection due to communication delays between the departments. After the
inspection is over, dispatch clearance is taken, and then the manufactured goods are dispatched.
Based on the steps explained above, the data was collected for a sample of four orders
executed last year and interpreted as follows. Of these, two were domestic orders and the other
two were export orders. In each of these categories, one representative order and another worst-
case order have been considered. Of the two export orders, one was the worst case in which two-
thirds of the order was cancelled because of late deliveries whereas the other is a smooth-going
case. The Relevant details of these work orders as well as their execution times are given in
Table 4.26.
TABLE 4.26 Order Execution Details: A Sample Analysis*
Note: *All the numbers in the table indicate the number of days for the activity.
SUGGESTED READINGS
F. Acebes, J. Pajares, J. M. Galán, and A. López-Paredes, (2014), ‘A new approach for project control under
uncertainty. Going back to the basics’, International Journal of Project Management, Volume 32 (3), April 2014, pp.
423–434.
R. Burke, Project Management Planning and Control, 2nd edition (New York: John Wiley & Sons, 1993).
D. I. Cleland, Project Management: Strategic Design and Implementation, 2nd edition (New York: McGraw-Hill,
1995).
E. M. Goldratt,The Critical Chain (Great Barrington, MA: The North River Press, 1997).
F. L. Harrison, Advanced Project Management, 3rd edition (Aldershot: Gower Publishing, 1992).
J. R. Meredith and S. J. Mantel, Project Management: A Managerial Approach, 4th edition (New York: John Wiley &
Sons, 2000).
B. Pollack-Johnson and M. J. Liberatore, “Project Management Software Usage Patterns and Suggested Research
Directions for Future Developments,” Project Management Journal 29, no. 2 (1998): 19–28.
W. A. Randolph and B. Z. Posner, “What Every Manager Needs to Know About Project Management,” Sloan
Management Review 29, no. 4 (1988): 65–73.
P. S. Royer, “Risk Management: The Undiscovered Dimensions of Project Management,” Project Management Journal
31, no. 1 (2000): 6–13.
M. P. Spinner, Project Management: Principles and Practices (New Jersey: Prentice-Hall Inc, 1997).
J. D. Weist and F. K. Levy, A Management Guide to PERT/ CPM, 2nd edition (Englewood Cliffs, NJ: Prentice-Hall,
1988).
Part II
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
What do you understand by the term “supply chain management"? What are the elements of a supply chain?
Will the structure of a supply chain have any impact on its overall performance?
What do you mean by the bullwhip effect? What causes the bullwhip effect in supply chains?
What are the metrics for assessing supply chain performance?
How can an organization design supply chains for innovative and functional products?
What is the role of third-party logistics in a supply chain?
A supply chain includes the chain of entities involved in the planning, procurement, and production and distribution of
products and services. The dabbawala network in Mumbai is an example of a supply chain in the service sector.
Source: Image taken with approval from www.mydabbawala.com, the official Web site of Dabbawala
Supply chain management (SCM) has gained considerable importance in recent years in the
manufacturing and services sector of our economy. Good supply chain management helps
organizations secure high quality input materials and purchase goods at competitive prices. It
also allows organizations to respond to customer orders faster. Often, with good management,
organizations have benefited from reduced costs throughout the supply chain.
Every organization is engaged in providing some goods and services to its chosen set of
customers. In a typical manufacturing firm, they begin this process with sourcing raw materials,
converting them into intermediate components, and later into finished goods. The goods are
finally delivered to the customer. Similar activities occur in the case of many service industries
also. Typical examples include hotels, healthcare, and automotive repair and maintenance. In the
case of hotels, raw materials and specialized services procured are converted into useful service
offerings and finally distributed to customers. Therefore, in every organization, materials flow
through a continuous chain, beginning from raw material suppliers, through intermediate
manufacturers to the final assembler, and to the distributors and retailers before it reaches the end
customer.
Supply chains are equally relevant to service organizations. Let us consider an example to
understand this. Consider the personal finance sector and a product such as fixed deposits
marketed by a firm such as Sundaram Finance Limited (SFL). Offering fixed deposits and
servicing various customer requirements involves several players, both within and outside SFL.
Typically, there are four major external players in SFL’s supply chain. Brokers play a key role in
providing reach and aiding SFL in business development by taking care of the distribution aspect
of the overall chain. Moreover, since they come into direct contact with customers, their
performance plays a significant role in overall customer satisfaction.
Banks play the role of handling financial transactions in the system. They also influence the
supply chain costs by providing additional features for the company in transferring funds from
one location to another. The registrar’s role in this business is crucial. The Reserve Bank of India
and other government regulations require that systems for processing fixed-deposit application
forms, printing fixed-deposit receipts and interest warrants, and maintaining relevant information
are adequately robust and foolproof. Finally, the Indian postal service and private courier
companies also form a part of the supply chain. They provide logistics support to the entire
operation.
A supply chain includes the chain of entities involved in the planning, procurement, and production and
distribution of products and services.
Another service example is the dabbawala network in the metropolitan city of Mumbai. The
forward supply chain begins at the households, where home-prepared food packed in “tiffin
boxes” is collected from housewives and ends at the point of delivery at the respective
customer’s office. Similarly, collecting the empty boxes from the customers’ office and
delivering them back at the respective household constitutes the reverse supply chain. In both
these cases, all issues of supply chain management need to be addressed. These include
designing the distribution network and appropriate capacities for the supply chain, demand
management and forecasting, development of appropriate metrics for supply chain performance,
and cost and quality control. Material flow, information flow and funds flow are characteristic
features of such supply chains, as is the case in any manufacturing supply chain.
Figure 5.1 presents a schematic representation of this idea. As shown in the figure,
procurement of input material and services, production of goods and services, and distribution to
the end customers are the three major processes in any supply chain. A supply chain, as the
name suggests, includes a chain of these entities involved in delivering products and services to
the end customer. Consider Liberty, the well-known shoe manufacturer. The dummy foot for the
shoes is created in Italy, Korea, Spain, or Germany. The mould for soles also comes from Italy
and Germany. The upper portion is designed using the company’s own CAD/CAM facility and
some specialized designs are bought from design studios in Europe. The final manufacturing is
done at their factory and marketing is done by the company’s exclusive dealers as well as
through other retail outlets.
Central to our definition of supply chains is the fact that every organization is just one entity in
a continuous chain of the value-creation process. Such a perspective helps every organization to
understand the complex interactions across organizationally and physically separated operating
units and the need to collaborate with them in order to create value for the ultimate customer.
Supply chain management principles require every entity that makes up the chain to work in
collaboration with the other entities rather than in isolation. In the absence of collaboration, the
value created in one part of the value chain is unwittingly destroyed in some other part of the
value chain. Therefore, planning is a major process in supply chain management.
Recent competitive pressures have made several Indian organizations realize that individual
companies by themselves cannot face up to competition. A well-designed product and sound
manufacturing technology may not ultimately deliver value to the customer. These strengths may
sometimes be offset by a poor set of suppliers or an inefficient distribution network. The key
requirement of supply chain management is the ability to network with several other business
entities having complimentary skill sets. This unique combination of several business entities
provides a value stream to the ultimate customer.
FIGURE 5.1 Schematic representation of a supply chain
A unique combination of several business entities engaged in various stages of the supply chain provides a value
stream to the ultimate customer.
For instance, consider Chennai-based Lucas TVS and Bangalore-based Bosch Limited, which
are manufacturers of auto-electrical components. It is well known that Lucas TVS and Bosch
market competitive offerings in a certain range of electrical components. Due to this, it is
customary to think that Lucas TVS competes with Bosch. However, principles of supply chain
management have transformed this impression. Lucas TVS competes with Bosch not merely on
the basis of what it does in its factory but also on the basis of its relationship with suppliers and
customers and the strength of its distribution system. In this case, the supplier, the customer, the
logistics partner, and Lucas TVS together constitute the supply chain and provide a unique value
stream. Therefore, it is reasonable to assume that the value stream provided by Lucas TVS
competes with the value stream provided by Bosch.
The Need for Efficient Supply Chain Management
Consider the following statistics: The number of new-product introductions in the U.S. food
industry increased from 2000 in 1980 to 18,000 by 1991. The number of watch models produced
by Titan Limited was 850 in 1993. It increased severalfold by 1999. In 2002–2003 alone, Titan
brought out 180 new models. Asian Paints maintains over 2,800 stock-keeping units. These
range from 50-ml packs of paints to over 5-kg cans. If the number of colours and shades are
added to this, the variety is mind-boggling. Department markdowns1 as a percentage of sales
went up in the United States from 8 in 1971 to 33 in 1995. The annual wastage arising out of
poor supply chain coordination in the United States was estimated to be USD 30 billion in 1999.
These statistics bring out the nature of issues that organizations the world over are facing today.
What happens when the number of new-product introductions increase? Amongst other things,
organizations face shrinking product life cycles. The Ambassador car has been sold for several
decades with very few variations and model ranges. However, with the introduction of more
products and models into the market, the life cycle of passenger car models today is not more
than five years. Most shoe manufacturers such as Liberty end up designing a product from
scratch and launch it every 18 months. They also continue to modify the product once in three
months. In certain sectors such as computers and electronic goods, the life cycle of the product is
shortening rapidly. Even in the services sector, we find that the life cycle of specific service
offerings are shrinking. For example, in the telecommunication sector, old tariff plans are
withdrawn and new schemes introduced often. Look at the new menus offered in the fast-food
industry or the new health and fitness programmes introduced by the fitness industry. These
share similar issues arising out of shrinking life cycles.
The second effect of faster new-product introductions is the dramatic increase in product
variety. When product variety increases, demand volatility also increases. Organizations find it
increasingly difficult to forecast demand and plan effectively for procurement, production and
distribution. Consider the case of consumer appliances such as colour TVs. Samsung and LG
offer several models of colour television sets to their customers. While it may be relatively easy
to forecast the demand for their brand of colour televisions at an aggregate level, getting the
exact numbers for each variation is difficult. Since estimating model-wise demand becomes more
and more difficult, organizations need to plan the production as near as possible to the point of
demand, in order to reduce the risk of non-moving inventory. This calls for shorter lead times for
procurement, manufacture, and delivery.
The third effect is the change in market dynamics. In an era of wide variety and faster new-
product introduction, organizations also face competitive pressures while trying to retain their
share of the market. Shrinking delivery windows at the marketplace demand greater
responsiveness on the part of organizations. Reducing the lead time and cost and recasting the
overall supply chain structure are some of the immediate steps required to face these competitive
challenges.
In the light of these challenges, today’s organizations face several questions pertaining to
managing the supply chain: Will the structure of the supply chain have any impact on the overall
performance? What are the strategies for improving inbound logistics? What kind of planning
tools are useful for supply chain management? Are there workable strategies for managing
scenarios such as product variety and short product life cycle? What are the appropriate measures
for supply chain performance? Good supply chain management practices seek to address several
of these questions and provide alternative viewpoints to develop a basis for addressing these
questions.
Supply chains are made up of three distinctive entities: the inbound supply chain, the in-house
supply chain, and the outbound supply chain. The inbound supply chain includes a host of raw
material and component suppliers. These suppliers respond to the production plan of the
manufacturer by supplying raw materials and components. The in-house supply chain engages in
the conversion process through which the raw material and components supplied by the vendors
are scheduled for production and final assembly. Finally, the outbound supply chain comprises
the distribution network. This may involve third-party logistics providers, distributors and
retailers.
The supply chain is made of three distinctive entities: the inbound supply chain, the in-house supply chain, and
the outbound supply chain.
The inbound supply chain pertains mainly to providing raw materials and components to an organization.
One method to organize the inbound supply chain is to create tiers of suppliers. Figure 5.3
provides an illustration of the tiers in an inbound supply chain in the case of the automotive
industry. Tier 1 consists of OEM suppliers who supply key subsystems. For instance, there will
be a few suppliers providing wiring harnesses, upholstery, transmission, and auto-electrical
subsystems. At Tier 2, there will be a set of component suppliers related to each subsystem. In
the case of transmissions, there will be suppliers of castings, forgings, spindles, and shafts. Tier-
2 suppliers coordinate with Tier-1 suppliers and respond to their production plans. Tier-2
suppliers often source their requirements from metal-manufacturing and ore-processing firms. In
the case of transmissions, the Tier-2 manufacturers may require alloy steel in the form of ingots,
rolled coils, and rods. They obtain these from iron and steel manufacturers. Similarly, in the case
of auto-electrical subsystems, copper manufacturers provide the basic material for Tier-2
suppliers.
FIGURE 5.3 Inbound supply chain: an example from the automotive industry
One method to organize the inbound supply chain is to create tiers of suppliers.
The in-house component of the supply chain relates to the physical configuration of the conversion process.
The core manufacturing layer consists of various physical resources for the conversion
process. The first operation is done in pre-manufacturing where the raw material is cut to the
required shapes and some pre-processing is done. In several cases, the conversion process
involves some fabrication and machining activity before the various components are assembled
and tested. An appropriate organization of these resources is required for the conversion process,
and maintenance of these resources is part of the in-house supply chain management.
Since the conversion process is a critical link in the entire supply chain, it will assume the
central role in the overall supply chain collaboration. For instance, the manufacturer of a
consumer durable will have to initially estimate the demand for the product and communicate
this information to the rest of the supply chain to devise a strategy for meeting the demand. This
may include directing the inbound supply chain to schedule delivery of certain components and
raw materials, ensuring the manufacture of the product according to plan, and scheduling
dispatch of the product through logistics infrastructure to the appropriate parts of the outbound
supply chain. The manufacturing support layer in an organization performs these activities.
FIGURE 5.4 A typical configuration of an in-house supply chain
Apart from the physical location of the resources for the conversion process, planning is a
crucial aspect of the in-house supply chain. The first major activity is demand management. It
includes forecasting and demand estimation, order entry and customer order fulfilment. In order
to meet delivery promises, a host of production planning and control activities will be required as
part of the in-house supply chain. Since the in-house supply chain plays the major role of
collaboration, both within and outside the organization, several functional areas are involved in
the process. The cooperation of several functional areas, including planning, materials and
procurement, marketing, production, and logistics will be a critical aspect of the working of an
in-house supply chain.
The outbound supply chain pertains to the distribution of goods and services to the end customers.
Figure 5.5 is a typical outbound supply chain for fast-moving consumer goods, such as soaps
and detergents, which consist of four levels. The first level is the factory warehouse where the
finished goods are stocked. At the next level, there will be regional distribution centres (DCs)
that cater to a wide geographical area. Each DC will, in turn, serve a number of sales depots or
regional stock points. For instance, the South Zone DC will cater to the requirements of Tamil
Nadu, Kerala, Karnataka, Pondicherry and Andhra Pradesh. At the fourth level, the sales depots
will serve the retail outlets. The sales depot located in Bangalore may, for example, serve over
500 retail outlets in and around the city. The retail outlets finally sell the product to the
customers.
The dealers served the end customer by dealing with them directly.
Distribution network design refers to the choices made with respect to the nature of entities
included in outbound logistics and the manner in which material flow and information flow are
managed across these entities. The network design could either result in centralized distribution
logistics or decentralized logistics design. For instance, a few years ago, a manufacturer of air-
conditioners was considering two alternative distribution network designs as shown in Figure
5.6.
In the first alternative, the two factories were directly sending the goods to the 16 depots. They
were held in the depots as inventory until invoiced to dealers. In the second alternative, the
sixteen sales depots would no longer be stock points. Instead, inventory would be maintained
only in the four zonal DCs. The sales depots would merely act as receiving, dispatching, and
invoicing points.
Network design is very crucial, as it will have a significant impact on supply chain
performance. The amount of inventory stored in the entire supply chain will vastly differ
between the two alternatives. If there are multiple locations in which inventory is stocked, then
there is a need to carry safety stock at each location in order to meet uncertain demand. On the
other hand, if the inventory is stocked at a centralized point, the safety stock will be much less.2
This will result in lower cost of operation. However, a centralized design will require an efficient
logistics setup to ship inventories to the point of demand. Therefore, the responsiveness of the
supply chain will depend on the logistics.
Distribution network design refers to various choices made with respect to the nature of entities included in
outbound logistics.
There is a noticeable trend in recent years to rationalize the supply chain structure. This is
partly attributed to advancements in IT infrastructure and availability of better transport
infrastructure. For instance, the manufacturing and distribution of HP DeskJet printer was done
using a lean structure with just two manufacturing facilities, one in Vancouver in Washington the
United States, and the other in Singapore.3 In the continental United States, vast markets were
served from fewer DCs as opposed to the earlier practice of having multiple stock points. Such
developments are yet to take place in India. As the logistics infrastructure—including the road
and rail network, availability of ports, and connectivity to the ports—and the efficiency of the
logistics process improve, there will be a greater shift towards fewer layers in the outbound
supply chain and the increased use of centralized DCs. The Golden Quadrilateral Project,
proposed and implemented by the Union Government during 1999–2001, is especially important
in addressing the supply chain issues of domestic firms.
FIGURE 5.6 Distribution network design: two alternatives
ideas at Work 5.2
The project resulted in several benefits to IOCL. These include better crude selection and
allocation which took into account product demands, refinery capabilities and effect of the
crude already procured; optimal refinery production planning; and optimal distribution
planning considering transportation costs, taxes, duties, and transportation constraints. On
account of these, IOCL was able to improve its margins and profitability,
IOCL experienced an improvement in their capabilities in certain areas after the
implementation of the Honeywell SCM solution. For instance there was an improved
visibility into its supply chain process across the five selected refineries. IOCL was able to
perform investment analysis for refinery units, pipelines, etc. and formulate strategies to meet
future scenarios like change in specifications. Faster and more effective decision making on
exchange strategies, imports and exports, and improved response and execution capability
were the other welcome changes after the implementation.
Source: http://hpsweb.honeywell.com/Cultures/en-US/NewsEvents/SuccessStories/success_IOCL.htm. Last accessed on June 15, 2009.
Logistics management
Logistics management refers to the set of activities involved in the physical movement of goods
across the supply chain. It may require planning shipments from one part of the supply chain to
the next, scheduling carriers, and route planning. The nature of the logistics required is a function
of the distribution network design. As demonstrated, a centralized distribution system will
require an efficient logistics system to meet the demands of customers spread over a wide
geographical area, as opposed to a decentralized design.
Channel management
Channel management is another key aspect of the outbound supply chain. The distribution of
goods and services often involves multiple organizational entities outside the manufacturing
firm. For instance, in the case of a fast-moving consumer goods (FMCG) manufacturer which
produces soaps and detergents, the channel will include dealers, stockists, and retailers. In such
situations, channel management involves the smooth passage of information and material flows
across the supply chain. The demand estimates and the point-of-sales data form crucial input to
production planning and scheduling. Hence, channel management requires systems for efficient
data capture and dissemination across the supply chain.
Channel management involves the smooth passage of information and material flows across the supply chain.
It requires systems for efficient data capture and dissemination across the supply chain.
Channel management also involves relationship management. Since the lowest level in the
channel has direct contact with the customer, they play a major role in customer relationship
management. Feedback from customers about the product, problems faced by the customers in
using the product, and specific preferences and requirements ignored in the current offering are
very important for the entire supply chain. Only when this information flows back quickly and
without loss can necessary corrections be done in the supply chain. Conflicts in the channel and
motivational issues of the channel partners also have a significant impact on the overall
performance of the supply chain.
In order to understand the various activities pursued in a supply chain, it is useful to take a
process view of the entire supply chain. A process approach helps clarify the various roles,
categorizes the processes into homogenous units, sets down the set of activities required to
complete the process, develops appropriate measurements, and uses them for improvement in the
overall performance of the supply chain over time. Figure 5.7 provides a process orientation to
the supply chain and its relationship to the components of the supply chain.
There are four generic processes involved in any supply chain: planning, sourcing, manufacturing, and
distribution.
There are four generic processes involved in any supply chain. These include planning for
operations, sourcing decisions, manufacturing-related activities, and distribution. Planning is the
main activity in the in-house supply chain. Planning establishes the premises for other
components of the supply chain. Detailed discussion on several aspects of planning is available
in the later part of the book (see Chapters 14–19 for various aspects of planning). Here, we shall
take a brief look at planning in the in-house supply chain.
FIGURE 5.7 Process view of the supply chain and its relationship to supply chain components
The inbound supply chain mainly comprises sourcing issues. Sourcing decisions call for
supplier development, supply management, purchasing policies, and a host of activities related to
cost reduction and new-product development. These have a significant impact on the overall
supply chain performance. Chapter 7 deals with these issues.
A considerable amount of planning is involved in in-house supply chain management. Since
supply chain entities cut across organizational boundaries, planning plays an important role in
supply chain collaboration. For example, on the basis of planning, customers are promised
delivery schedules. Schedules for distribution and third-party logistics are also arrived at in a
similar way. On the other hand, delivery schedules for component and raw material suppliers are
also drawn on the basis of planning.
Demand management is the most crucial amongst all the elements of a supply chain. Based on
the information available on various components of the supply chain, planning for future
operations is done in order to fulfil a certain demand. Demand management triggers a host of
other planning processes including planning for materials, resources, capacity, and distribution.
The supply chain structure addresses issues pertaining to the number of layers that make up a supply chain, the
composition of each layer, the type of information flow across the layers, and the nature of integration achieved
between the layers.
The various aspects of supply chain structure that we just discussed culminate in the bullwhip
effect in supply chains. In our example, each layer will employ its own practices in updating the
demand forecasts. For instance, when the retailer may be revising its forecasts on the second
Wednesday of every month, the distributor at the next level may be doing it on the first Saturday
of every month. This introduces avoidable confusions into the system and may result in
misleading signals in the system. While the retailer may be estimating a rising demand, the
distributor may estimate falling demand due to alternative time periods of estimation.
The bullwhip effect denotes the increasing severity of distortions in demand information and ordering patterns
as the information travels from one layer of the supply chain to the next layer.
The problem may get more complicated if ordering patterns differ across the layers. Since at
each layer the review of demand is done at different times, one would expect orders also to arrive
at different times. Continuing our example, while distributors estimate lesser demand and place
orders for fewer quantities, they may in fact receive larger orders from retailers. This will cause
further problems in the supply chain and each supply chain layer will begin to respond to these
confusing signals by revising the order either upwards or downwards. Moreover, the severity of
these distortions in demand information and the subsequent ordering pattern increases as the
information travels from one layer of the supply chain to the next layer.
Very soon, it will lead to a situation where the ordering pattern and magnitude will not be
representative of the reality. This phenomenon eventually leads to a bullwhip effect in the
supply chain.4 The bullwhip effect often creates unnecessary strain in the entire supply chain and
is responsible for sudden surges in inventory or severe shortages. Recent studies have shown the
existence of the bullwhip effect in several supply chains and the possible causes for the same.
Minimizing the bullwhip effect calls for several decisions cutting across the supply chain and
greater collaboration among supply chain partners. Reducing the number of layers could
minimize the bullwhip effect. Very recently, several global firms have resorted to this approach.
They try to directly reach a vast number of customers from just one or two strategically located
distribution centres with the help of third-party logistics, provided by well-known players such as
FedEx.
The other strategy is to reduce the delay of information flow in the supply chain. The use of
point-of-sales data-capturing systems, Electronic data interchange (EDI), and the Web helps
organizations to cut delay in information flow. Organizations need to encourage supply chain
partners to share sales, capacity, and inventory data with other supply chain members. This task
is easier said than done as it requires behavioural changes among the supply chain members.
Suitably modifying the incentive structure could be one method to achieve this objective. During
1998, when Bangalore-based colour-TV manufacturer BPL electronically linked the top
distributors and retail outlets with the factory planning system, the finished goods inventory level
fell dramatically almost immediately.
Supply chain members can reduce the bullwhip effect further by reducing the lead time of
their business processes. From Figure 5.8, one can see that from the time the retailer decides to
place an order with the distributor, it takes eight days (two days to transmit the order to the
distributor, one day to prepare shipment, three days to transport goods, and another two days to
account the receipt and shelve it). This lead time could be reduced using better business
processes, information technology, and closer working arrangements with logistics providers and
distributors. The bullwhip effect can also be minimized through a reduction in fixed costs of
ordering.
Supply chain performance measures seek to provide a quantitative basis for understanding the
performance of the supply chain and indicate potential areas for taking corrective measures.
Since improvement projects have a certain time lag, current period results may indicate the effect
of previous time periods’ efforts in managing the supply chain efficiently. Hence, measures
could also help the manager understand the nature of improvement efforts currently undertaken
during the measurement period. Supply chain performance measures could be both post-process
and process indices.
Supply chain performance measures could be both postprocess and process indices.
Post-process Indices
Post-process indices are those that rely on past data to assess the performance of the supply chain
function during the relevant period of time. These measures typically use information available
in annual reports of companies to compute indices. Since inventory is the most representative
item in the annual report, post-process indices for supply chain performance compute inventory
measures. The following measures for supply chain performance could be computed from annual
reports:
Traditionally, the inventory turnover ratio and the number of inventory turns are employed to
assess the performance of the supply chain. While these measures are quite useful for accounting
and control functions, they do not offer substantial help to the operations function. On the other
hand, the number of days of inventory is a useful measure for operations. It enables the operating
personnel to relate the numbers to ground realities and helps them make improvements.
For example, consider a company having sales of ₹300 million and a total inventory
investment of ₹50 million. The inventory turnover ratio is 16.67 per cent and the number of turns
of inventory is six. Alternatively, one can express inventory in number of days, which is two
months in this case. An investment of two months of inventory clearly indicates that the lead
time of various activities in the company pertaining to procurement, manufacturing, and
distribution is in the range of two months. Reducing inventory in this case requires that the lead
time is cut proportionally, and clarifies to the operating personnel the nature of the activities to
be undertaken to reduce the lead time and the relationship of these to the overall investment in
inventory. Similarly, to the operational personnel, the DSO, DPO, and CCD indicate the credit
terms enjoyed by the company and offered to the customers, and their impact on the overall
working capital needs.
One can obtain a sharper focus by computing the number of days of inventory for each class
of inventory, that is, raw material (RM), work in progress (WIP), and finished goods (FG) and
relating them to specific activities in the supply chain. The formulae for computing each category
of inventory is as follows:
EXAMPLE 5.1
Extracts from the annual reports for the years ended March 2010 and March 2011 of an auto-
component manufacturer in Tamil Nadu are given in Table 5.1. Compute the relevant post-
process indices of the supply chain performance. Are there any significant inferences that one
can make based on the computation?
TABLE 5.1 Extracts from Annual Reports (Values in ₹Million, Unless Otherwise Specified)
Solution
Calculation of number of days of inventory for the year ended March 2011:
Table 5.2 presents the post-process indices for supply chain performance.
TABLE 5.2 Post-process Indices for Supply Chain Performance
2011 2010
Inventory turnover ratio (TO) (%) 8.15 6.67
Number of inventory turns (TN) 12.28 15.00
Total inventory (TID) 29.73 24.33
Sales outstanding (DSO) 60.37 48.88
Payables outstanding (DPO) 47.00 50.42
Cash-to-cash cycle time (CCD) 43.10 22.79
Sub-classes of inventory
Raw material inventory (RMD) 34.08 21.04
WIP inventory (WIPD) 3.24 2.94
FG inventory (FGD) 6.64 8.12
Some Inferences
During the year, there is an increase in TID, DSO and CCD. Clearly, the company is under
the pressure of a high inventory and has apparently given generous credit terms to its
customers. Therefore, although the finished goods inventory has come down, there is a
significant increase in DSO, and correspondingly in CCD. Due to this increase, the working
capital requirement has gone up.
The number of days of inventory in the case of raw material has significantly gone up
during the year. Interestingly, during the same time, the credit term extended to the suppliers
(indicated by DPO) has come down. This could be attributed to any of the several causes,
including an increase in procurement lead time, mismatch between planning and production,
poor performance by suppliers leading to increase in safety stock, and hedging inventory to
protect against impending price increases. It could also point to the need for studying the
inbound supply chain for possible performance deterioration during the year.
*An alternative base for computation of this measure could be cost of goods sold.
Process Indices
Improvements in supply chain performance occur only when the processes related to the supply
chain are studied and corrective measures taken. Therefore, a set of measures is needed to
understand the nature of improvement activities pursued in the supply chain. Process indices
serve this purpose for an organization. By virtue of the improvements in the process, the supply
chain may improve its responsiveness, cost, quality, or reliability.
For instance, when an organization develops superior suppliers with a high degree of delivery
reliability and quality performance, the investment in raw material inventory may come down on
account of reduced safety stock and lead time for procurement. Similarly, the component may
also cost less in the long run. Some of the process indices used for assessing the performance of
the supply chain are discussed here.
EXAMPLE 5.2
Eastern Castings is a crucial supplier to a machine tool manufacturing company. Since the
delivery reliability of the castings has a direct bearing on the scheduling of the production
system and the ability of the machine tool manufacturer to commit to the customers, they
have instituted OTD as a measure of performance. Table 5.3 shows the performance of
Eastern in the last 10 deliveries. Compute OTD index for Eastern Castings.
Solution
The number of points for each delivery is tabulated in Table 5.4.
Designing an appropriate supply chain calls for a good understanding of the product profile for
which the supply chain is configured. Let us compare two product categories: one an FMCG
item, such as toothpaste, and the other a trendy sportswear, such as World Cup cricket T-shirts
conforming to specifications of the official design. Toothpaste is a functional product, or a
product that is primarily intended to serve a function and is independent of the time of marketing
the product. These products are stable, satisfy a basic need, and therefore will have a predictable
demand over a longer time. Because of these features, there is more competition in such
products, leaving very little profit margins for companies operating in this product.
The World Cup T-shirt, on the other hand, is an innovative product. Other examples include
Tanishq jewellery by Titan and other fashion goods such as shoes. Liberty claims to introduce
1,000 new styles every year and withdraw several old styles. An innovative product is novel,
very sensitive to the time frame within which it is being marketed, and may often have a short
life cycle as opposed to a functional product. Because of its innovative nature, competitors may
try to imitate it. Moreover, the innovation may also involve creating a large variety of offerings.
All these may ensure greater profit margin, but may also introduce additional problems for an
organization. As mentioned at the beginning of this chapter, increased variety results in demand
volatility and greater need to predict demand accurately. Otherwise, the cost of operating the
system may be very high. Other examples of innovative products include fashion goods,
computers and high-tech gadgets.
The key differences between a functional product and an innovative product are as follows:6
Innovative products command a higher contribution margin (20%–60%) than functional products (5%–20%) and have
much more product variety.
Due to variety, forecast errors in the case of innovative products will be higher (40%–100%) than functional products
(10%), leading to product markdown at the end of the product lifecycle. Usually, there are no forced end-of-season
markdowns for functional products. Innovative products, on the other hand, may have markdowns between 10 per cent
and 25 per cent.
Further, short product life cycles of 3 months to 1 year in the case of innovative products also imply that these products
require shorter lead times (1 day to 2 weeks) in the supply chain.
Because of the shorter lead times in the case of innovative products, responsiveness is the most important requirement.
A responsive supply chain will be able to curb the demand–supply mismatch and significantly reduce the end-of-period
markdowns. In the case of functional products, efficiency issues are important while configuring the supply chain due to
their narrow profit margins.
Since demand is likely to exist for a longer time, the continuous replenishment of products in a
cost-effective manner will require efficient information sharing across the supply chain.
Developing robust inventory control mechanisms to accurately fix the reorder point and order
levels are important to ensure efficient supply chains.7 Integrating material planning and control
systems with other related systems in the in-house supply chain using corporate-wide
information systems such as enterprise resource planning (ERP) is yet another requirement for
efficient supply chains.
The design of a responsive supply chain begins with the basic premise that uncertainty in demand and large
forecast errors are often the reality.
Assembly postponement
Low levels of investment in FG
Ability to handle a large variety through modular
design (Example: Dell computers)
Manufacturing postponement
Final stages of manufacturing delayed until firm
orders are received (Example: Benetton's dyeing of
fabrics)
Reduced lead time
Other strategies employed in configuring an efficient supply chain include changes in product
design. Deploying standardization, use of modular design and product platforms are some
examples. The computer and automotive industries provide several examples of these. Table 5.5
lists postponement strategies and their implications to organizations.
Firms selling innovative products practice three types of postponement: packaging postponement, assembly
postponement, and manufacturing postponement.
Firms selling through the Web reach out to the customer directly. They engage in a variety of
direct selling techniques. For instance, they may display a catalogue and request the customers to
select a product or service of their choice and pay for it using a digital payment system. For
example, Indian Railways allows prospective travellers to book tickets through their Web site
www.irctc.co.in. Similarly, LG Electronics encourages its customers to order directly through
their Web site www.lgezbuy.com. Web-based firms may alternatively put up an auction and
encourage customers to bid for their offerings online. A very good example of this is the well-
known auction site www.ebay.com. In all these examples, once the customer makes his/her
selection of product through a Web site and place an order, the firm reaches the product to the
customer at the delivery address provided during order booking using a third-party logistics
provider.
The advent of Web-based e-commerce has projected a crucial role for third-party logistics providers.
In all these cases, the customer exercises his/her choice on the Web site and makes a payment.
Order fulfilment is a crucial requirement in Web-based e-commerce. During the Christmas
festival season in 1999–2000, Web-based firms took online payments from several customers
and failed to make delivery of gifts before Christmas. In lieu of this, they offered to pay back the
money and also gave gift cheques for USD 20–USD 50. Widespread protests by unhappy
customers eventually led to the downfall of several Web-based firms. Therefore, the advent of
Web-based e-commerce has projected a crucial role for third-party logistics providers.
SUMMARY
Planning, procurement of input material and services, production of goods and services, and distribution to the end
customers are the major activities in any supply chain.
The components of a supply chain include inbound, in-house and outbound. While the inbound supply chain addresses
issues relate to raw materials and component suppliers, the outbound supply chain pertains to the distribution network.
Supply chain structure refers to the manner in which various entities pertaining to a supply chain are configured. It
includes the number of layers that make up a supply chain, the composition of each layer and the nature of integration
among the layers.
The bullwhip effect occurs in a supply chain due to several reasons. These include:
The number of layers in the supply chain
Delays in information flow across these layers
Variations in decision making patterns at each layer
The independence of each member in the supply chain
Several measures are available to assess the performance of supply chains. These can broadly be classified into process
indices and post-process indices.
The design of a supply chain must take into consideration the nature of products. Innovative products require
responsive supply chains and functional products require efficient supply chains.
Due to an increase in the use of the Web and IT infrastructure, third-party logistics providers play an important role in
the outbound supply chain of many firms.
REVIEW QUESTIONS
1. Identify the key supply chain elements in the following cases:
2. Consider Asian Paints and Berger Paints, the two close competitors in the paint industry. Which are the elements of supply
chain management that would create differences among their competitive offerings in the market?
3. How should an organization manage an inbound supply chain that is organized into tiers? What are the benefits of organizing
suppliers into these tiers?
4. What do you mean by supply chain structure? What is the relationship between the supply chain structure and supply chain
performance?
5. Which sectors of industry will have a pronounced bull-whip effect? Why?
6. Why is the bullwhip effect undesirable for organizations? What are the various alternatives that organizations have to
minimize the bullwhip effect?
7. What are the key aspects of managing a global supply chain? Can you give some examples of global supply chain
management?
8. Identify an appropriate type of supply chain for each of the following product categories:
a. Jewellery watches
b. Toothpaste and brushes
c. Earth-moving equipment
d. Designer clothes
e. Mobile telephony services
f. Executive education programmes
g. Bicycles
h. Passenger cars
i. Heavy commercial vehicles
FORMULA REVIEW
NET-WISE EXERCISES
1. Visit the URL: http://www.hul.co.in/sustainable-living-2014/casestudies/Casecategory/Project-shakti.aspx and browse through
the page. Answer the following questions:
a. What are the supply chain management issues pertaining to accessing rural markets? What are the opportunities that
organizations can tap while addressing these issues?
b. How does i-Shakti address the rural markets? List the strengths and limitations of their offerings.
c. Google for the e-Choupal initiative of ITC and draw a comparative study of i-Shakti and e-Choupal.
2. Asian Paints is India’s largest and Asia’s third largest paint company, with a turnover of ₹109.70 billion. Asian Paints
operates in 17 countries and has 23 paint manufacturing facilities in the world, servicing consumers in over 65 countries. Visit
the URL http://asianpaints.com/index.aspx. From the main menus, click on the pull-down menus on the top and pursue the
links there.
Answer the following questions:
a. Based on your understanding of the products and services offered, enumerate the supply chain management challenges
that Asian Paints is likely to face.
b. Can you relate the concepts discussed in the chapter to the Asian Paints situation? Identify some strategies that Asian
Paints can utilize to manage their supply chain better.
c. How have they addressed some of the SCM challenges?
MINI-PROJECTS
1. Select two close competitors in any sector of industry and study their annual reports for the last three years.
a. Compute relevant measures of supply chain performance for both of them for these three years and report your
significant observations from the study.
b. Do you see any significant change in the performance in these companies during this period? If so, collect additional
data and list down the possible reasons for such change(s).
c. According to you, who is practicing supply chain management better? Can you support your claim by additional facts
and figures?
2. Select one company from each of the following sectors of the industry: petrochemical, automobile, machine tools, fast-moving
consumer goods (FMCG), and turnkey project management. Study their annual reports.
a. Compute relevant measure of supply chain performance for all the four companies and prepare a comparative table.
b. What are your significant observations? Do you see any pattern in these computations?
c. Relate the significant differences in the performance measures to the operating characteristics of each company.
d. Do you have any recommendations to each of these companies?
e. If these companies have to improve their supply chain performance, what are the likely areas of primary focus for each
company?
SUGGESTED READINGS
M. A. Cohen and H. L. Lee, “Strategic Analysis of Integrated Production-Distribution Systems: Models and Methods,”
Operations Research 36, no. 2 (1998): 216–228.
C. Elrod, S. Murray, P. E. and S. Bande, “A review of performance metrics for supply chain management,” Engineering
Management Journal 25, no. 3 (2013): 39–50.
E. Feitzinger and H. L. Lee, “Mass Customization at Hewlett-Packard: The Power of Postponement,” Harvard Business
Review 75, no. 1 (1997): 116–121.
M. L. Fisher, “What is the Right Supply Chain for Your Product,” Harvard Business Review 75, no. 4 (1997): 105–
116.
M. L. Fisher, J. H. Hammond, W. R. Obermeyer and A. Raman, “Making Supply Meet Demand in an Uncertain
World,” Harvard Business Review 72, no. 3 (1994): 83–93.
C. K. Hahn, E. A. Duplaga and J. L. Hartley, “Supply Chain Synchronization: Lessons from Hyundai Motor
Company,” Interfaces 30, no. 4 (2000): 32–45.
H. L. Lee and C. Tang, “Modelling the Costs and Benefits of Delayed Product Differentiation,” Management Science
43, no. 1 (1997): 40–53.
H. L. Lee, C. Billington and B. Carter, “Hewlett-Packard Gains Control of Inventory and Service Through Design for
Localisation,” Interfaces 23, no. 4 (1993): 1–11.
H. L. Lee, V. Padmanabhan and S. Whang, “The Bullwhip Effect in Supply Chains,” Sloan Management Review 38,
no. 3 (1997): 93-102.
J. Magretta, “The Power of Virtual Integration: An Interview with Dell Computer’s Michael Dell,” Harvard Business
Review 76, no. 2 (1998): 73–84.
B. S. Sahay, A. Maini, A. K. Gupta and R. K. Gupta, “Changing the Mindsets for Success in Supply Chain
Management,” Vikalpa 27, no. 3 (2002): 39–43.
T. Sastry, “Supply Chain Strategies for Small Firms Facing High Demand Uncertainty and Seasonality,” Vikalpa 24,
no. 4, (1999): 7–15.
J. Shah and A. Balram, “Improving Supply Chain Performance Through Postponement Strategy,” IIMB Management
Review 11, no. 2 (1999): 5–13.
J. Shah and N. Singh, “Benchmarking Internal Supply Chain Performance: Development of a Framework,” The Journal
of Supply Chain Management (Winter, 2001): 37–47.
M. Shukla and S. Jharkharia, “Agri-fresh produce supply chain management: A state-of-the-art literature review,”
International Journal of Operations and Production Management 33, no. 2 (2013): 114–158.
CHAPTER 6
Facilities Location
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
What role does globalization play in the facilities location problem that a firm faces?
What are the factors that influence the facilities location problem?
How is the facilities location problem solved?
What are the qualitative and quantitative methods available for choosing one location from multiple possible
locations?
How does one use the transportation model for solving the location problem?
Many facilities are located in multiple markets so that it becomes easy to access customers. HSBC, for instance, has locations
at many places all over the world.
Source: Wikimedia
ideas at Work 6.1
The controversy surrounding the Tata Nano project in the Singur special economic zone (SEZ) in
West Bengal highlights the importance of location decisions for an organization. Ratan Tata’s
ambitious objective of building the world’s cheapest car became a victim of a tussle between the
CPI(M) leadership and the Trinamool Congress. For the CPI(M), the Nano project could have
been a symbol of the new spirit of industrialization in the state. This was important because, over
the years, West Bengal had been losing out to new hubs of corporate excellence such as
Bangalore and Hyderabad. For Trinamool Congress leader Mamata Banerjee, however, it was an
opportunity for gaining the goodwill of the farmers as she felt that they had been given a raw
deal when their farmland had been acquired. Thus, location decisions seem to have a political
dimension in a country like India due to its growing population and the paucity of land.
Location decisions are an integral part of designing a supply chain for an organization as it
determines the flow of materials from the raw material suppliers to the factories, and finally to
the customers. Location decisions affect the overall profitability of a firm. This is because the
cost of manufacturing and the total cost of logistics and distribution are direct fallouts of location
decisions. For instance, locating manufacturing facilities far away from the market could raise
the logistics and distribution costs. On the other hand, locating manufacturing facilities in
regions with high factor costs may make the firm unattractive as the cost of manufacturing may
turn out to be high. Moreover, once location decisions are made, they are likely to stay longer as
huge investments are made. Therefore, location decisions are strategic in nature.
Location decisions pertain to the choice of appropriate geographical sites for locating various
manufacturing and/or service facilities of an organization. In reality, a firm may have to locate
several operating units to cater to the requirements of the market. At one extreme, it is possible to
have a single location in which all facilities could be located. Products are manufactured in this
facility and sent to the markets using a distribution system. Companies in the aerospace and
defence sectors, such as Boeing and Airbus, are examples of this category. Another example is
the printer-manufacturing facility of Hewlett-Packard. At the other extreme, many facilities are
located in multiple markets so that it becomes easy to access customers. Automobile
manufacturers such as Ford and Toyota are examples of this category. Essel Propack, the Indian
manufacturer of tubes for toothpastes, creams, etc., has factories in all continents except
Australia. Both the extremes may prove inappropriate in most situations. However, they
highlight the basic trade-off involved in solving the location puzzle.
Location decisions pertain to the choice of appropriate geographical sites for locating the various manufacturing
and service facilities of an organization.
Locating facilities in regions that offer attractive cost advantages is one aspect of the trade-off.
It may be expensive to set up a distribution system that is both efficient and responsive (it leads
to an increase in transportation costs and the costs of coordinating and communicating with the
supplier about products and availability). On the other hand, locating facilities very close to the
markets provides the advantages of quicker responses and better customer service.
Operations managers need to evaluate intermediate solutions to arrive at an optimal location of
facilities. Location planning essentially addresses this issue and provides the operations manager
the needed tools and techniques to study the location problem. Location decisions affect other
functional areas of business as well. Choice of locations is an important input to marketing as the
management needs to plan their marketing strategy in relation to the location of facilities vis-àvis
the customers. In this chapter, we shall look at these issues in some detail and understand how
operations managers could make decisions with respect to location of the organization facilities.
Location issues have become more prominent in recent years due to the increased pace of
economic reforms in several countries and the consequent globalization of markets. The
globalization of markets opens up new opportunities to multinational corporations on issues
related to location, which they have not faced before. For instance, ABB decided to identify
factories that could produce world-class products at internationally competitive prices.
Moreover, the factories needed to have high levels of technical capability and domain expertise.
This exercise resulted in selecting plants at Vadodara and Nasik for circuit breakers of ranges
above 72.5 kV.1 Similarly, their motor division in Faridabad, the only one available east of the
Suez Canal, is capable of making variable drive motors for global markets.
These examples amplify the close relationship between globalizing operations and new
dimensions to the location of manufacturing facilities. Therefore, we shall examine in detail the
factors that have caused globalization of operations and how these influence the location
decisions of firms. Figure 6.1 presents a schematic representation of the impact of globalization
on the location decisions of organizations.
Regulatory Issues
The most significant factor that drives globalization is the ongoing economic and regulatory
reforms in several developing countries. Beginning from 1991, we embarked on a set of
regulatory changes that made India much more attractive in terms of locating a manufacturing
facility. Two events have been broadly responsible for this. First is the reduction in customs and
excise tariffs and a move towards the single-point value-added tax (VAT) regime. The other is
the delicencing of several sectors of industry and the progressive removal of the cap on foreign
direct investment (FDI). Further, there has been a progressive simplification of the procedural
aspects of setting up operations here and running them on a day-to-day basis.
The removal of entry barriers and reduction in the cost of manufacturing due to tariff reductions will make India
an attractive destination to which manufacturing bases can be shifted.
These changes have several implications for the location choices of multinational firms. The
removal of entry barriers and reduction in the cost of manufacturing due to tariff reductions will
make India an attractive destination to which manufacturing bases can be shifted. In the case of
multinational firms with manufacturing facilities already located in India, these facilities will be
attractive candidates for further development and growth.
FIGURE 6.1 Impact of globalization on location decisions of organizations
Another relevant point for location planning with respect to regulatory issues is the emergence
of regional trading blocs. A trading bloc is essentially a group of geographically separated
nations that provide advantageous access to markets, manufacturing facilities, and technologies
within the member countries. If required, it could also have a common currency, as in the case of
the European Union. Such an arrangement provides multiple options for locations for an
organization operating in such markets. India is a member of two trading blocs, the South Asian
Association for Regional Cooperation (SAARC) and the Association of Southeast Asian Nations
(ASEAN).
Factor Advantages
The globalization of manufacturing is also triggered by factor advantages that an organization
can enjoy by operating in specific locations. Generally, developed countries in Europe and in the
United States are characterized by the high cost of labour. On the other hand, developing
countries offer significant advantages to a firm due to the availability of cheap labour. Therefore,
in location planning, these advantages are likely to be considered. The recent shifting of
manufacturing bases to China and India and the large-scale shifting of BPO activities to India are
primarily related to the factor cost advantages that these countries offer to firms in the western
world. For a similar reason, the bulk of semiconductor and electronic appliances manufacturing
has been shifting to countries such as South Korea, Taiwan and Malaysia.
Factor advantages may encourage a firm to globalize its operations.
Factor advantages also accrue to a firm on account of the availability of skilled labour and
other resources required in plenty for manufacturing. The other resources may be power and
water (as in the case of the chemical processing industry) or the availability of technical
infrastructure in the form of well-developed ancillary industries. Essentially, when these
resources are available in plenty, firms develop confidence that the location of their facilities will
benefit them in the long run and enable them to make further enhancements in the product
offering without much difficulty.2 Moreover, the plentiful availability of these resources will
introduce a healthy competition among them, paving the way for cost reduction and the overall
improvement of quality and delivery reliability. These factors may significantly influence the
location decisions of a firm and encourage the firm to globalize its operations.
From a cursory analysis of the table, one may come to the conclusion that it is relatively easier
to do business in China compared to India. This is because there are fewer procedures related to
starting or closing business in China. Furthermore, it is easier to enforce contracts. We also
notice that the costs associated with running the business, closing, or enforcing the contracts are
much lesser in China. It is evident from this example that one can develop a set of factors that
affect the location decisions of organizations and use it as a basis for the site selection.
In order to study the factors that affect a location decision in an organization, it is useful to
categorize them and understand how each of these are relevant to the location problem. Table 6.2
shows a four-part classification of the various factors that affect the location decision.
Market-related Issues
The first set of factors in Table 6.2 comprise market-related issues. The existence of a market for
the firm’s products and services significantly influences the location decision. In general, it is
very desirable to locate the operating system close to either the market or the source of raw
material and other critical inputs for the system. Locating systems far away from these will often
make the system unviable due to large transportation and distribution costs arising out of
bringing the product to the market where it is ultimately consumed. Availability of raw material
and other inputs also significantly affect the overall cost of the system, and hence the
profitability. In India, successive governments have been trying to woo investors to backward
and underdeveloped regions by providing tax rebates and other sops. In spite of that, firms have
not been attracted to these places. This is due to the market-related issues of location planning
and the absence of infrastructure facilities. The importance of market-related issues is much
more in the case of a service organization.
The other related issue is the size of the market. A large market provides several advantages to
an organization. It has higher demand for products and services. It also assures ample supply of
key human resources for the operating system. Large markets are very attractive locations, as we
currently see in the case of countries such as China and India. Furthermore, the nature of
competition also determines the suitability of a location. If a market is already saturated and
there are already several established firms competing for the market share, it requires
considerable effort to successfully establish operations in such a market.
TABLE 6.2 Factors Affecting Location Decisions
Cost-related Issues
Cost-related issues capture the desirability or otherwise of competing locations on the basis of
the cost of operating the system in alternative locations. Typically, logistics and distribution
costs are often considered for the analysis simply because these costs are direct, tangible, and
easy to measure and analyse. However, depending on the availability of data and sophistication
of the analysis, other costs, such as wages or input factor costs and tax and other tariff benefits
could also be included. Our earlier discussion on the hesitation of firms to locate their operating
systems in far-off and underdeveloped regions could be easily justified by taking a cost-oriented
approach to location planning.
Typically, logistics and distribution costs are often considered for location analysis because these costs are
direct, tangible, and easy to measure and analyse.
Other Issues
Finally, other issues that influence a location decision include cultural, linguistic, and climatic
aspects, along with the overall quality of life. These factors provide a certain assurance to firms
that operating systems can work without much difficulty, as the key input for the system, in the
form of human resources, will be readily available.
Any location planning decision needs to take these factors into consideration. Depending on
the scope of the location planning and the type of modelling of the problem, location planners
can judiciously make use of these factors for selection from among competing candidates.
6.3 LOCATION PLANNING METHODS
Generally, location decisions could be one of the two: one facility–multiple candidates and
multiple facilities–multiple candidates. In the former, the decision boils down to selecting one
location out of a group for a single facility. For example, a firm may want to choose one location
out of several alternatives to build its factory. In the process of selecting the ideal location, one
can suitably incorporate any of the factors discussed so far and assess the attractiveness of each
location on their basis. We shall discuss two methods used to perform this exercise.
Generally, location decisions could be one of the two: one facility–multiple candidates and multiple facilities–
multiple candidates.
The second type is one of selecting k locations out of the n alternatives available. A typical
example could be a firm wanting to locate four regional warehouses to serve the demand
throughout the country. Therefore, it may have to choose four locations out of the 10 alternatives
available to it. Some sort of network flow modelling is involved in this case to decide on an
appropriate set of locations. In addition to choosing k locations, the location planner must also
decide how these locations will serve multiple markets that are spread out geographically. One of
the popular network models is the transportation method, which could address this issue.
Factor rating is a simple methodology to assess the attractiveness of each potential location.
The advantage of this method is its ability to incorporate any factor into the analysis as long as
the decision maker can assess its relative importance. It is also simple to compute and
comprehend. However, the limitation of this method is that it can be used only for initial
screening and broad-level ranking. A detailed cost-based analysis may be desirable to assess the
quantum of benefit or impact of each candidate location before a final decision is made on the
location.
The four-step process of identifying an appropriate location is as follows:
1. Identify and list all the relevant factors for the location decision. We have already discussed some of the factors to be
considered. One can choose several factors from the list and add a few more that are specific to the context.
2. Establish the relative importance of each factor in the final decision. The easiest method to establish the relative
importance of each factor is to rate each factor on a scale of 0 to 100. Once all the factors are rated, these ratings can be
normalized to obtain the relative weights. A simple example of normalization is provided in Table 6.3 for a hypothetical
case of six factors.
3. Rate the performance of each candidate location using a rating mechanism.
4. Compute a total score for each location, based on its performance against each factor, and rank them in the decreasing
order of the score.
EXAMPLE 6.1
A manufacturer of garments is actively considering five alternative locations for setting up its
factory. The locations vary in terms of their advantages to the firm. Hence, the firm requires a
method of identifying the most appropriate location. Based on a survey of its senior
executives, the firm has arrived at six factors to be considered for final site selection. The
ratings of each factor on a scale of 1 to 100 provide this information. Furthermore, based on
the detailed analysis of both the qualitative and quantitative data available for each of the
locations, the ratings of the locations against each factor have also been arrived at (on a scale
of 0 to 100). Using this information (see Tables 6.4 and 6.5), obtain a ranking of the
alternative locations.
Solution
Establishing the Relative Importance Through Normalization
The first step in the solution is to establish the relative importance of each factor using a
normalization procedure. The sum of all ratings is 325. Therefore, by dividing each factor
rating by 325 one can obtain the relative weight of the factors. The workings are presented in
Table 6.6.
TABLE 6.4 Factor Ratings
Factors Rating
Availability of infrastructure 90
Size of the market 60
Industrial relations climate 50
Tax benefits and concessions 30
Availability of cheap labour 30
Nearness to port 65
Note: The overall scores computed in the table are more accurate as the relative importance of the factors are not rounded to two decimals, as shown
in the sample computation for Location 3.
The centre-of-gravity method merely indicates the ideal location in the grid map that would ensure that the
weighted distances travelled on the whole is minimum.
The centre-of-gravity method merely indicates the ideal location on the grid map that would
ensure that the weighted distance travelled on the whole is minimal. However, location planners
may have to include other practical considerations before making a final decision. For instance,
the coordinates indicated by the centre-of-gravity method may point to an infeasible site to set up
a plant. In some cases, it may be very expensive to acquire the land and develop the site. There
could be regulatory constraints in locating a factory (as in the case of having a factory in a
residential area). Therefore, one method to solve this problem could be to evaluate multiple sites
in the area and select one among them.
EXAMPLE 6.2
Solution
Let us represent the given information using the following notations:
Number of existing demand (or supply) points in the grid map = n
Coordinates of Location i in the grid map = (xi, yi)
Quantum of shipment between existing demand (or supply) point i and proposed facility = Wi
Coordinates of the centre of gravity in the grid map = (X,C, YC)
In our example, the location of supply point B is represented as (xB, yB) = (350, 400).
The coordinates of the centre of gravity are given by:
Using the information available in Figure 6.3 for Locations A to D and Eq. 6.1 one can compute
the centre of gravity for the above example to be (366,376). The figure also has the plot of the
centre of gravity. Therefore, the manufacturer might want to locate the new facility at this point.
FIGURE 6.3 Grid map of the target market
The Load–Distance Method
As opposed to the centre-of-gravity method, the load–distance method enables a location planner
to evaluate two or more potential candidates for locating a proposed facility vis-à-vis the demand
(or supply) points. The load–distance method provides an objective measure of the total load
distance for each of the potential candidates. Choosing the location with best load distance
among these will satisfy the objective of identifying an appropriate location for the proposed
facility. Distance is measured using a Cartesian measure.
The load–distance method enables a location planner to evaluate two or more potential candidates for locating a
proposed facility visà-vis the demand (or supply) points.
EXAMPLE 6.3
Consider Example 6.2. Assume that the manufacturer came to know that there are constraints in locating the new
facility. Based on an initial survey of possible sites for the proposed facility, the manufacturer identified four
candidates. Figure 6.4 has the location coordinates of the four candidates (numbered 1 to 4). What is the best location
for the proposed new facility?
FIGURE 6.4 Location coordinates of the four candidates
Solution
Let us first summarize the coordinates of the existing supply points and the candidates for the
proposed facility by tabulating them from the grid map for the target market. Let us also
tabulate the quantum of shipment from these supply points to the proposed facility. Table 6.8
presents this information.
Using the information available in Table 6.8 and Eq. 6.2, one can compute the Dij values
between every pair of supply point and candidate for locating the new facility. The
computation for A-1 is as follows:
TABLE 6.8 Coordinates of Supply Points and Proposed Locations
Computing for all the pairs in this manner, one can obtain a matrix of Dij, as shown in Table
6.9.
TABLE 6.9 Dij Values
Using Eq. 6.3, one can obtain the LDj values and select the candidate with the least value
of LD. The computation of LDj is available in Table 6.10.
In the case of multiple plants and multiple demand points, the location problem can be solved using the standard
transportation model.
The transportation model is an interesting variation of the basic linear programming (LP)
model, which could be used to optimally identify a subset of supply points from a potential list
that can satisfy the demand at various demand points. There is a unit transportation cost of
shipping material from every supply point to every demand point, and by an optimal choice of
supply points, the total cost of transportation is minimized. In the process of doing this, it is
ensured that none of the capacity constraints in the supply point is violated. Similarly, it is also
ensured that at every demand point the requirement is fully satisfied. Figure 6.5 shows the
typical representation of a transportation problem.
FIGURE 6.5 Typical representation of a transportation problem
In this example, there are four factories that could supply to five markets. The last column in
the table shows the capacity available in each factory. For instance, Factory C has 700 units of
capacity whereas Factory D has just 200 units of capacity. The last row shows the demand for
the product in each of the five markets. The demand in Market 1 is 200 units and the demand in
Market 5 is 400 units. Since the markets and the factories are physically separated, one could
expect different costs of transportation between each pair of markets and factories. This
information is available in the cells of the table. The numbers inthe cell (upper right hand side)
denote the cost of transporting one unit from a factory to a market. In our example, it costs ₹100
to transport one unit from Factory A to Market 1.
An important decision in this problem is to decide which of the factories will serve the
markets and how many units should be transported from a factory to a market. The transportation
model provides the optimal combination of a factory with a particular market and also identifies
how many units need to be shipped. Location problems in the case of multiple plants and
multiple demand points follow exactly the same pattern; therefore, one can use the standard
transportation model to solve the multi-location problem too. The transportation model is
discussed in any standard and elementary level textbook on operations research. However, the
application of the transportation model in a location problem is illustrated using an example.
EXAMPLE 6.4
Solution
Constructing the Transportation Table
Let us first construct the initial transportation table by adding the available demand and
supply information. Figure 6.7 has the initial table. In this case, we find that the demand and
supply are exactly balanced. If the supply and demand are unbalanced, the first step is to
make the transportation problem anced, the first step is to make the transportation problem
balanced by adding dummy supply (or demand) points. If the demand is greater than the
supply, then we need to add a dummy supply point with zero cost of transportation.
Similarly, if the supply is more than the demand, we need to add a dummy demand point
with zero cost of transportation.
FIGURE 6.7 The initial transportation table
There are several heuristic methods to solve the transportation problem. Optimal methods
are also available. Typically, the heuristic methods are used as the starting solution for the
optimal method. One heuristic, Vogal’s approximation method (VAM) is known to provide
close-to-optimal solutions to the transportation problem.4
Computing the Difference Between the Minimum Cost Cell and the Next Higher Cost
Cell
The maximum value of the differences as computed in Figure 6.8 is 20. Since it occurs at two
places, we break ties arbitrarily and choose Market 2. We shall assign the maximum to the
least cost cell in the column. Using this logic, we assign 1500 to Warehouse C in that column
and adjust the demand and supply values. Since the entire requirement of Market 2 is met
with, we strike off the column. We also note that after this allocation, the balance capacity
available with Warehouse C is 2200. After making these adjustments, we return to Step 1 of
the algorithm and recompute the differences and proceed with the allocation in the same way
as mentioned above.
FIGURE 6.8 The first iteration of VAM
The final allocation table obtained in this manner is shown in Figure 6.9.
The final solution and the cost of the plan is as shown in Table 6.11.
FIGURE 6.9 The final allocation table
TABLE 6.11 The Final Solution and the Cost of the Plan
Recent trends in international markets point to a shift towards fewer facilities that could serve
markets worldwide. For example, the HP DeskJet printer manufacturing facilities are available
only at two places, Vancouver in Washington State in the United States, and Singapore. Several
other appliance manufacturers are gravitating towards fewer and fewer distribution centres.
These developments point to the fact that there are two aspects that could affect the location
planning problem significantly. One is the availability of good transportation infrastructure and
the other is the use of the Internet and IT infrastructure. Today’s operations managers must know
how these two tools affect location design.
The availability of good transportation infrastructure enables firms to invest in fewer and larger facilities and
provide a good level of service to the customers.
The availability of good transportation infrastructure enables firms to invest in fewer and
larger facilities and provide a good level of service to customers. Infrastructure includes
availability of an all-weather high-speed road network, good railway infrastructure that is
capable of transporting bulk cargo, availability of ports that have faster turnaround times and
good connectivity to the ports. The recent attempts by the Central Government pertaining to the
Golden Quadrilateral Projects seek to address this issue. Once the project is complete, one can
expect significant changes in the approach to location planning.
In the era of the Internet and IT, providing useful interfaces for customers may offset some of
the shortcomings of location decisions. In recent years, the advent of the Internet has resulted in
significant reductions in the cost of communication and coordination among trading partners.
This has great relevance for the location planning problem that operations managers face. Due to
the reduced cost of communication and coordination, it is possible to have fewer locations and
still provide better customer service. Developing home pages that provide for high levels of
customer interfacing and interaction may enable an organization to directly deal with the
customers and provide good service even without a physical facility within close proximity of
the customer.
In a service-intensive industry, nearness to the market will be more important than the cost of the site.
Another issue pertaining to location is the supply management practice of leading Japanese
manufacturers such as Toyota. Japanese manufacturing management techniques lay an emphasis
on location planning in the overall context of the just-in-time manufacturing philosophy. In
Japan, it is often the practice to locate a good number of suppliers in the vicinity (20–40km
radius) of the manufacturer. Locating suppliers who are close by provides several advantages to
the firm. It enables them carry almost zero inventory. Typically, a milk-van-type routine is
followed, in which a truck goes round and collects hourly or half-day requirements of material
from each of these suppliers and delivers them at the factory premises. Several Indian automobile
manufacturers have similar practices with respect to locating their suppliers. The Ambattur and
Guindy industrial estates in Chennai cater to the requirements of all major automobile
manufacturers in Chennai. Similarly, Maruti has several suppliers located in the Gurgaon–
Faridabad belt on the outskirts of Delhi. Locating suppliers close by helps, especially when the
quality of the infrastructure is not good. Further, it helps minimize disruptions arising out of
bandhs, strikes and natural calamities.
Finally, location issues in a service-intensive industry are different from those in a
manufacturing industry. Since service quality is largely determined by the responsiveness of the
service delivery system, locating service outlets as close to the demand point as possible is an
important requirement in a service system. Therefore, nearness to the market will be more
important than the cost of the site, as opposed to a manufacturing system, where the cost of the
site would take precedence over location.
SUMMARY
Location issues have become more prominent in recent years on account of the globalization of markets. Multinational
corporations have more opportunities to identify candidate locations for their manufacturing facilities.
Factor cost advantages and expanding markets in developing countries have made these nations attractive for locating
new facilities.
Simple qualitative methods are useful for quickly screening an initial set of candidates and narrowing down the choice
to one or two.
The load–distance and centre-of-gravity methods help evaluate the suitability of candidate solutions from a perspective
of distance and the quantum of items to be transported between a location and the demand points.
The transportation method helps in optimally identifying a set of k locations out of n candidate solutions.
The availability of good transport infrastructure and recent developments in Internet technology suggest that it is
possible to have fewer locations and still provide better customer service.
Location decisions in service systems must address the requirement of speed of responsiveness. Therefore, locating
service outlets as close as possible to the demand points may be highly desirable in service systems.
REVIEW QUESTIONS
1. How important are location decisions in operations management? What are the consequences of a bad location decision?
2. What factors drive the globalization of operations? Do these have any bearing on the location decision that an operations
manager makes?
3. Will the choice of factors for selecting an appropriate location vary with the geographical spread of the potential candidates
for location? Why?
4. Identify three important factors that a location planner may consider with respect to each of the following:
5. What are the pros and cons of using the location factor analysis method for location planning? Do you have any
recommendations on how to use this method for location planning?
6. A manager intending to use the location planning model for analysis is not sure about which of the two models, the centre-of-
gravity method and the load–distance method, should be used for his analysis. Prepare a one-page note that will help him
identify an appropriate model for his consideration.
7. What are the key differences between using a transportation model and a load–distance method for solving a location
problem?
PROBLEMS
1. A firm is considering three alternative locations for setting up its new factory. Each of these locations provides some
advantages and some limitations. Therefore, it is necessary that some method of assessing the attractiveness of each site be
arrived at. Based on a survey of its top management, the company has identified seven factors that will determine the
appropriateness of a site for setting up the new factory. These factors and the score out of 100 for each of them are shown in
Table 6.12.
The company collected some data pertaining to each site that it is considering, and, based on
some criteria, was able to arrive at how each site rates against a score of 100 against each factor.
In some cases, actual data was estimated on the basis of some projections. Table 6.13 has details
pertaining to these:
TABLE 6.13 The Ratings for Each Site Against Each Factor
Identify the most appropriate site for locating the new factory on the basis of the above information.
(Hint: Suitably convert the actual data into scores out of 100 and then analyse)
2. A chain of stores trading in a host of retail goods for urban customers is planning to set up operations in one of the
metropolitan cities in the country. The exact location for the new outlet is to be decided. In order to do that, the company
conducted a market survey and found five pockets of consumption in the city. The distance coordinates of these five pockets
from a reference point in the grid map pertaining to the city are as follows: Zone 1 (15, 22), Zone 2 (10, 40), Zone 3 (35, 15),
Zone 4 (50, 5), and Zone 5 (40, 35). The annual demand for the goods (in tonnes) in the five zones is 200, 130, 80, 170, and
120, respectively.
a. What is the centre of gravity for setting up the operation in the city?
b. If the cost of transporting the goods is ₹300 per tonne, what is the cost implication of locating the stores at the centre of
gravity?
3. The chain store officials found that locating the operations in the centre of gravity point, as obtained in the above problem,
would cost them a lot. The operational costs of renting and maintaining the site will be ₹2,500,000 per annum. In view of this,
they are now considering other options. They have identified three other candidate locations whose cost of operations will be
less than the centre of gravity location by 10 per cent, 25 per cent and 18 per cent, respectively. The locational coordinates for
these three are (20, 40), (40, 25), and (35, 45), respectively. What should the chain store do in this case? Does it make sense to
set up operations in any of these three alternatives locations?
4. A manufacturer of certain household goods caters primarily to the south Indian market. At present, the manufacturer has
located six stock points in the major south Indian cities of Hyderabad, Mysore, Chennai, Kochi, Pune, and Vizag to cater to
the demand. There are four warehouses from which the components are distributed. In order to save on the cost of operations,
the warehouses are typically located away from these cities in different locations. The unit cost of transporting the items from
each of these warehouses to the cities (₹tonne), the capacity of the warehouses, and the demand at the city stock points per
unit time (in tonnes) are all summarized in Figure 6.10. Identify the least cost schedule for deciding which warehouse will
send the items to the cities.
5. Suppose, in the above problem, the company has decided to shut down its Kochi operations and to serve the requirements of
Kochi from Mysore. Will the decision change? What will the cost impact of this decision be?
6. Consider Problem 4. Suppose the company has received a proposal for setting up another warehouse (Warehouse E) with a
capacity of 900 units. The location of the warehouse is such that the unit costs of transporting to the six cities are as follows:
Hyderabad–45, Mysore–75, Chennai–30, Kochi–65, Pune–90 and Vizag–80. The company would like to know how the
allocation will change if the warehouse is made operational and the likely cost impact on the operation of the system.
(Hint: Its introduction of will result in an unbalanced transportation problem. Convert it into a balanced transportation
problem and solve.)
a. Location issues play a very important role in setting up a manufacturing plant. Can you analyse why the plants have
been set up at these places on account of the availability of electricity, water, transport facilities, etc?
b. Do you think that the spread of potential customers plays an important role in selecting the location?
MINI PROJECT
1. Several cities now have chain stores that sell a variety of goods. These chain stores may be like Vivek’s (selling household
appliances and white goods), Garment manufacturers (such as Raymonds), supermarkets (such as Food World and Big
Bazaar), high-end malls (such as Shoppers Stop and Life Style), and eating outlets and joints (such as Café Coffee Day).
Select two such categories of chain stores for the study. Visit the chosen chain stores, interview the operations manager and
the owner of the franchisee and elicit information on the various factors that are important for choosing an appropriate site for
locating their retail outlet. For the important factors identified, obtain the relative importance of the factors as perceived by the
companies.
Based on this exercise, prepare a final report that covers the following aspects of the location
decision-making framework
a. What are the important factors that drive the choice of location in each of the two examples that you have studied? How
important are they to these examples?
b. Are there significant differences between the importance of the factors in the two examples that you have studied?
Explain why these differences occur.
c. If a new player wants to set up an operation in the same area as the one that you have looked at, explain how you will
advise the new player on its choice of an appropriate location for their operation.
CASE STUDY
The Tamil Nadu Small Tea Growers Industrial Cooperative Tea Factories Federation Limited,
usually called Indcoserve, was established in the year 1965 and is functioning under the
administrative control of the Department of Industries and Commerce, Government of Tamil
Nadu. With the establishment of the first Industrial Cooperative Tea Factory at Kundah in 1962,
the small growers in the Nilgiris region in Tamil Nadu were able to realize a better price for
their green leaf. In the four decades of its existence, Indcoserve has enabled its 15 member
industrial cooperative tea factories to command 17 per cent of the total production of tea in the
Nilgiris District by offering its expertise in technical, managerial, and commercial competence.
The 15 factories are functioning in the Nilgiris district with an aggregate installed capacity to
produce over 27.3 million kg of tea per annum. Nearly 20,000 small growers supply green leaf
to these factories that are geographically spread out in the region. Table 6.14 presents some
basic details on these factories. The table has data for 17 factories. However, two factories—
Kotagiri and Kilkotagiri—have discontinued their operations. At the outset, it is evident from
the table that the 15 factories vary considerably on several parameters. For instance, Salisbury
has the highest value in all the parameters. Nearly 20,000 small growers supply green leaf to
these factories that are geographically spread out in the region.
Indcoserve factories have a serious problem of price realization. Except for the period 1997–
1999, the price realized is much lower. The cost of production has gone up to 300 per cent of the
1989–1990 levels. This disparity between price and cost trends is highly undesirable and may
threaten the viability of Indcoserve. Therefore, the management would like to assess the
performance of these factories in order to make certain choices with respect to location of the
factories and their future plans.
The quantity procured across factories fluctuates wildly in every period. During all the 29
months, Salisbury procured the highest quantity of leaves (at an average of 0.824 million kg per
month as opposed to the overall average of 0.309 million kg per month). For nine out of 15
factories, the maximum-to-minimum ratio was in excess of 300%). Such marked variations
between Salisbury and other factories cannot be explained in any manner other than the
acceptance that there were significantly varying procurement practices among the 15 factories.
All the 15 factories exhibited significant price differences for the green leaf that they procured.
These differences existed in each factory throughout the 29-month period of observation.
Salisbury not only paid higher rates, but also procured the maximum quantity during the
observed period.
Production of Tea
The data on production-related details are available in Table 6.16. Based on the data available, it
was found that Salisbury was the highest producer in 28 out of 29 months of data that was
analysed at an average production of 0.207 million kg per month (as opposed to an overall
average of 0.08 million kg per month). Ebbanadu (0.027 million kg per month), Karumbalam
(0.032 million kg per month) and Kinnakorai (0.035 million kg per month) are the lowest
producers of tea during the observed period.
The level of technology used in tea manufacturing continues to be low in Indcoserve factories.
Therefore, a useful measure for comparison is the labour productivity. Labour productivity is
measured using “the number of kg of tea produced per person employed in the factory”. The
relevant data is available in Table 6.16. It is evident from the table that productivity differences
across the factories are very significant and that Salisbury scores higher than all others in the
group. Salisbury employs the maximum number of peoples among the 15 factories.
All factories seem to have very similar, albeit high cost of production (excluding the cost of
GL) close to the average of ₹18.68 per kg (Table 6.17). The only exceptions to this are Salisbury
(₹12.93 per kg), Pandalur (₹14.97 per kg) and Frontier (₹15.27 per kg). A comparison of the
maximum and minimum values observed during the 29 months of observation also shows that
these factories had better cost control (evidenced by a narrower range).
TABLE 6.15 Green Leaf Purchased in the 15 Factories During April 2004–March 2006
TABLE 6.16 Tea (Production) in the 15 Factories During April 2004–March 2006
TABLE 6.17 Production Cost Excluding Green Leaf Cost in the 15 Factories During April 2005–March 2006
TABLE 6.18 Average Sale Price in the 15 Factories During April 2004–March 2006
SUGGESTED READINGS
J. M. Apple, Plant Layout and Material Handling, 3rd edition (New York: John Wiley & Sons, 1977).
M. E. DeForest, “Thinking of a Plant in Mexico?” The Academy of Management Executives 8, no. 1 (1994): 33–40.
Z. Drezner and H. Hamacher, Facility Locations: Applications and Theory (Berlin: Springer Verlag, 2002).
R. Dubey, “Clustering for Competitiveness,” Business Today, 22 April 1999, pp. 50–63.
R. L. Francis and J. A. White, Facilities Layout and Location: An Analytical Approach (Englewood Cliffs, NJ:
Prentice-Hall, 1977).
R. D. Klassen and D.C. Whybark, “Barriers to the Management of International Operations,” Journal of Operations
Management 11, no. 4 (1994): 385–396.
M. Koksalan and H. Sural, “EFES Beverage Group Makes Location and Distribution Decisions for its Malt Plants,”
Interfaces 29, no. 2 (1999): 89–103.
J. Magretta, “Fast, Global and Entrepreneurial: Supply Chain Management Hong Kong Style,” Harvard Business
Review 76, no. 5 (1998): 103–114.
I. V. E. Martin, “Centralised Bakery Reduces Distribution Costs Using Simulation,” Interfaces 28, no. 4, (1998): 38–46.
“New Engines of Growth,” Business India, 18–31 May 1998, pp. 126–134.
A. Pal, “ABB Global Factories,” Business Today, 22 June 2003, pp. 122–126.
M. R. Porter, “Clusters and the New Economics of Competition,” Harvard Business Review 76, no. 6 (1998): 77–90.
A. Ravindran, D. T. Philips and J. J. Solberg, Operations Research: Principles and Practice, 2nd edition (New York:
Wiley, 1987)
H. A. Taha, Operations Research:An Introduction, 8th edition (New Delhi: Pearson Education, 2008).
CHAPTER 7
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Good sourcing and supply management practices enable an organization to obtain good-quality products and services at
competitive prices. They also help organizations build relationships with trading partners.
Source: Faraways.Shutterstock
ideas at Work 7.1
Sourcing and supply management are important functions in any organization. In recent years,
the focus on supply management has increased significantly on account of several developments
in the market and the competition. Some of the salient aspects of these developments are
discussed in this section.
A defect-free input is not achieved merely by rock-bottom price negotiations with suppliers.
When the bulk of the cost incidence of a product is fixed a priori by input costs, emphasis on input cost
reduction becomes important.
Quick-response Requirements
As competition increases in markets, customers become more demanding. Firms are required to
respond to the market requirements much more rapidly. Good supply management practices are
central to any quick-response system. Organizations need to have less lead time and greater
flexibility to respond to the changes in the marketplace. This calls for closer ties with the
suppliers and the ability of the suppliers to match the requirements placed by the buyers. It also
calls for alternative methods for cutting down the procurement lead time.
The requirements of faster new-product introduction and good supply management practices
have placed several new demands on the purchasing and supply management practices in every
organization. Figure 7.1 briefly captures the important changes that have taken place in this
process in response to these requirements. Organizations have moved from multiple suppliers to
fewer (and often one supplier) for an item. Typically, cooperative and long-term relationships are
established with these suppliers. Due to these new arrangements, suppliers are able to deliver
quick-response capabilities to an organization through short lead time and frequent deliveries.
Moreover, the quality level is also found to increase significantly. Before we analyse the nature
of activities undertaken to implement the transitions indicated in Figure 7.1, let us understand the
notion of strategic sourcing and the various steps involved in a typical procurement process. A
good understanding of the procurement process is crucial to appreciate the impact of the new
procurement methods and the supply management practices, which will be discussed in a later
section.
FIGURE 7.1 Transitions in purchasing and supply management practices
7.2 STRATEGIC SOURCING
Product and component categorization: Sourcing strategies often focus on a specific part, component, service, or
product. A one-size-fits-all sourcing strategy makes little sense across a company’s category portfolio. Suppliers might
be plentiful and eager to compete in some categories, while for other categories, supply options might be limited. Some
categories requested by internal customers may have a larger impact on business in terms of total expenditure or on a
company’s strategic position, while the impact of other categories might be relatively small. In many cases, very
specific characteristics of the parts, services, or products requested by internal customers will drive the sourcing
strategy, while the supplier perspective or supply-market perspective will be primarily driven by those requirements. In
some cases, products or services might only be sourced from a specific supplier due to worldwide intellectual property
rights or unique access to resources, giving the relationship with this supplier a more prominent role in the sourcing
strategy. Finally, factor cost advantages or local content requirements might lead to the decision to source a specific
spend volume from a single supply market.
Often, an analysis of the components that need to be procured will lead the efforts in strategic sourcing into two
directions:
Focus on components that are critical to the product and the company is good at making.
Outsource all other components where suppliers have a distinct competitive advantage, greater scale, lower
cost, or greater performance incentives.
In a large multi-product multi-division organization such as Bosch Limited, the approach to supply management
depends on the classification of parts and suppliers. The suppliers are classified as preferred suppliers (P), technology
specialists (T), customer-specified suppliers (D), new suppliers (N), Bosch Limited approved essential suppliers (E),
suppliers for elimination (X), and suppliers without new business (W). After an initial classification of suppliers as P, T
or D, the other classifications are arrived at using a structured supplier evaluation methodology.
New source development and supply management: Supply management considerations include all the activities
necessary to develop and implement sourcing objectives with the overall supply base and individual suppliers. Key
strategy elements are the number of suppliers and the kind of relationship, such as arm’s-length transactions or
collaborative partnerships. Furthermore, the number of suppliers for a specific product or service in the form of sole
sourcing, single sourcing, dual sourcing or multiple sourcing is important. A good supply management programme will
endeavour to constantly identifying reliable and capable sources that provide good-quality inputs at lower costs. Further,
continuous efforts will be invested in developing the existing vendor base in terms of technology, quality, and process
upgradation. Shrinking the vendor base in order to reduce the transaction costs arising out of the procurement process
will also be another aspect of the supply management process.
Logistics and supply organization: The sourcing function will enable all activities necessary for developing and
implementing the sourcing objectives of the organization, as well as the other functions involved in the sourcing
process. The sourcing function requires an organizational set-up that helps in the alignment of requirements and
specifications across multiple product categories and geographical locations. Relevant activities are standardization and
bundling. Standardization aims to find or increase similarities between products and services which need to be sourced
in order to improve performance measures such as cost reduction or process improvements. Bundling efforts will
increase the sourcing volume for specific categories across different business units or companies in order to improve the
negotiation power towards suppliers and realize economies of scale.
From a logistics perspective, requirements such as stock delivery, just-in-time delivery or just-in-time purchasing need
to be addressed.
Performance metrics for sourcing and supply management: Traditional measure such as purchase price is grossly
inadequate to measure the performance of the sourcing and supply management function. Several organizations have
realized that while price is a major component in the sourcing decision, other attributes are equally important.
Furthermore, the performance of the sourcing function should not be merely judged by tactical parameters. Measures
indicating the capability of the suppliers, the competitive advantage that they bring to the business, and the rate at which
they improve on certain key parameters are important. New metrics such as total cost of ownership are useful to
measure the sourcing and supply management function in an organization. We discuss the metrics in a later section of
the chapter.
Irrespective of the type of organization, the ownership pattern, and the nature of items procured,
the procurement process consists of a standard set of activities that needs to be performed in a
chronological order. Figure 7.3 outlines the steps in the procurement process.
FIGURE 7.3 Steps in the procurement process
Step 1: Need identification. The first step in the process is need identification. Unless the users of the material express a
desire to consume more than what they have during a time or the inventory records show the potential depletion of
inventory, a procurement process never gets initiated. In the case of regular production items, inventory control policies
that are in vogue will trigger a procurement process. In the case of non-routine requirements such as the manufacturing
of a customized order, maintenance, and R&D requirements, users will usually raise a purchase request. The purchase
request will be the basis for procurement. The purchase request will have detailed specifications of the material required.
Based on the inventory control policy, appropriate order sizes are established.
Step 2: Vendor selection. Once the exact requirements are identified, the procurement process focuses on identifying an
appropriate source for supply. This usually begins with a search for potential vendors.5 The existing set of vendors
dealing with the organization is the usual starting point. However, additional sources could be identified thorough
professional contacts, trade directories, and B2B portals. The sources are contacted and are subjected to a request for
proposal (RFP) and request for quotation (RFQ) processes. Several alternatives are available for an organization to place
the order. These include placing a repeat order with an established vendor, selection through a tendering process, post-
tender negotiations for price and use of very recent methods such as reverse auctions through an electronic market.
These procedures vary markedly between government organizations and private-sector organizations.
Step 3: Order placement. After source selection and negotiations, the vendor is selected and certain administrative steps
are taken to place an order with the selected vendor. This includes fixing all the terms of the contract, such as price,
payment, and delivery terms, along with other engineering specifications, and obtaining the required authorizations for
spending. Purchase-order generation is inextricably linked to the authorization procedures in vogue in the organization.
For orders with very high monetary value, the decision making is done at the highest level in the organization.
Therefore, the purchase order is generated only after the competent authority has approved the purchase.
Step 4: Order receipt. Once the order is placed with the selected vendor, follow-up of the order is done to ensure that the
order is received as per the specifications and on time. Traditionally, organizations have spent much time on this step if
its suppliers are not reliable enough. Multiple communication channels such as phone, e-mails, faxes, personal visits,
and face-to-face meetings are held to follow up the purchase order. However, due to better sourcing and supply
management practices, many organizations have been able to cut time, cost, and deployment of organizational resources
in this stage of the procurement process. The reasons for this transition and the systems put in place in order to ensure
this constitute the modern supply management knowledge. We shall understand this in some detail in a later section in
the chapter.
Step 5: Order acceptance. The received material is subjected to some inward-goods inspection and a decision is taken to
either accept or reject it. Once a lot is accepted after inspection, the relevant inventory records are updated in the stores
and the material becomes available to the user. Therefore, this signifies the end of the procurement process for a
purchase request initiated at the beginning of the procurement process. In recent times, due to modern sourcing and
supply management practices, the inspection of incoming material is either eliminated altogether or minimized to a large
extent.
Although the steps in the procurement process are standard, there are several procedural
variations with respect to purchasing in every organization. These are largely dictated by the
purchasing policies in vogue in the organization. The simplest purchasing procedures pertain to
off-the-shelf buying, oral orders over phone, and placing repeat orders with an existing supplier.
In all these cases, the crucial steps of identifying an appropriate source of supply and price
negotiation are skipped. This results in large reductions in lead time and simplicity in operation.
Typically, low-value items are procured in this manner.
Although the steps in the procurement process are standard, there are several procedural variations with respect
to purchasing in every organization.
Electronic markets for reverse auctions and JIT purchasing are two recent methods adopted for procurement.
Procurement lead time is made up of three components, as shown in Figure 7.3. First, there is
an internal lead time due to the various administrative processes involved in selecting a vendor
and placing an order with the vendor. Once the order is placed, the lead time taken by the
supplier constitutes the external component of the procurement lead time. Finally, a set of
administrative activities is performed within an organization to accept the order, which relates to
the third component of procurement lead time. Although not evident from the figure and the
description, process complexities can significantly influence the lead time. If an organization has
a very complex process for order placement, then one could expect a very high procurement lead
time. Public-sector organizations sometimes suffer from such procedural complexities, leading to
very high procurement lead time.7
JIT purchasing philosophy suggests that by developing a good relationship with suppliers, it is
possible to simplify the purchasing process in an organization and thereby reduce cost and lead
time.8 Traditional supply management philosophy will not help an organization develop
suppliers who can deliver today’s requirements. On the contrary, there are several other
advantages for an organization in developing exclusive sources. In order to appreciate this point,
we shall first take a look at how traditional methods and the recent methods of supply
management differ. The traditional approach to supply management is best described as multiple
sourcing. On the other hand, the recent approach to supply management can be referred to as
single sourcing. These approaches are based on fundamentally different assumptions. Figure 7.4
illustrates the important differences between the two approaches.
The multiple-sourcing model, as the name implies, deals with multiple suppliers every time an
item is procured and enjoys some gains from the competitive scenario that a supplier faces in the
process. Moreover, information about future business is not shared as there is no requirement for
it. There is no guarantee that the same supplier will win the procurement contract in the future. In
a nutshell, buying organizations hope to benefit from the weakened position of the suppliers and
are confident of obtaining lower prices. It was true to a certain extent that the buyers did obtain
lower prices for the items procured in this manner. However, the cost associated with obtaining
the items as per specifications and in time was often high. Excessive follow-up was required.
Moreover, suppliers performed miserably on other parameters such as quality, dependability
during emergencies, quick response, and development of new technology and capabilities. These
issues were simply not captured in the price-negotiation process, so the suppliers had no
motivation to deliver on these parameters.
FIGURE 7.4 Multiple- and single-sourcing approaches to supply management
In contrast to the multiple-sourcing model, the single-sourcing model is based on the premise
that by developing a long-term collaborative relationship based on mutual trust and
dependability, it will be possible to obtain not only lower prices but also several other benefits
that are missed out in a multiple-sourcing model. This cooperative relationship includes sharing
vital business information pertaining to production plans, future contracts, and cost information
so that both the supplier and the buyer can gain substantially from business certainties. Suppliers
will be in a position to plan their capacity additions and technology and skill upgradation in
anticipation of future business from buyers. Buyers, on the other hand, benefit from lower costs,
better technology, quick-response capabilities, dependability in case of emergencies, and the
long-term strategic engagement of the suppliers in matters such as new-product development.
The traditional approach to supply management is known as multiple sourcing, and the modern approach to
supply management can be referred to as single sourcing. The single-sourcing model is built over the premise of
mutual trust.
Single sourcing, the Japanese prescription for supply management, has attracted considerable
attention in recent years, both in practice and in theory. Japanese manufacturers claim that this
new arrangement makes business sense. They report several benefits from single sourcing,
including a substantial reduction in inventory, reduction in planning and procurement costs,
continuous cost reduction, and faster new-product introduction. Manufacturing organizations in
other Western countries such as the United States and the United Kingdom have followed suit
and claimed similar benefits. In India, major automobile and component manufacturers have
adopted the single-sourcing model successfully.
One of the concerns that managers have about single sourcing is that when there is no
competition among suppliers, it is difficult to fetch a lower price. On the contrary, the price-
fixing mechanism in single sourcing is more rational and is based on a “win–win” proposition. In
a single-sourcing agreement, the focus actually shifts to cost. The buyer and the supplier will
freely exchange information and arrive at the cost of making the item through negotiation. Once
the cost structure is arrived at, the required margin for the supplier is added to it to arrive at the
final price. Such an arrangement is in the interest of both parties. Suppliers show greater interest
because their margins are built into the price. Buyers show interest because this new arrangement
makes it easy to engage in cost-reduction exercises. For example, once the cost structure and the
margin for the supplier are agreed upon, both parties can engage in cost-reduction exercises
jointly. The reduced cost gets passed on to the buyer, while the margin that is required for the
supplier to conduct business is preserved. The cost structure and the required margins could be
reviewed once per annum. During the intervening period, both parties can engage in joint cost-
reduction exercises.
Single sourcing, the Japanese prescription for supply management, has attracted considerable attention in recent
years, both in practice and in theory.
Another approach to solve this issue is factoring the geographical proximity of suppliers in
prioritising single-sourcing initiatives. If the suppliers are geographically closer, many of the
problems we have discussed could be easily sorted out. Perhaps, this explains why automobile
manufacturers develop suppliers in a 50-km belt around their place of operation. Moreover, there
are other benefits from the geographical proximity of suppliers. Earlier studies reported that as
the distance between suppliers’ and automakers’ plants decreased, automakers’ inventories as a
percentage of sales also decreased.10
Despite these benefits and success stories, adopting a single-sourcing model has not been easy
for organizations. In some sense, it amounts to putting all one’s eggs in one basket, which is
perceived to be risky. Jumping over to single-sourcing agreements overnight will create havoc in
business. The single-sourcing model is built on the premise of mutual trust. Establishing trust
between the two partners invariably takes time. However, organizations can approach the single-
sourcing model using a two-pronged initiative. One is to reduce the number of suppliers from
many to two immediately and the other is to set up the required mechanisms for developing
reliable vendors. We shall focus on the latter aspect in some detail in the next csection.
Despite several benefits and successful stories, adopting a single-sourcing model has not been easy for
organizations. The single-sourcing model is built on the premise of mutual trust.
The examples in Ideas at Work 7.3 show that there are no hard and fast rules for developing an
atmosphere of trust. It helps greatly if the buyers take the initial steps in this direction. The extent
to which a buyer can develop trustworthiness in the minds of the suppliers is limited only by its
corporate mandate and the commitment and creativity on the part of managers involved in this
process.
Trusting Suppliers?
The most difficult aspect of developing good suppliers is the building of trust between the
buyer and the supplier. Unfortunately, there are no certification procedures for
“trustworthiness”. Neither the buyer nor the supplier can begin to trust the other overnight.
Trust-building takes time, but the essential question is—how do the two parties begin to
develop trust? Unfortunately, this does not lend itself to a simple approach.
Organizations have taken recourse to a variety of means that might help the buyer and the
supplier develop mutual trust over a period of time. Often, these include the buyer
organization voluntarily taking steps to offer infrastructural and technical support to the
suppliers and addressing the payment-related issues of the supplier. Maruti Udyog Ltd, has
been continuously engaged in a variety of initiatives that would help the suppliers improve
their quality, technology, and business. For instance, Maruti imported machine tools under
the Export Promotion Capital Goods Scheme (EPCG), and gave them away to its suppliers.
This actually meant upgrading the machining capabilities of the suppliers, providing
infrastructural and technical support, and making an export commitment to the government
on behalf of the suppliers.
A Chennai-based company took several measures during the 1990s to speed up payments
to its suppliers. According to one executive, this alone was responsible for bringing down the
inventory investment by one-third. A leading computer manufacturer invested in all the
tooling required for the manufacture of tower cabinets. It went further and encouraged the
supplier to increase the production and sell it to its competitors as well.
Ranbaxy Limited took several initiatives that were much beyond the traditional supplier–
buyer relationship. These included improving the core competencies of the suppliers, sharing
several benefits, equity participation, financial assistance, and joint ventures for their
expansion. Some of their endeavours included conducting joint training programmes on
“customer–supplier relations” using reputed consultants, and actively participating in quality
improvement projects in vendor companies.
As a result of these initiatives, one supplier of the drug Alpha-D was able to increase the
purity level from 97.2 per cent to 98.5 per cent, as well as his business with other customers.
The company also helped another supplier of printed aluminium coils in changing its sealing
machines. In another instance, while solving a quality control problem at a sugar plant
belonging to one of its suppliers, Ranbaxy helped change certain piping sections by
providing a loan. The suppliers were highly motivated by these efforts of Ranbaxy.
All modern approaches to sourcing and supply management have a fundamental prerequisite,
namely, developing reliable suppliers. Therefore, every organization should invest in supplier
development and have methods by which it can assess the reliability of its suppliers.
Supplier Development
A good supplier development programme can be implemented in three stages. In each stage, a
set of activities are undertaken to identify the most appropriate source for the organization.
Figure 7.5 is an illustration of these steps. The first stage involves drawing up a list of potential
sources that could be considered for developing long-term relationships. In several organizations,
the search begins and ends with the existing set of suppliers with whom they have been dealing.
Analysing the volume of business and past history will reveal the suitability of considering some
of the suppliers. However, it is also not uncommon for organizations to enlist new vendors with
whom they have not dealt with earlier. In such cases, detailed questionnaires are sent to them and
some initial visits are made to assess the suitability of such suppliers. Trial orders are also given
to assess their capabilities, technical and otherwise. Bosch Limited categorizes its suppliers to
classify them and make various decisions with respect to their development. For instance, “X-
category suppliers” are those identified for elimination while “N-category suppliers” are new
ones who have been found to have the competitive capability to replace suppliers facing
elimination.
A good supplier development programme can be implemented in three stages: initial screening of potential
sources, coming closer to potential sources, and working towards signle-source self-certified deliveries.
During the second stage, the organization needs to take several steps to build trust and assess
the suitability of having long-term contracts in realistic terms. This is achieved by a series of
measures, as shown in Figure 7.5. Providing various types of assistance, encouraging mutual
frequent visits, and information exchange are some of the crucial steps in the process. Frequent
personal contacts and mutual visits by both suppliers and buyers will invariably result in faster
product development cycles and more reliable products. After some years of working with a
select set of suppliers, an organization would be reasonably confident of establishing much
closer and long-term relationships. Therefore, in the third stage, several measures are taken to put
checks and balances in place for the proposed single-sourcing arrangement. These include a
robust certification mechanism and a good vendor rating system. Once these mechanisms are in
place, it is possible for the organization to engage in a single-sourcing relationship with the
supplier.
The steps outlined in this section are broad and suggestive. Several organizations have adopted
a similar approach for vendor development. The differences are mainly in the time taken to
develop self-certified deliveries. While in some cases it takes as little as nine months, in several
other cases it could take two to three years to reach this stage. The time taken for vendor
development also depends on the complexity of the item for which a vendor is being developed.
The certification programme assesses, amongst other things, the financial and equipment
capabilities of the supplier under active consideration. Past balance sheets are analysed to
realistically assess if there is a possibility of the supplier going bankrupt. Similarly, the supplier’s
equipment capabilities, quality assurance systems, and propensity to engage in value analysis
efforts are also analysed. Other issues studied include understanding the cost structure and a
detailed scrutiny and finalization of the contract performance. The contract performance is an
elaborate exercise in which various contingencies likely to emerge are envisaged and the
expected behaviour on the part of both parties is spelt out. This includes, for instance, expected
performance in quality and delivery parameters and the time frame for corrective action to
restore deteriorating performance.
The certification programme assesses the financial and equipment capabilities, the cost structure, the contract
performance, the quality assurance system, and the value analysis effort of the supplier under active
consideration.
Once a supplier is subjected to the certification programme and is found to meet the
requirements, there will be no need for an inward goods inspection and the supplier will be
entirely responsible for quality and delivery performance. It is technically feasible to have direct
“online” sourcing of the items procured from the supplier. This will provide enormous savings in
procurement lead time and costs for the organization, which the multiple-sourcing model cannot
deliver.
Vendor Rating
Selecting the right vendor is a difficult but important part of the sourcing and supply
management process in any organization. Suppliers have varied strengths and weaknesses as
much as organizations have varied emphasis on different parameters pertaining to procurement.
For example, one organization may give greater importance to delivery reliability and another
may place greater emphasis on quality. Therefore, a systematic and careful method of assessment
of suppliers’ performance is required to enable an organization to monitor its suppliers’
performance and to align them with its requirements. Vendor rating systems are used for this
purpose in every organization. As suppliers make repeated deliveries of items in response to the
purchase orders that have been placed, they provide an opportunity for buyers to evaluate their
performance using the vendor rating system.
Vendor rating is a systematic method to evaluate suppliers’ performance using data from the delivery of items
in response to the purchase orders placed.
The first step in developing a vendor rating system is to establish the criteria to be used for
assessing the vendors. For example, an organization would be keen to understand if the suppliers
performed well in quality, delivery reliability, and dependability parameters. Although several
criteria could be used, the following criteria are used frequently: quality, price, service, technical
capability, financial strength, geographical location, reputation, and reciprocal arrangements.11
Further, when several criteria are used, it is also necessary to determine how far each criterion
influences the vendor rating. The relative importance of the factors is established, typically by
assigning weights to the factors such that the sum of all factor weights is 100, as illustrated in
Table 7.1.
In a weighted points plan, vendor rating is done using a two-step procedure. The first step is to
design the vendor rating system. As already discussed, it involves identifying the set of factors to
be considered, establishing the relative importance of each of these factors and the development
of yardsticks for each of these factors. Once the yardsticks are established, the performance of
individual vendors could be rated using the yardsticks and the weights for each factor. If a
vendor has excellent performance in all the factors, then it will score the maximum points in the
vendor rating. In reality, vendors score very high on some factors and low on others. Therefore,
one can compute the vendor rating as the ratio of the sum of the actual score that they have
obtained for all the factors to the maximum possible score.
TABLE 7.1 A Sample Vendor Rating System
Note: FOB means free on board.
In Table 7.1, five factors have been considered for vendor rating and the relative importance of
these manifests as weights for the factors. In this example, quality has 28 per cent weightage and
delivery reliability has 24 per cent weightage. For each factor, there is a 5-point scale used to
measure the performance of the vendors. Therefore, the maximum that a vendor can get is 500
points. In the lower part of the table, the actual performance of two vendors is compared. The
factor scores are arrived at on the basis of the yardstick for each factor. Each factor rating is
multiplied by the factor weight to get the factor score. For example, Vendor 1 gets an excellent
rating for quality. Therefore, his factor score for quality is 140 (28 × 5). The sum of all the factor
scores is given towards the end of the table and the ratio of this to 500 is the final vendor rating
(in percentage). From the table, we find that Vendor 1 has a higher rating (81 per cent) compared
to Vendor 2 (72 per cent).
The traditional purchasing method and approach was that supply management did not require
any measure other than the price of procurement for an item. The management accounting
system measured price variances using a standard costing system and purchasing managers were
evaluated on the basis of this information. This method of measuring performance based on just
the price of procurement had a serious drawback. Even when managers were getting an
apparently low price for the items that they procured, in reality the actual costs of eventually
owning the material was significant. Unfortunately, existing measurement systems totally
ignored this aspect. Moreover, as we have seen, the current approach to supply management
requires several measures other than simply assessing the price for how well the procurement
process is performing. These measures pertain to the actual procurement processes and the
outcomes of these processes rather than mere financial numbers. Some of the measures worth
considering are listed in Table 7.2.
Long-term partnership
Percentage of suppliers who have entered into long-
term contracts
Percentage of value of purchase under long-term
contracts
Rate of shrinkage of the supplier base
Number of certified deliveries
Total man-days spent in technical assistance for
supplier problem solving
Quantum of financial assistance to suppliers
The total cost of ownership is a comprehensive measure that factors, in addition to the price, other costs incurred
in eventually owning the material.
Involving suppliers in R&D efforts, reducing/ sharing of R&D costs, reducing processing costs, and co-developing
proprietary materials
Minimizing supplier costs by sharing long-term volume forecasts and jointly evaluating vendor capital expenditure
plans
Based on further analysis and corrective decisions identified in the study, the company
hoped to reduce the supply base from 68 to 10. The potential savings identified out of the
proposals made was to the tune of ₹6.898 million.
Source: P. B. Suriyanarayanan and K. Sivakumar, “Supplier Management Strategies for Enhancing Manufacturing Competence,” unpublished MPT
project report submitted at IIM Bangalore.
One of the frequently encountered issues in purchasing is the “make-or-buy” decision. In the
early period of the twentieth century, several firms were making all components in-house. The
Ford Motor Company was extracting iron ore at its own plant and fabricating all the components
required for assembling a passenger car all by itself. On the other hand, by 1980, Toyota was
buying over three-fourths of its requirement from suppliers and was focusing only on assembling
the components and integrating various sub-systems to produce a defect-free passenger car.
Clearly, every item that an organization requires in its manufacturing or service set-up could be
made either in-house or bought from a supplier. Therefore, the issue to resolve is what to make
in-house and what to buy from an external source.
In recent years, we are seeing a greater trend towards buying from external vendors and
focusing on certain core activities in-house. Several service businesses such as catering, event
management, software maintenance, maintenance of machine tools, billing and collection,
computer facilities management, fleet management, recruitment, front-office functions, and
several aspects of customer relationship management have all been shifted from in-house to a
third-party vendor. The growth of business process outsourcing (BPO) has been extremely rapid
over the last few years. Companies across the world are not only outsourcing some of their
peripheral and support business processes, but are also seriously looking at outsourcing
possibilities across the entire value chain. As of 2012, the Indian IT outsourcing industry
employed around 2.8 million people. Annual revenues were around USD 11 billion. The global
outsourcing market has grown from USD 111 billion in 2002 to USD 234 billion in 2005 and
USD 310 billion by 2008.13
In recent years, we have seen a greater trend towards more and more buying from external vendors and focusing
on certain core activities in-house.
The recent spurt in BPO signals the tendency of organizations (cutting across different sectors
of industries and services) to buy rather than make their requirements in-house. In such a
scenario, the importance of supply management further increases as suppliers’ performance is
likely to have an even greater impact on delivery and quality commitments made by an
organization to its customers. The trend towards outsourcing in several organizations is driven
primarily by four factors: cost, core versus non-core activities, management of capacity
expansion, and strategic restructuring.
The trend towards outsourcing in several organizations is driven primarily by four factors: cost, core versus non-
core activities, management of capacity expansion, and strategic restructuring.
Cost
The primary reason for outsourcing is that the cost of the product or service is much less if
procured from a third party. Several reasons are attributed to this. Some of them include the
benefits of large capacity investments made by the third party, the low volume of in-house
activity, and the need for continuous upgradation of the technology and process behind these
activities to obtain cost and productivity benefits. Consider the transportation of finished goods
to the market situated all over India. If an organization such as Voltas invested in its own fleet of
trucks to transport its air conditioners and refrigerators, it would suffer from all the drawbacks
that we identified. On the other hand, a trucking company will be ready to constantly invest in
new technology and improved processes. It will also have justifiable volume of business to
spread its cost over a wider set of activities. Another example is the emergence of software as a
service (SaaS) and cloud based software services.
Strategic Restructuring
Even when many of the opportunities do not exist for an organization, sometimes “make-or-buy”
decisions are taken as a part of a strategic restructuring exercise. During such an exercise, the top
management may contemplate some future expansions and market penetration strategies and in
response to these it may take a fresh look at the nature of activities to be performed. Core and
non-core activities are identified and plans drawn for outsourcing.
7.8 E-PROCUREMENT14
Local sourcing was a standard sourcing strategy for many years because the transaction costs of
sourcing across long distances were high. Furthermore, potential suppliers were seldom known
and contract enforcement was a problem. Transporting goods was costly and highly risky as
well-developed global logistics and supply networks did not exist. With the advent of modern
communications technologies and globally integrated markets as well as efficient transportation
services, global sourcing is now a reality. Multinational companies in particular are using global
sourcing strategies to overcome specific local factor shortages in their home countries and to
benefit from the availability of specialized skills elsewhere. They also benefit from factor cost
advantages.
All these have led to a new dimension in procurement—e-procurement. The growth of
electronic markets has dwarfed historical growth patterns of other sectors of the industry. Over
the years, several market mechanisms have sprung up in the electronic marketplace to tap new
value creation opportunities. Free markets introduced the idea of conducting reverse auctions
using the electronic media during 1995. Similarly, the big three auto players in the United States
(GM, Ford, and Chrysler) created COVISINT, a consortia-based market structure, in February
2000. Many others such as Oracle, i2, SAP and Ariba developed a variety of supply chain
collaboration tools. It is evident that there are several types of markets structures operating in the
B2B electronic market domain. Each B2B site typically operates multiple market structures. The
market structures fall under three distinctive categories: collaborative mechanisms, quasi-market
mechanisms, and neutral market mechanisms.
B2B market structures fall under three distinctive categories: collaborative mechanisms, quasi-market
mechanisms and neutral market mechanisms.
Quasi-market Mechanisms
In the second set of market structures, one or a small group of either the buyers or the sellers will
initiate the marketplace, host, and monitor, enrol market participants, and moderate the market
behaviour if required. In a buyer-centric marketplace, buyers take the initiative to host the market
and appropriate greater benefits than other market participants, just as the suppliers do in a
supplier-centric market. Two well-known variations of seller- and buyer-centric marketplaces are
forward and reverse auction sites, respectively. In the case of a forward auction a seller auctions
an item and several buyers will bid for the item. On the contrary, in a reverse auction, several
sellers bid to win a procurement contract from a buyer. A host of industrial procurement
contracts fall under reverse auctions.
In a quasi-market mechanism, the market structures have an inherent bias towards one of the market participants,
namely, the buyer or the supplier.
On the other hand, a few buyers or a few suppliers belonging to a sector of an industry could
create a consortium marketplace. Examples include www.covisint.com (automobiles) and
www.metaljunction.com (steel industry).
While e-procurement offers several new advantages, it also requires that organizations draw a blueprint on how
they will derive value by exploiting e-procurement opportunities.
While e-procurement offers several new advantages, it also requires that organizations draw a
blueprint of how they will derive value by exploiting the e-procurement opportunities thrown
open by the various electronic market structures. For certain products and services, purchase
managers will endeavour to build tighter integration through electronic market structures. For
instance, key suppliers and dealers will be encouraged to participate in a TPN. Enlisting trading
partners in the TPN will be a challenging task. Managers need to build incentives for
participation in such markets. These may include sharing vital commercial and technical
information, primarily through the network, sharing cost benefits accruing out of reduced
transaction costs, and hosting private auctions of excess and obsolete items in such marketplaces
even before tapping other traditional channels.
E-procurement necessitates restructuring of the logistics and supply chain network of an organization.
In certain other cases, purchase managers will discover more value in engaging existing
trading partners with whom they developed close relationships in more competitive mechanisms.
For instance, a supplier of custom components with low or transferable tooling may be asked to
participate in a reverse auction, where they will compete with a handful of other suppliers
carefully pre-selected through a robust RFQ process. In yet another class of items, managers will
see immediate value arising out of sourcing from competitive neutral markets. Some suppliers
may be replaced with more competitive sourcing decisions involving neutral electronic markets.
Electronic markets may also cause dis-intermediation of some traditional channels. For instance,
neutral auction sites may replace channels for the disposal of second-hand capital goods, and
surplus inventory. New opportunities will emerge in recasting the distribution network in several
industries. Resellers may be eliminated and the role of distributors and other channel partners
will be redefined.
Managers will find it necessary to look at procurement practices afresh. Restructuring the
logistics and supply chain network of an organization will become inevitable. However,
launching several pilot projects to tap these alternative market structures in the mid-term of about
two years will be a valuable step in reconfiguring their value chain.
SUMMARY
Important developments in operations management in the recent past have redefined the role of sourcing and supply
management in organizations. These include new quality management initiatives, changing nature of cost structures,
creating lean organizations, and quick-response requirements.
A typical procurement process consists of six steps: (i) need identification, (ii) vendor selection, (iii) order placement,
(iv) order receipt, (v) order acceptance, and (vi) vendor rating.
Traditionally, organizations utilized a multiple-sourcing model for vendor selection. However, in recent times,
organizations employ a single-sourcing model. The single-sourcing model provides several advantages over the
traditional model.
Supplier certification and vendor rating are important components of modern day supplier development and
management initiatives.
The “make-or-buy” decision is made on the basis of the criticality of the item under consideration, cost, capacity issues,
and strategic restructuring initiatives in an organization.
E-procurement offers several advantages to an organization such as reduced cost and lead time. However, it requires
that organizations develop a new blueprint for dealing with the existing suppliers as well as the new suppliers likely to
be discovered though the electronic markets.
REVIEW QUESTIONS
1. Why has sourcing and supply management assumed importance in today’s organizations?
2. How is procurement lead time determined? What factors contribute to the procurement lead time in an organization?
3. An organization has high investment in inventory. Although it is not clear what factors contribute to this, the organization
would like to know if the procurement lead time has any impact on inventory. Assume that you have been asked to conduct a
study on this issue, and prepare a one-page proposal for the organization, detailing the various aspects that your study would
cover.
4. What are the key differences between the traditional and contemporary approaches to supply management?
5. What benefits accrue to an organization through the single-sourcing model?
6. What are the steps involved in a supplier development programme?
7. What is a supplier certification programme? Why is it required?
8. Identify some measures of performance for the purchasing and supply management function. How would an organization
benefit in the long run by performing better on these measures?
9. What do you mean by “make or buy”? Can you give some examples for how organizations utilize this concept?
10. How is business process outsourcing related to purchasing and supply management practices?
11. What do you understand by the term e-procurement? Identify three reasons for an organization to opt for e-procurement.
CASE STUDY
TABLE 7.5 Ratio of Non-Moving Stock to Average Closing Stock of the Respective Sources
Note: *The non-FXB column includes all imported items other than FXB.
TABLE 7.6 Classification of Non-moving Items (Stock Value > ₹5,000) in Stores 00
Since FXB components form the bulk of the purchase, they also contribute to a large
percentage in the non-moving category. However, in the case of low-value items, i.e., items with
stock value less than ₹5,000, the local items also form a larger proportion of the non-moving
stock. In spite of being an imported category, in both the cases, the non-FXB components
(source code K) do not contribute substantially to the non-moving stock. This phenomenon
perhaps suggests that more attention need to be given to the nature of interaction with FXB.
Similar analysis has been carried out with respect to Stores 03. The product-wise and the
source-wise classification of the inventory point to more or less the same situation as it exists for
Stores 00. The FXB components contribute to about 46 per cent of the total non-moving
inventory in Stores 03. The details are available in Table 7.7.
Relatively lesser lead times of procurement for the nonFXB imported items
Ability to order variable (and sometimes smaller quantities) with non-FXB suppliers
Better interaction and coordination possible with nonFXB suppliers
Poor planning of procurement leading to excess ordering of all items and especially the indigenous items
Restrictions on minimum order quantity
Poor relationships with the local vendors leading to inability to reap the advantages that may accumulate due to
relatively short lead times of procurement
Sudden withdrawal of orders, sometimes, by the customers, leading to accumulation of non-moving inventory
Changes in technology and product obsolescence. The instruments manufactured/assembled and sold by BCLM were
initially analog. However, the changes in technology resulted in digital and Intelligent Automation (IA) systems. Such
changes in technology have always resulted in pile-up of non-moving inventory.
Wrong forecasting. Another source for non-moving inventory, according to a few executives, has been the overly
optimistic assumptions about the product performance leading to gross overestimation of the demand. However, when
the product failed to meet the expectations, the inventory piled up. The examples of the product BMOS seem to
substantiate this phenomenon. At the time of introduction, it was felt that BMOS was a premium product. Initially, 25
units were supplied to the customers and orders were placed to manufacture/assemble about 300 sets. However, the
high price and the inferior design of BMOS could not substitute the standard, universally known, low-priced and easy-
to-sell FXB 760 models. It appears, in this case, that very little feedback has been obtained from the marketing and
branch offices. Knowledge of these factors would have helped in avoiding the build-up of non-moving inventory.
Product failure in the market. When new products were introduced, the performance of some of the components has not
been satisfactory, leading to customer dissatisfaction. Changes made in such products have also contributed to pile-up
of non-moving inventory of older designs and versions. The example of BMOS fits into this category. Similarly, when
the plasma indicators were introduced, the customer feedback pointed out to the reliability problems. Hence, it was
decided to changeover to fluorescent display in 2002–03. Such sudden changeover resulted in a non-moving inventory
of about 100 sets. Currently, it is estimated that about 49 displays and about 60–65 kits may be in stock.
Withdrawal of products either by BCLM or FXB. BCLM has been regularly withdrawing various product groups for
several reasons. Some of them, such as change in technology and product failure in the market, have already been
mentioned. In addition, BCLM withdraws products because FXB withdraws them from their product line. Withdrawal
of products is not unusual to many manufacturing organizations. The frequency of withdrawals in industries such as
BCLM may be high. Under such conditions, it is important to develop certain systems and procedures with respect to
withdrawal of the products. At the time of the study, enough evidence of such systems and procedures was not found.
For example, it was given to understand that FXB announces its withdrawal plans about six months in advance. At
BCLM, information on such announcements and the action taken on these was not available. Table 7.8 has a list of
products withdrawn by BCLM in the recent past.
The observations made by various executives during the discussions indeed confirmed the
analysis and conclusions drawn based on the database of non-moving items. In addition, a few
significant observations were made with respect to the build-up of non-moving inventory. All
these point to the fact that a few areas need to be strengthened in order to prevent the build-up of
non-moving inventory.
SUGGESTED READINGS
F. Ahmed, “Managing Vendors,” Business Today, 7–21 March 1995, pp. 132–136.
M. Bensaou, “Portfolios of Buyer–Supplier Relationships,” Sloan Management Review 40, no. 4 (1999): 35–44.
V. Chhachhi, “Make or Buy?” Business Today, 7–21 April 1996, pp. 60–69.
M. A. Cusumano and A. Takeishi, “Supplier Relations and Management: A Survey of Japanese Transplant and US
Auto Plants,” Strategic Management Journal 12, no. 8 (1999): 563–588.
G. W. Dickenson, “An Analysis of Vendor Selection Systems and Decisions,” Journal of Purchasing 2 (1966): 5–17.
J. H. Dyer, “Dedicated Assets: Japan’s Manufacturing Edge,” Harvard Business Review 72, no. 6 (1994): 174–178.
J. H. Dyer, “How Chrysler Created an American Keiretsu,” Harvard Business Review, 74, no. 4 (1996): 42–56.
L. Ellram, “Purchasing: The Cornerstone of the Total Cost of Ownership Concept,” Journal of Business Logistics 14,
no. 1 (1993): 163–184.
C. Fill and E. Visser, “The Outsourcing Dilemma: A Composite Approach to Make or Buy Decision,” Decision 38, no.
1 (2000): 43–50.
J. L. Hartley, B. M. Grear and S. Park, “Chrysler Leverages its Suppliers’ Improvement Suggestions,” Interfaces 32,
no. 4 (2002): 20–27.
J. Hazra, “Supply-chain Management in the Internet Era,” IIMB Management Review 16, no. 4 (2004): 54–74.
P. D. Larson and J. D. Kulchitsky, “Single Sourcing and Supplier Certification: Performance and Relationship
Implications,” Industrial Marketing Management 27 (1998): 73–81.
J. P. MacDuffie and S. Helper, “Creating Lean Suppliers: Diffusing Lean Production Techniques Through the Supply
Chain,” California Management Review 39, no. 4 (1997): 118–151.
B. Mahadevan, “Enhancing Manufacturing Competence Through Supplier Co-partnership,” in The New Manufacturing
Architecture (New Delhi: Tata McGraw-Hill, 1999), pp. 133–134.
B. Mahadevan, “Making Sense of the Emerging Market Structures in Business-to-Business E-Commerce,” California
Management Review 45, no. 3 (2003): 86–100.
T. Minahan, “Is This the Future of Purchasing?” Purchasing, 12 March 1998, pp. 42.
T. Nishiguchi, “The Toyota Group and the Aisin Fire,” Sloan Management Review 40, no. 1 (1998): 49–59.
E. W. Schuster and S. J. Allen, “Raw Material Management at Welch’s Inc.,” Interfaces 28, no. 5 (1998): 13–24.
J. P. Womack, D. Jones and D. Roos, The Machine That Changed The World (New York: Rawson Associates, 1990).
S. Yahya, and B. Kingsman, “Vendor Rating for an Entrepreneur Development Programme: a Case Study Using the
Analytical Hierarchy Process Method,” Journal of the Operational Research Society 50 (1999): 916–930.
Part III
Designing Operations
CHAPTER 8
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Process mapping is a tool to understand the various steps involved in a business process. Data about the process can be
collected through brainstorming among the various players connected to the process. Employees working in an area of the
process discuss and develop a representation of the actual process on the basis of their working knowledge.
Source: stock.xchng
One of the important decision points of an operations system concerns the capacity to be
deployed in the system. We often hear about excessive delays and waiting time in service
systems such as a teller counter in a bank. Similarly, we hear that some factories work with near-
100 per cent utilization of their resources. All these pertain to certain decisions taken with
respect to the nature and amount of capacity that an operating unit builds into the system and the
manner in which the existing capacity is put to use. Improper choice in the amount of capacity
deployed and poor planning of the existing capacity will lead to loss of productivity and overall
reduction in the profitability of the operating system. Therefore, a set of tools and techniques are
required to address these issues. In this chapter, we shall look at various issues related to capacity
planning. The basic building block of capacity analysis is process analysis. After all, the process
design determines the capacity of a system. Therefore, we shall begin with an understanding of
processes before we address the issue of capacity planning.
Understanding any operations systems invariably begins with an understanding of the processes
that are integral to the operations system. A process is the basic building block of operations. It
consists of a set of activities that need to be performed by consuming some resources and time.
This eventually determines the performance of the operations systems in terms of cost, time,
productivity, profit, etc. A few examples will help us understand this better.
Consider a fast-food joint such as Hotel Saravana Bhavan in Chennai. When the customer
goes to the counter and places an order for the food items he or she would like to eat, the
outcome is determined by the process employed to satisfy this customer demand. The steps
involved in serving the demand, the number of people involved, the nature of resources
consumed and the time taken to serve this customer will all depend on the process design. The
process design will in turn determine aspects such as the time taken to serve the customer, the
cost involved, the productivity of the people and the utilization of the resources. These are the
performance characteristics of the process in question. Viewed from this perspective, the
performance of the processes finally dictates the performance of an operations system itself.
Consider an automobile garage that deals with all aspects of the repair of wheels and tyres
(such as wheel alignment, wheel balancing, tyre and tube replacements, and repairing of
punctured tires). Here, too, one will encounter a set of issues similar to those at Saravana
Bhavan. The design of the process will determine the performance of the automobile garage.
Once can extend this logic to larger manufacturing and service systems.
An operations manager may want to understand some of the following issues in the process
design:
Do I have adequate resources to meet the demand? If I need to add some extra resources, where should I add them?
What is the extent of the utilization of my resources?
If I need to increase the capacity of my system, how should I modify the process? Should I add some more resources?
What will happen to the cost of my operations?
One can find the answers to these questions through process analysis. By process analysis, we
refer to the use of some analytical mechanisms to understand the impact of the process design on
output, cost, or any other performance metric. It also amounts to understanding the impact of
alternative process configurations on the chosen performance metric.
The design of a process is a matter of detail. Every activity that constitutes a process must be
identified. It is also important to know the time taken for each activity and the nature of flow of
materials/information in the process. A pictorial representation of all this information could be
developed using process flow-charting. Process flow-charting employs a set of standard symbols
and graphical tools to represent all the information about the process. The symbols used are as
follows:
A step in the process
Transportation (movement)
Storage or inventory
Figure 8.1 is a simplified representation of a process flow chart for shirt manufacturing, which
consists of five steps in the shirt-manufacturing process and three storage points. The store
stocks raw cloth and other materials such as zippers, buttons, and thread. The cloth is moved to
the first stage of processing where it is systematically spread out and relevant markings are
made. After this, it is sent for cutting. Once the cutting operation is complete, various individual
pieces that make up one shirt when put together are segregated into two bundles. One bundle
goes to the first stitching operation while the other bundle goes to the second stitching operation.
After the stitching is complete, they are temporarily stored. The assembly process consists of
matching the pieces and finally attaching them into a completed shirt. After this, the shirt is
inspected, pressed, packed, and sent to the stores for final dispatch.
A process flow chart helps an operations manager in many ways. Some of these are as
follows:
a. It provides a pictorial and compact representation of the process and enables a quicker and better understanding of the
various aspects of the process.
b. By superimposing additional information such as the time taken for each stage of the process, certain useful measures
can be computed. For instance, one may be able to estimate the time required to complete the process. Using this
information, one can also identify the bottlenecks in the process and the productive capacity of the process.
Furthermore, one can also assess the amount of resources that need to be deployed at each stage in order to meet the
projected requirements.
FIGURE 8.1 A simplified process flow chart for shirt manufacturing
These aspects are what constitute process analysis. We shall see in some detail how this
process analysis is carried out.
Process and capacity analysis has a closer relationship to the planning premise that the system
uses. For example, for a turnkey manufacturer such as Bharat Heavy Electricals Limited (BHEL)
or ABB, a question such as “how many power plants can I erect in a planning horizon of, say, a
week or a month?” is not relevant. On the other hand, for a volume manufacturer such as Kinetic
Motors, such a question will be of great relevance. Therefore, before embarking on an exercise
of process analysis, we need to have some understanding of the type of planning premise that an
operations system is utilizing. Three generic planning premises are in use in operations
management: make to stock, make to order, and assemble to order.
From our description of the MTS planning methodology, it is clear that such systems are more
amenable to organizations with fewer product varieties and high production volume, as in the
case of continuous and streamlined flow systems. In such a situation, the demand is likely to be
continuous and substantial, so that production can proceed on the basis of a few assumptions
about the nature of the demand. In the case of demand being higher or lower than what was
estimated for a particular period, the stockpile of finished goods will absorb the fluctuations, and
organizations do not run the risk of carrying excess inventory for a long time.
From a process and capacity analysis perspective, MTS is suitable for mass production
systems. Therefore, the relevant questions for process and capacity analysis are:
a. What is the productive capacity of my process per shift or per day?
b. Where is the bottleneck for this process? If I need to increase the capacity of the process, at which stage of the process
should I invest in capacity expansion?
c. Can I improve the productive capacity by means other than investing in new resources?
Another method of planning for MTS is using “takt time” for scheduling products at various stages in the
production system.
Traditionally, organizations use a variety of production planning and control tools for MTS
systems. These systems compute the materials and capacities required for production and
schedule various operations in the manufacturing system to produce as per plan. One of the
recent approaches in implementing an MTS-based planning system is to use a pull-based
scheduling system. Another method of planning for MTS is using “takt time” for scheduling
products at various stages in the production system. We shall discuss these in detail in Chapters
13 and 18, under just-in-time (JIT) manufacturing and scheduling of operations.
It is clear from the previous discussions that the performance metrics for process analysis are
different for MTS and MTO systems. We shall look at some relevant performance measures for
these systems.
Throughput time: Throughput time (TPUT) is the elapsed time from the first stage of the process to the last stage of the
process. It is also known as lead time. In the case of a new customized print order that a printing press is executing, if
the order was launched into the system on Monday morning at 8.00 a.m. and the job was completed by Tuesday evening
at 4.00 p.m., then the throughput time for this order is 32 hours.
Cycle time: Cycle time is the elapsed time between two successive outputs from a process that is continuously operating
in a given period of time. For instance, in a bread-making process, if a loaf of baked bread comes out of the system
every 20 seconds, then the cycle time for the process is 20 seconds. In other words, in a steady-state operation of the
process, three loaves of bread will come out every minute. If the process is operational for 450 minutes (7½ hours) then
the productive capacity of the process is 1,350 loaves daily.
Bottleneck: That stage of the process that dictates the output of a process is the bottleneck. Let us assume that in the
bread-making example, the baking of the bread in the oven takes 20 seconds and all other processes take less than 20
seconds. In this case, the baking process is the bottleneck. The processing time at the bottleneck is the cycle time for the
process.
Throughput time is a relevant measure for MTO systems. Measures such as cycle time and
bottlenecks, on the other hand, are relevant in the case of MTS systems.
At an overall level, the assumptions in Example 8.1 could be considered for the purpose of
capacity computations. However, in reality, several of these assumptions are to be checked and
necessary modifications should be made. When there are statistical fluctuations, these could be
addressed at the scheduling stage. We have discussed these issues in Chapter 18 (see
synchronous manufacturing in Chapter 18 for more details).
EXAMPLE 8.1
A toy manufacturer receives crafted toys from local carpenters and performs the final
operations before stocking it for sale. The process consists of five steps. The first step is to
arrange a set of four toys in a pallet. After this, the pallet moves to the next station where the
toys are pre-treated. This is to increase the toys’ life, prevent them from attacks of pests in
the future, and also to improve the effectiveness of the painting operation. The next step is to
send it to the spray-painting chamber, where it is painted as per the specifications. At present,
there is one spray-painting machine. After painting, it is left in an open area for drying. The
painting process and the pre-treatment process are specialized so the paint dries quickly.
Finally, the toys are inspected and packed. The steps in the process, along with the relevant
details, are as follows:
Solution
The process flow diagram is presented in Figure 8.2.
FIGURE 8.2 The process flow diagram
a. The throughput time for the process is the sum of all processing times. In this example, the throughput time is 55
minutes. This implies that if all the required resources are available, then from the time the job is launched at the
first step, a pallet consisting of four toys will come out of the system after 55 minutes.
b. The bottleneck is that stage of the process that dictates the output of the process. In our example, the spray painting
is the bottleneck.
c. The cycle time is determined by the process time at the bottleneck station in the process. In this example, cycle
time is 20 minutes. The implication of this is that when the process operates in a continuous manner, then one can
expect a pallet of finished toys to come out every 20 minutes. This is shown in Figure 8.3.
d. In order to compute the productive capacity of the process, we shall compute the production rate at each stage of
the process. This is illustrated in Figure 8.4. Stage 2 has a processing time of 12 minutes. It means that the
productive capacity is 5 per hour. Similarly, the painting stage has a productive capacity of 3 per hour. It does not
make sense to compute the productive capacity for the drying operation as it does not use any constrained resource.
For practical purposes, one can stack as many pallets as one wants to on the floor for drying. Based on this
computation, we conclude that the manufacturer can produce three pallets (12 toys) per hour. If the manufacturer
works with an 8-hour operation, he/she could produce up to 24 pallets per day.
e. These computations are based on certain assumptions about the process. These include the following:
The process is a continuous operation (that is, it operates at a steady state with no start-up and shutdown issues).
An adequate amount of resources (including material, men, machines, and other resources) is available.
Once the operation starts, there are no delays and breakdowns in the system.
There are no variations and statistical fluctuations (so that the time estimates are fixed and constant).
The process produces a standard set of items. Therefore, there are no additional time losses due to set-up and
changeover other than those already factored into the time estimates.
FIGURE 8.3 Graphical representation of the process with start and end times
EXAMPLE 8.2
Consider Example 8.1. Suppose the preparation process has two parts to it. The first part is
setting up of pallets, which requires 4 minutes, and the second part is the actual time of
loading the pallet with toys. Each pallet consisting of four toys requires 4 minutes to load.
Furthermore, the painting booth can hold up to three pallets during the spraying operation.
There are enough number of pallets available in the system. Given this additional
information, examine the capacity of the system under the following conditions:
a. The revised process will paint two pallets at a time.
b. The revised process will paint three pallets at a time.
c. There are two pre-treatment units available.
What are the main inferences that one can draw from this exercise?
Solution
Effect of Batch Size
The additional batch that one can process affects the preparation, pre-treatment, inspection,
and pack stages of the process. If we process two pallets, then the preparation time will be
the total time taken to setup and load toys or [4 minutes + (2 × 4) minutes] = 12 minutes.
Therefore, every 12 minutes two pallets of toys will move out of this stage of process. This
will translate into 10 pallets per hour. The pre-treatment stage will continue to be five pallets
per hour and the inspection and pack stage will also continue to be 12 pal-lets per hour. In the
case of painting, in 20 minutes, both the pallets will be painted. This means two pallets will
come out every 20 minutes, there by indicating a capacity of six pallets per hour.
In the case of three pallets, in the first stage, it takes 16 minutes [setup time + toy loading
time = 4 minutes + (3 × 4) minutes] to prepare three pallets. This translates into a capacity of
11.25 pallets per hour. Table 8.1 shows the capacity details for two pallet and three pallet
batch sizes, in comparison with one pallet size.
When the batch size is increased to two or three pallets, the output increases from three
pallets per hour to five pallets per hour as the bottleneck shifts from the spraying operation to
the pre-treatment operation. Figure 8.5 illustrates these alternative scenarios.
Some Inferences
From this analysis, we notice that batch sizes play a crucial role in determining the bottleneck
of a process. Furthermore, as several choices are made with respect to the resources and the
batch size, the bottleneck shifts from one stage of the process to another. This is referred to
as a wandering bottleneck. In our example, addressing the pre-treatment process is crucial to
improving the productivity of the process. Before adjusting the batch sizes and deploying
additional resources at painting, improving the pre-treatment process becomes crucial.
FIGURE 8.5 Production capacity under varying batch sizes
As we have seen in Example 8.2, a process analysis exercise often leads to the issue of
improving the process. For instance, in this example, the issue we may want to address is how
the process can be improved. Can we increase the production without much investment in the
capacity? Can we reduce the process times? Can we redesign the process such that the bottleneck
station can increase its output? How can we increase the overall utilization of the stations? One
way to improve the process is to understand the existing process in detail and to explore
alternatives to simplify it. Let us look at some well-known tools for improving and redesigning a
process.
Business process re-engineering (BPR) is the fundamental rethinking and radical redesigning
of business processes to achieve dramatic improvements in critical measures such as cost,
quality, service and speed. Hammer and Champy, in their book Reengineering the Corporation,
have laid out the broader framework of BPR.1 According to them, the scope for improvement
through BPR is evident in an organization if any of the following are happening:
Business process re-engineering (BPR) is the fundamental rethinking and radical redesigning of business
processes to achieve dramatic improvements in critical measures such as cost, quality, service and speed.
Extensive information exchange, re-keying of data, and data redundancy, pointing to arbitrary fragmentation of a natural
process.
Large investment in inventory, buffers, and other assets, indicating that the existing process is unable to cope with
uncertainty.
High proportion of checking, control, and progress monitoring, implying excessive fragmentation of the process.
Inadequate feedback in the process resulting in excessive rework and iteration.
Increase in complexity of operations, exceptions, and special cases, also indicating the need for simplifying the process
by revisiting the process fundamentals.
There are alternative frameworks available for undertaking a BPR exercise.2 Despite some
differences in the number of steps involved, all of them fundamentally involve a three-stage
process, the understand–simplify–automate (USA) approach to process improvement. Clearly,
the first step is to develop a sound understanding of the process. Process mapping is a very
valuable tool to develop this understanding as it provides a detailed snapshot of how the process
is currently performed. Once the process is understood, the next step is to simplify the process.
Non-value-added (NVA) analysis is a valuable tool for this purpose as it helps the improvement
team question the relevance of each step identified by the process mapping exercise.
Once the relevance of some of the activities is questioned, it enables the improvement team to
look for opportunities for eliminating them altogether, or combining them with some other
activity. In some cases, the process may be simplified if the activities are rearranged. In reality,
firms develop what is known as the “as-is” version of the process usingprocess mapping.
Furthermore, after some NVA analysis they develop a “should-be” version of the process. The
final step is to automate the process and reap further efficiency and productivity gains from the
new process by using IT tools. The use of a host of Web-based workflow packages, electronic
data interchange (EDI), and enterprise resource planning (ERP) are some of the common
automation tasks done as a part of the BPR exercise. Figure 8.8 depicts the three-stage
methodology for BPR as described here.
The BPR methodology enables an organization to make significant improvements in
processes, thereby leading to better performance as certain implementation guidelines are
followed in a BPR exercise. These guidelines are briefly explained here:
Shifting the focus of the entire exercise from activities to results: What this means is that the basis for organizational
structure is outcomes and not tasks. Therefore, the organizational structure will shift from functional departments to
process teams.
Performing work where it makes the most sense: The key implication of this is that those who use the output of the
process should perform work. In such a scenario, they will be able to control the process and make the required
improvements in it. The accountability of the process team towards the outcome of the process will also improve.
Creating flat organizational structures: Poor process performance is due to multi-layered (deep) organizational structures
that deny the process teams an opportunity to own the process and have the required mandate and control to improve the
process. Moreover, too many “change of hands” take place and eventually nobody takes ownership for the process
performance and outcomes. Instead, whenever problems crop up, they end up pointing fingers at one another. All these
issues are eliminated in a flat organizational structure.
Capturing information once at source: Advances in Web-based tools should be exploited in any process re-engineering
exercise. In an era of shared databases and client server architecture, it is important to capture information only once and
store it for multiple uses. Capturing the same information at multiple points by multiple entities creates unnecessary
complexities due to data integrity problems and makes the processes less efficient. It also introduces unnecessary delays
and requires commitment of more resources to sort out the mess.
Process mapping, is a tool to understand the various steps involved in performing a business
process. The basic premise behind process mapping is that in every organization there are several
business processes and these processes consume resources and time, finally driving the cost.
Therefore, improvements in cost, lead time and resource utilization essentially boil down to
understanding how processes are performed and identifying unnecessary steps in the process for
possible elimination3. In a process mapping exercise, data about the process is collected either
from available records, an ERP system, or by brainstorming among the various players
connected to the process. In the brainstorming approach, the employees working in an area of the
process are asked to discuss and develop a representation of the actual process on the basis of
their working knowledge. It is possible to construct the complete process steps in this manner
over, say, a three-hour session.
Process mapping, is a tool to understand the various steps involved in performing a business process.
In a process mapping exercise, the data is collected on the extent of usage of manpower and other resources.
Process mapping is the starting point for any improvement effort in an organization as it
equips the improvement team with a wealth of data. While it details how the process is currently
performed, it does not indicate whether the various steps are required. Therefore, the next step in
the process is to categorize the various activities into alternative heads and analyse their
relevance. As explained earlier, NVA analysis is a method by which the relevance of some of
the existing activities is questioned. Various categories of activities used in NVA analysis are
discussed here.
NVA analysis is a method by which the relevance of some of the existing activities is questioned.
Every operating system uses a variety of resources including labour, machines, tools, and
fixtures. All these resources are available in fixed quantities for use by the system (unlike
materials, availability does not dwindle by the use of these resources). In this sense, capacity is a
fixed investment for repetitive use by the system, until it requires replacement on account of
depreciation or wear and tear. Capacity denotes, in general, the extent of availability of these
resources for use by various processes. It also denotes the maximum output of products and
services one can achieve by using these resources. For example, let us consider a machine
capable of stamping circular washers that are used in the final assembly of a product. The
capacity of the machine can be expressed in one of the following two ways: the number of
machine-hours or the number of washers that could be produced.
One can say that in a month, 200 machine-hours are available for processing. Alternatively, if
we assume that one can stamp 2000 washers per hour, then the capacity of the machine can be
expressed as 400,000 (20032000) washers per month. The former definition is useful when we
analyse capacity issues in a localized fashion. On the other hand, at an aggregate level (such as a
factory or a division) it will be useful to measure capacity in output terms. Thus, at the Nasik
factory of ABB (which manufactures circuit breakers), the firm will compute capacity in terms of
the number of circuit breakers (or cubicles/panels) that it can produce per month. On the other
hand, at the fabrication plant, it will compute the capacity of the welding machine and hydraulic
press in terms of machine-hours.
The notion of capacity is very similar in service industries also, For example, a financial
services firm may measure their capacity in terms of the number of loan applications processed
per day. A management consulting or an accounting firm may measure its capacity by the
number of man hours of consultant time available. A hospital will also measure surgeons
capacity in a similar fashion.
Irrespective of the method of defining capacity and measuring it, there are certain important
issues associated with capacity that an operations manager needs to understand. Capacity has a
significant impact on the cost of operation of the system. Capacity investments are large and
fixed in nature. Therefore, they manifest as fixed costs of the system. Since the total cost can
come down by using a fixed investment more, firms can experience cost benefits from using the
available capacity to produce more by judicious planning. A well-known economic principle,
economies of scale, indicates the relationship between cost and capacity in an operating system.
When the output increases in an operating system, the system is likely to experience cost
advantages on account of several factors. These include spreading of the fixed costs of capacity
over a larger number of units, improvements in the utilization of several resources in the system,
cost benefits in procurement on account of increased volumes, and efficient use of supervisory
and managerial resources and other common resources in the system. Due to these, the average
unit cost of the system begins to fall. However, as the output increases, the marginal returns from
these gains begin to reduce such that the average total unit cost begins to increase beyond a
certain point. Figure 8.9 is a graphical representation of this effect.4 Knowledge of these costs
and benefits helps an organization to choose appropriate levels of capacity deployment and
output rates.
The second aspect of capacity and cost is the learning curve effect. When a new technology or
process is deployed, the operating personnel are not conversant with the best operating
conditions. Moreover, organizational policies and standard operating procedures would not have
been established. Therefore, in the initial stages, the productivity levels are low and the capacity
tends to be underutilized. The average unit costs of operations tend to be high. However, over
time, learning takes place and capacity utilization improves. The output in the system also
increases, leading to lower unit costs. Operations managers should be aware of these issues and
understand the need to accelerate the learning effects and benefit from using capacity effectively.
FIGURE 8.9 An illustration of economies of scale
The decision regarding how much capacity is required is not addressed just once by an
organization. Firms need to revisit the issue of capacity from time to time, in response to
emerging market scenarios. The company may have decided on a market expansion strategy and
therefore may have to understand whether adequate capacity is available in line with the strategy.
In other cases, in response to a competitor threat, a company may decide to invest in more
capacity to pre-empt the entry of new players. Government policy changes and changes in the
regulatory framework may also induce a company to revisit the capacity issue. A case in point is
the Indian textiles industry. In response to the dismantling of quotas worldwide from 1 January
2005, several firms are currently expanding their capacity. According to the Centre for
Monitoring Indian Economy (CMIE), the value of new investments in the textile sector has gone
up by 35 per cent as on 30 April 2004.
In a proactive mode of capacity build-up, capacity is added in anticipation of the future. In a reactive mode of
capacity build-up, capacity is added in demand.
Irrespective of the context for capacity expansion, capacity is generally added in large chunks.
It is not possible to add capacity in small increments at frequent intervals. This is because
capacity additions happen in certain integral units and invariably call for larger capital outlay and
detailed capital budgeting exercises. Further, they also involve longer lead time to source and
install in the system. Therefore, capacity decisions are made at less frequent intervals.
Consequently, firms go through a cycle of overcapacity–appropriate levels of capacity– under-
capacity. This is a typical mode of capacity build-up.
In the reactive mode of capacity build-up, firms wait until demand materializes sufficiently before augmenting
capacity.
There are two other alternative methods of capacity build-up available for firms. In the
reactive mode of capacity build-up, firms may want to wait until demand materializes
sufficiently before they augment capacity into the system. In contrast, in a proactive mode of
capacity build-up, capacity is added in anticipation of future demand. In a highly competitive
market where temporary loss of sales directly translates into loss of market share, firms may
have to plan for a proactive mode of capacity build-up. However, it calls for high accuracy in
forecasting future demand. On the other hand, in monopoly situations, firms can afford to adopt a
reactive mode of capacity build-up. In several cases, firms tend to utilize a typical capacity build-
up mode. Figure 8.10 presents a graphical representation of alternative modes of capacity build-
up.
Firms go through a cycle of overcapacity, appropriate levels of capacity, and insufficient capacity. This is a
typical mode of capacity build-up.
The approach taken to measure capacity depends on several factors. First is the issue of volume
and variety of products and services offered by the operations system. A high-volume system
may find it appropriate to use output measures of capacity and a low-volume–high-variety
system may use input measures of capacity.
A petrochemical plant such as Bharat Petroleum Corporation Limited (BPCL) measures its
capacity in terms of the number of barrels of crude oil it can process per day. The refining
capacity of its refinery in Mumbai is 260,000 barrels per day. Similarly, a PCB manufacturer
measures its capacity by the amount of square metres of board that it can fabricate. A television
manufacturer often measures its capacity per million picture tubes produced. The Hazira plant of
Reliance Industries Limited measures the capacity of its polyethylene plant in terms of tonnes of
polyethylene processed. A tool room facility will measure its capacity in terms of machine-
hours. All these examples share a common feature, that is, these measures are input measures of
capacity. Several firms operating in a low-volume–high-variety situation find it easier to measure
capacity on the basis of the quantum of input that it can process in a unit time. One can observe a
similar phenomenon in a service setting also. A management consulting firm or a legal
consulting firm will measure its capacity by the consulting man-hours available per month.
Similarly, a hospital will measure its capacity in terms of the number of beds.
Capacity utilization is measured as a ratio of the capacity put to use to the total available capacity in a unit of
time.
The most commonly used method for measuring the effectiveness of capacity usage is the
utilization of resource(s). Capacity utilization is measured as a ratio of the capacity put to use to
the total available capacity in a unit of time.
Depending on the method adopted to measure capacity, this definition can be suitably modified.
For instance, if output measures of capacity are used, then the numerator will indicate the actual
output and the denominator will indicate the installed capacity in terms of the number of units of
output. Alternatively, if input measures are used, the numerator and the denominator could be
correspondingly redefined.
When the volume of production is higher and the variety is lower, it is appropriate to use output measures of
capacity.
One of the issues that an operations manager needs to have some clarity on is the manner in
which the total capacity available is determined. There are several assumptions involved in
arriving at this number. The simplest measure is the theoretical maximum capacity available.
Consider a machine such as a hydraulic press in a fabrication shop. Considering a month to be of
30 days, the theoretical maximum capacity available is 720 hours per month. However, the firm
may have a policy of operating only two shifts. Moreover, there will be some time earmarked for
scheduled maintenance of the machine. Excluding all these will result in the effective capacity
available during the month. For instance, if 20 hours per month are set aside for scheduled
maintenance then a two-shift operation will reduce the maximum available capacity to 460 hours
[(30 day × 16 hours) − 20 hours].
The actual capacity realized will be much lower than the effective capacity of the resource.
The difference between the effective capacity available and the actual capacity realized is due to
several factors that result in causing the capacity to be idle. One way to explain this difference is
to make use of a simple equation pertaining to capacity, as practiced by Japanese manufacturers:
The Canon production system describes nine different types of waste that occur in a production
system. These include waste in: (1) planning, (2) operations, (3) start-up, (4) equipment, (5)
defects, (6) materials, (7) indirect labour, (8) human resources, and (9) expense.5
Of the nine wastes mentioned here, the first five directly relate to the capacity of the resources.
Improper planning of resource usage will bring down the utilization of resources, as they will
wait for material, operating instructions, and process sheets. Similarly, waste in operations and
start-up may take away valuable time from the resource and bring down the actual capacity of
the system. The most frequently encountered example of this is a very long set-up time for a
machine before a batch of components can be processed. Waste in equipment results due to
improper methods of using the equipment as well as poor maintenance, resulting in significant
loss of the time available for processing. Unscheduled breakdowns and longer times to restore
the machine to operation may take away capacity from the system.6 Finally, waste in defects
indicates the time lost in processing the defective components as they are eventually scrapped.
Capacity planning issues vary markedly with respect to the time horizon in which the decisions
are made. It is useful to divide the time horizon into the long term, medium term, and short term
to understand the nature of issues to be addressed with respect to capacity planning. Table 8.3
illustrates the salient features of capacity planning under the three time horizons.
In the long term, the emphasis in capacity planning is on making the right amount of capacity
available to meet the projected growth. Typically, organizations initially make a certain
investment of capacity. However, as operations stabilize and market share increases, firms need
to take decisions in advance to plan for augmenting capacity. Alternative methods are available
to augment capacity. These include de-bottlenecking of certain stages of the process, addition of
more resources in certain parts of the operating system, large-scale expansion of capacity by
building additional plants, and mergers and acquisitions of other operating units. These options
differ in the amount of additional capacity that is brought into the system and the cost and
technological aspects of capacity build-up. De-bottlenecking is a commonly employed method in
process industries as it is easy to identify the flow and bottleneck points in the system.
In the medium term, the focus in capacity planning shifts to balancing available capacity with the demand.
Source: Based on R. Balaji, “Waging War on Waste,” Unpublished report submitted in the MPT programme at IIM Bangalore.
Capacity planning in the short run is very different from capacity planning in the medium or
long run. In the short run, the maximum available capacity is fixed. Therefore, the capacity
planning exercise is confined to making effective use of the available capacity and ensuring that
unforeseen developments such as equipment breakdowns and poor planning methods do not lead
to keeping the capacity idle. The focus in capacity planning in the short run is on maximizing
resource availability and efficient use of the resources. Several operations management tools are
available to manage capacity in the short run. These include planning, scheduling methods, and
maintenance management methods. Some of these topics are discussed in detail in Chapters 15
and 18. Waiting-line models are a generic class of tools used to analyse capacity issues in both
service and manufacturing firms. We discuss this in Chapter 10. Similarly, a detailed analysis of
the impact of capacity on the working of the system can also be studied using simulation
modelling of the system. See the supplement of the end of the chapter 10 for more details on
simulation.
Capacity planning in the short run is confined to making effective use of the available capacity.
Capacity planning is a systematic approach to three issues pertaining to capacity: estimating the
amount of capacity an operating system requires, evaluating alternative methods of augmenting
capacity, and devising various methods to use existing capacity effectively.
Although this computation appears simple, in reality, there are certain operational difficulties in
estimating SL and EL. The standard labour hours are supposed to be established by a standard
setting routine in every organization. This will call for studying each process, estimating normal
allowances and incorporating them. Moreover, as the process is put into practice, some
assumptions are required about the efficiency of the workers. This efficiency is likely to be
somewhat low during the initial stages. However, as the employees experience the learning curve
effect, the efficiency of the process improves, thereby calling for a revision of these standard
labour hours. Despite this, revisions are not done as frequently as the process warrants. This is
because the earning potential of the employees is inextricably linked to standard hours and
efficiency assumptions. By keeping the standards loose and efficiencies low, it is possible for
some workers to earn productivity bonuses even for normal work. Due to these reasons, in a vast
majority of industrial organizations, SL and EL have an industrial relations angle as well. In
recent years, however, the notion of productivity and the manner in which employees are
rewarded for good performances in productivity are not linked to these issues. Operations
managers need to be aware of these limitations while making a judicious choice of these
parameters.
One can use a similar expression for computing the machine requirements by using the
subscript M in the place of L.
EXAMPLE 8.3
A manufacturer of medium-voltage circuit breakers is planning for a capacity build-up of 8
cubicles and 13 circuit breakers per day. A year consists of 305 working days. The
fabrication division is responsible for manufacturing metal sheet components that are welded
together to form the cubicle. Some metal sheet components are welded to host the circuit
breakers inside the cubicle. The components are painted after welding. While the fabrication
uses a CNC turret press, painting is a manual job. The standard time required at the CNC
turret press for fabricating a cubicle is 150 minutes and the time for the breaker housing is 36
minutes. A cubicle requires 43 m2 of area to be painted and a breaker housing requires 2.60
m2 of painting. The standard time required to paint 1 m2 of area is 18 minutes. The machines
work at 80 per cent efficiency and the manual labour works at 90 per cent efficiency. Using
the given data, compute the labour-hour and machine-hour requirements.
Solution
Capacity planning is done for the planning horizon. In our example, the basic data for the
problem pertains to the planning horizon.
Since the demand during the planning horizon is the basis for computing the capacity
requirement, we compute the demand using the data. Since eight cubicles are to be
manufactured per day, in a year the required capacity is to fabricate (83 × 05) cubicles, or
2,440 cubicles. The computation is as follows:
Planning horizon: 1 year
Number of working days in a year: 305 days
Therefore, the demand for the planning horizon will be as follows:
The standard time for fabricating one unit of cubicle and breaker housing are available.
Also known is the efficiency of the machine. Using Eq. 8.4, we can therefore compute the
machine-hours required. The computations will be as follows:
Efficiency of CNC turret press: 80%
One can use Eq. 8.3 to compute the labour hours required in a similar manner. The labour-
hour calculation will be as follows: Efficiency of workers: 90%
Standard man-hours for painting 1 m2: 0.30
Cubicles Breaker Housing Total
Square metres of area to be 43.00 2.60
paintedper unit
Total area to be painted 104,920.00 10,309.00 115,229.00
during one year (m2)
The comparison acts as a basis for the operations manager to understand the consequence of the capacity expansion
initiative.
It helps to categorize the resources into those with adequate capacity and those with insufficient capacity and helps to
focus on the latter category for problem solving.
It provides an impetus for process-plan changes and improvements for uncovering waste, thereby leading to the
discovery of more capacity at some of the bottlenecks
It helps the manager to draw out the capital budgeting and investment requirements of the capacity expansion initiative.
EXAMPLE 8.4
Consider the fabrication shop of Example 8.3. Suppose the factory works on a two-shift basis
with six workers in the paint shop in each shift. There is only one CNC turret press currently
available. Suppose prior data shows that the equipment at the shop has a downtime of 12 per
cent and the absenteeism rate of the employees is 5 per cent. Assess the impact of the
capacity expansion initiative in the plant.
Solution
The capacity requirements have already been computed in Example 8.3. They are as follows:
Labour hours required at the paint shop = 38,409.67 Machine-hours required at CNC turret
press facility = 10,598.75
Now, let us compute the available capacity using Eq. 8.5. The relevant computations are as
follows:
Rounding to the nearest integer, three more workers per shift and two CNC turret presses
need to be added to meet the targeted capacity.
System availability
A product is manufactured in a shop using a five-stage process. The first step in the process
is to cut the sheet metal to required shapes and sizes using a shearing process. After the
shearing process, the components are subjected to pressing operations to alter the shape of
the flat sheet as per the design. In the third stage of the process, welding is done to join the
components. The next step in the process is a painting operation. After painting, the
components are packed and kept ready for dispatch. The time taken for each of these
operations is 20, 30, 15, 12 and 6 minutes, respectively. Presently, each stage has only one
machine for operation. Map the process and analyse the capacity with respect to the
following scenarios:
a. If the shop works on an 8-hour shift with an effective available time of 450 minutes, what is the production
capacity of the shop?
b. Where is the bottleneck in the system? If we want to add one machine, where should we make the investment?
c. Identify the additional capacity required for a daily production target of 25 units. Compute the utilization of the
machines as per the revised capacity calculations.
d. What are the key inferences of this exercise?
Solution
Based on the description in the question, one can map the process and identify the extent of
time the resources will be used at each stage of the process for manufacturing. Figure 8.13
shows the process map. The numbers in parentheses denote the time required at each stage.
a. These timings can be translated into production capacities by dividing the total available time per shift by the
required time at each stage. The production capacities are:
The smallest number in these calculations limits the production capacity for the shop. Therefore, the current
production capacity is 15 units per day.
The smallest number in these calculations limits the production capacity for the shop. Therefore, the current
production capacity is 15 units per day.
b. The slowest process is the bottleneck in the system and it dictates the production capacity. Therefore, it is prudent
to channelize the investment at the bottleneck process. By adding one more machine, the situation changes as
shown in Figure 8.14.
Since two machines are available at the pressing stage, the effective time per unit will now be 15 minutes.
Therefore, the bottleneck now shifts to shearing and the revised production capacity will be 22.50 units a day.
c. The production target is now 25 per day. Since a day has 450 units, the maximum time that the process can take at
each stage is 18 minutes. The packing, painting, and welding sections have timings less than 18 minutes. Therefore,
they do not need any more investment in capacity. By adding one more machine at the pressing stage, the effective
time will be less than 18 minutes. Similarly, by adding one more machine at the shearing stage, the effective time
will be 10 minutes. Using Eq. 8.1, the capacity utilization for the machines is as follows:
d. Key inferences: The computation of individual capacity availability and requirement is the first step (and an
important one) in the capacity planning exercise. The value of this exercise lies in our ability to use this
information to identify bottlenecks. A process mapping exercise will help us achieve this objective. The utilization
of the shearing process has dropped to 56 per cent due to the addition of a machine. Clearly, the addition of one
more unit of resource may not always be the best solution. This points to the need for exploring alternative methods
of augmenting capacity.
Example 8.5 pointed out the need for looking at alternative methods for capacity augmentation
during a capacity planning exercise. The straightforward option is to add more resource units.
Although it is simple from an operational decision-making point of view, in reality, the
implications could be far too many. The cost of the resource in question, the utilization of the
resource, and the ease of implementation of the capacity expansion plan are some of the issues
that will influence the choice of this option. Moreover, as more and more units of resources are
added, it may significantly impact the long-term operational costs of the system. For example, if
several new workers are added, then the cost of the system increases not merely in terms of
salary but also in terms of support costs, benefits, and health and post-retirement provisions.
Similarly, as more machines are added, equipment maintenance costs will go up in the long run.
Furthermore, if the resource in question is very expensive and its utilization is likely to be low, it
is a very unattractive option, and operations managers should look at other ones.
Waste Elimination
One method to increase the capacity of the resources in the system is to employ Japanese
methods of resources planning and management, as described earlier in the chapter. In a nutshell,
it is for the operations manager to make use of Eq. 8.2 and progressively uncover capacity from
the system by elimination of waste (see Ideas at Work 8.2 for an illustration of this method of
increasing the capacity in the system). Process industries increase capacity by de-bottlenecking
operations. Since in process industries, the system is a continuous flow of material from the raw
material to the finished goods, merely identifying the bottleneck stage in the process and de-
bottlenecking it can increase the capacity of the system. Japanese approaches to waste
elimination and/or addition of new capacity in that stage will increase the capacity of the system.
However, the bottleneck will shift elsewhere and the process could be repeated.
Subcontracting/Outsourcing
Capacity augmentation need not always be done in-house. An alternative approach to capacity
augmentation is to subcontract the work to competent vendors and service providers. The logic
of the subcontracting decision closely follows that of a “make-or-buy” decision that purchase
managers and management accountants address in their respective domains. Several
considerations influence the subcontracting decisions in firms. Primary among them is the lack
of capacity to meet the current demand. The other consideration is the technological intensity and
criticality of the item for which capacity is being subcontracted. When the item is of low
technical intensity and criticality, then it is less risky to subcontract capacity. The firm will not
lose any valuable trade secret or know-how pertaining to the product being manufactured. The
third issue is one of cost. When the cost of performing an activity in-house is much higher than
what is available outside, then it is appropriate to outsource such activities and use the released
capacity for other important activities.
Firms can react much faster to market requirements by using subcontracting. Subcontracting is also very useful
for managing peak-hour demand.
Subcontracting offers several advantages to a firm. First is the flexibility to handle fluctuations
in demand. By investing in in-house capacity, the firm may run the risk of under-utilization
should the demand for the product/service come down in the future. In-house capacity
augmentation is a time consuming process. On the other hand, firms can react much faster to
market requirements by using subcontracting. Subcontracting is also very useful to manage peak-
hour demand. By having an in-house capacity equal to the average demand during a period,
organizations can address peak-hour requirements using subcontracting.
Despite these advantages, subcontracting has some drawbacks. The major challenge in sub-
contracting is to identify an appropriate vendor for providing subcontracted services. If the
selection of the vendor is not done carefully, poor performance of the vendors will impact the
firm’s business and market standing adversely.
The capacity planning exercise requires methods by which alternative options are evaluated. Two
metrics are useful to perform the evaluation. In the cost-based method, each alternative can be
evaluated from the perspective of the costs and benefits accruing out of the alternative. For
example, a firm may be considering three options: not to do anything about capacity, add a new
machine, or opt for subcontracting. Another firm may have three options of varying
technological and operational capabilities for capacity addition, resulting in different capital
costs, operating costs, and useful life of the resource. In each of these examples, one can evaluate
the alternatives from the perspective of costs and benefits.
In the cost-based method, each alternative can be evaluated from a perspective of cost and benefits accruing out
of the alternative.
The other method is to use operational performance based methods for comparing
alternatives. If the choice is to go for multiple resources, one can analyse the impact of the
alternatives in terms of utilization of capacity and the waiting time of the jobs. If more machines
are added, the waiting time as well as the utilization of the resource will come down. It is
possible to analyse and select the best alternative on the basis of these measures.
In operational performance based methods, alternatives are analysed in terms of utilization of capacity and the
waiting time of the jobs.
Decision trees are useful to evaluate alternative capacity choices on the basis of the cost of the
capacity and the benefits. Further, the inherent uncertainty in the demand for the resource can
also be modelled to assess the impact of the decisions. A decision tree is a schematic model in
which different sequences and steps involved in a problem and the consequences of the decisions
are systematically portrayed. Decision trees comprise nodes and branches. Each node represents
the decision point and the branches represent the potential outcomes of the decision. The
consequence of each outcome is measured as the cost of the impact, and the uncertainty
associated with each outcome could be associated with the requisite branch. Using this basic
information, the tree is constructed. After the tree is constructed, each branch in the tree is
evaluated with respect to the costs, benefits, and uncertainty. The tree is evaluated from right to
left (from end to beginning). As we move from right to left, unattractive portions of the tree are
eliminated to arrive at the final decision. The use of a decision tree for evaluating capacity
alternatives is explained with the help of Example 8.6.
EXAMPLE 8.6
Solution
Decision Point 2
Revenue from adding new capacity = ₹850,000 × 4 = ₹3,400,000
Cost of adding new capacity = ₹787,500
Net revenue from this option = ₹2,612,500
Revenue from the subcontracting option = ₹350,000 × 4 = ₹1,400,000
Therefore, the best option at this stage is to go for adding new capacity.
Node A
Revenue in case of high demand = ₹850,000 × 5 = ₹4,250,000
Revenue in case of moderate demand = ₹400,000 × 5 = ₹2,000,000
E[A], the expected revenue at Node A, will be as follows:
E[A] = (4,250,000 × 0.40)1(2,000,000 × 0.60) = ₹2,900,000
Cost of adding new capacity = ₹750,000
Net revenue from this option = ₹2,150,000
FIGURE 8.15 The decision tree
Node B
Revenue in case of high demand = ₹550,000 × 5 = ₹2,750,000
Revenue in case of moderate demand = ₹300,000 × 5 = ₹1,500,000
E[B], the expected revenue at Node B, will be as follows:
E[B] = (2,750,000 × 0.40) + (1,500,000 × 0.60) = ₹2,000,000
Cost of expanding the capacity = ₹275,000
Net revenue from this option = ₹1,725,000
Node C
In the case of high demand, our evaluation of the decision point shows that it is better to add
new capacity after one year than continuing with the subcontracting option. This will fetch a
net revenue of ₹2,612,500 during the last four years. Moreover, in the first year, it would
have fetched a revenue of ₹350,000.
Therefore, Revenue in case of high demand = ₹2,962,500
Revenue in case of moderate demand = ₹180,000 × 5 = ₹900,000
E[C], the expected revenue at Node C, will be as follows:
E[C] = (2,962,500 × 0.40) + (900,000 × 0.60) = ₹1,705,000
There are no costs of subcontracting. Therefore,
Net revenue from this option = ₹1,705,000
Since the expected value of Node A is the highest, we go with the decision of adding new
capacity at the beginning itself rather than just de-bottlenecking or waiting for a year to sense
the demand. In the decision tree in Figure 8.15, the best option at each decision point is
shown with bold lines.
SUMMARY
Process analysis is the mechanism used to understand the impact of process design on output, cost, or any other
performance metric using some analytical tools. It also amounts to understanding the impact of alternative process
configurations on the chosen performance metric.
A pictorial representation of all process-related information can be developed using process flow-charting. Process flow
charting employs a set of standard symbols and graphical tools to represent all the information pertaining to the process.
Three generic planning premises are in use in operations management: make to stock (MTS), make to order (MTO), and
assemble to order (ATO).
The basic approach to planning in an MTS system is to schedule production for the purpose of replenishing stock to
some predetermined level. On the other hand, the MTO planning methodology is used by organizations that typically
manufacture products with high variety and low volumes.
Organizations use input- and output-based measures for defining capacity. High-volume–low-variety manufacturers use
output-based measures for defining capacity.
Capacity can be augmented in organizations in alternative modes. Typical modes of capacity augmentation follow a
cycle of insufficient capacity–excess capacity–appropriate levels of capacity.
Capacity = Work + Waste. Therefore, one way of augmenting capacity in organizations is to eliminate waste from the
system.
The capacity planning framework changes with the time horizon. The emphasis in the short term is to maximize
capacity availability through efficient use of resources. On the other hand, in the medium term, the emphasis is on
matching the supply with the demand on a period-by-period basis.
The capacity planning framework consists of a three-step process:
Estimate the capacity requirement for the planning horizon
Identify the quantum of capacity to be augmented
Select an appropriate alternative for capacity augmentation
Selection of an appropriate alternative for capacity augmentation could be done using a decision tree model.
1.
3.
4.
REVIEW QUESTIONS
1. What do you mean by process analysis? Why should organizations conduct process analysis?
2. An automobile garage owner wants to perform process analysis in order to improve her customer service process. You have
been engaged as a consultant. Prepare a one-page plan of action for the automobile garage owner.
3. What do you mean by process mapping? Explain with the help of an example.
4. Suppose you want to visit your bank to deposit your salary cheque and then withdraw some money from your account. Use
your knowledge of process mapping and draw the process.
5. For the example given in Question 4, perform an NVA analysis and prepare a one-page report to the bank manager about the
current status of the process. What are your key recommendations to the bank manager with respect to improving the process?
6. What do you understand by the term “business process”?
7. Give two examples of business processes in the following organizations:
8. What do you mean by BPR? Why is it an important process improvement tool for an organization?
9. Outline the key steps involved in a BPR exercise. Are these steps generic enough to apply to any business process?
10. Suppose that a travel agency has performed a process analysis with respect to attending customer requests for airline and
hotel bookings. The analysis shows that it takes 30 minutes to finish this process and has identified the insufficient number of
computers, people, and information as the possible reasons. Suppose the project requires redesigning to bring it to 15 minutes,
how will you go about the task?
11. How do you define capacity in an operating system? Are input measures of capacity more appropriate than output measures
of capacity?
12. Identify an appropriate measure of capacity in operating systems for each of the following:
13. What are the alternative methods by which capacity could be augmented in an operating system? Explain with some
examples.
14. How does knowledge of capacity help an operations manager?
15. What are the alternative metrics available to evaluate available alternatives for capacity planning/ expansion?
16. What are the alternative methods available for building up capacity in a system? Which method is the most suited for
implementation?
17. Is capacity management a short-term or a long-term issue? Explain with examples.
18. What are the implications of adding capacity in huge chunks at less frequent intervals?
19. How do you define capacity utilization? What factors influence the utilization of resources in an operating system?
20. Why is capacity planning important? Outline the steps involved in a capacity planning exercise.
21. What do you mean by waste elimination? Does it have any relevance for capacity management?
PROBLEMS
1. Ram runs a small roti shop. People can place their order over the phone and come and collect the roti. Orders are taken for a
standard pack of 5 rotis. The entire process is done in batches of 5 rotis only. Roti making and delivery has five steps.
Preparing the dough (3 minutes), rolling (5 minutes), heating (4 minutes), packing (2 minutes), billing and cash (1 minute).
Given this process, answer the following:
a. Prepare a process flow chart and identify the bottleneck for the shop.
b. If somebody calls Ram over the phone, when should he ask them to come?
c. Suppose you go to his shop at 9:00 a.m. and then order 15 rotis. When will you collect them?
d. If Ram keeps his shop open for 4 hours, how many rotis can he hope to make?
2. There are six stages in a semi-automated process and the relevant details are given in Table 8.6.
a. If there are 24 workers in the fabrication shop and 12 each in the other two shops, what will the utilization of the
workers be at the current level of operation?
b. How many workers are required if the over-all utilization of the factory is targeted at 90 per cent?
c. What is the additional number of workers required in each of the shops if the annual production increases to 25,000?
(Assume a targeted utilization of 80 per cent in all the shops.)
5. An ortho-specialty hospital conducting micro-surgeries is now operating at a capacity of 20 surgeries a day. On average, each
surgery takes 45 minutes of the surgeon’s time, one hour of the pre-operative staff’s, and two hours of the post-operative
staff’s in the hospital. A surgeon works for four hours a day while the staff work for eight hours a day. The hospital currently
estimates that the surgeons work at a utilization of 94 per cent and the other staff at 85 per cent utilization. The hospital plans
to increase its capacity by 20 per cent and it has all other infrastructure facilities to handle this proposed capacity expansion.
What it needs to plan for is the human resources required during this expansion. It also proposes to bring down the utilization
of surgeons to 85 per cent and that of the other staff to 75 per cent. The excess capacity will be used for training and
development activities. If the cost of locating new staff, training, and inducting them into the hospital is ₹20,000 per person
and that of a surgeon is ₹80,000, what will the initial costs of the expansion plan be?
6. A firm employs a process that has three steps to complete the process. Step one takes 10 minutes, step two takes 12 minutes,
and step three takes 14 minutes. There is setup involved in steps one and three, which takes 4 minutes each. Once set up, step
one can process up to two batches and step three up to three batches. Currently they operate with a batch size of one.
a. What is the hourly capacity of the process now? Which is the bottleneck in the process?
b. What is the effect of adding one more unit in the bottleneck step? Which is the new bottleneck?
c. Instead of adding one more unit at the bottleneck, if we increase the batch size to two what will be the process capacity?
d. What will be the process capacity if the batch size is three?
e. Which option would you recommend?
7. A restaurant works in a busy locality in Kanpur. At the moment they have adequate cooking capacity and space to seat the
customers. They would like to know the adequacy of the service manpower available with them. The order takers spend on an
average 15 minutes per arriving batch of customer and the servers spend 48 minutes.
a. If there are five order takers and 12 servers, how many batches of customers they can serve every day?
b. If a batch of customer arrives on an average every 4 minutes, can they handle the demand with the existing staff?
c. The restaurant found that during peak hours a batch of customer arrives every 3 minutes. Will they be able to handle the
demand?
d. How many order takers and servers should they employ if they want to work with an utilization of 75% (during non-
peak hours)?
8. A travel agency attends to various travel planning requirements of its clients. They work six days a week and are open
between 9.30 am and 6.00 pm. They have lunch break between 1.30 pm and 2.30 pm. Typically, servicing a client requirement
takes about 22 minutes. There are five travel consultants currently employed. There is 10% absenteeism of employees.
a. If they receive service requests at the rate of 8 per hour, what is the utilization of the service consultants?
b. They want to expand their operations so that they can handle 100 requests per day. They would also like to work with a
targeted utilization of 80%. Identify the number of travel consultants to be employed under the following conditions:
9. TSMC manufactures semiconductor chips for companies such as Nvidia and Qualcomm. A simplified process for chip
manufacturing at TSMC consists of three steps. Step one is Depositing, step two is Patterning and step three is Etching. Table
8.7 gives the required processing time and setup time. Assume the unit of production is a wafer. The plant operates for 8 hours
a day. The demand is 800 wafers per day. Also assume that a setup can only begin once the batch has arrived at the machine.
All units of a batch need to be processed before any of the units of the batch can be moved to the next machine.
10. An eye hospital has made an elaborate study on its process and the arrival pattern of the customers to the hospital for eye
examination and corrective procedure for a certain kind of illness. Since the patients are directed from a pre-screening section,
all of them are subjected to a similar treatment. All the arriving patients go through the same set of steps. However, only one
in four go through the HFA and OCT stages. On average, it was found that 12 patients arrive every hour during the operation
of the hospital. The details are given in Table 8.8. The time (minutes) is an average over some period of observation.
a. Develop a process flow chart for this problem. Use the process flow chart and indicate the capacity of the various stages
of the system.
b. How long will an arriving patient be in the system?
c. Estimate the capacity of manpower and other resources required for the hospital.
d. Are there sufficient numbers of resources in the system?
e. What is the cycle time for the system? Which is the bottleneck stage?
f. If the hospital is willing to make further investments in manpower and equipment, can you help them prioritize their
investments?
Note: *Two minutes are spent by the nurse in applying drops to a patient. Nurses conduct both the tasks in an inter-changeable fashion.
MINI PROJECT
For the organization that you have chosen, undertake an improvement project along the
following lines. First, talk to the head of the organization to understand the major problems and
concerns. Based on your understanding of the problem and consultation with the members in the
organization, identify a set of measures that needs to be improved in order to solve the problem
and set some meaningful targets for the same over the next six months. Collect data to assess the
current status of these measures, the business processes affecting its performance and the
activities that currently constitute these business processes. Perform the necessary analyses of
the various business processes and write a final report detailing the steps and your
recommendations to the organization regarding the nature of efforts to be undertaken for
improving the situation and solving some of the problems listed. Also, clearly identify the
challenges in the process and the nature of the precautions that the unit head should take while
implementing your recommendations.
CASE STUDY
Introduction
Vasudhri Tooling Limited (VTL) is promoted by a technical training foundation and managed
by a board of directors. VTL has two production units, one in Bangalore in Karnataka state, and
the other near Vellore in Tamil Nadu. Both have independent and sophisticated tool rooms and
component divisions for the manufacture of plastic-moulded and sheet-metal components. The
unit at Bangalore has a separate CNC division and product engineering division for the
development of products. VTL has dedicated ancillaries where all the assembly-line know-how,
equipment tooling and technical supervision and control are held.
The main activities of VTL include the design and manufacture of press tools and injection
moulds, along with the manufacture of sheet-metal parts and injection-moulded/ compression-
moulded parts. The company’s main activities fall into the category of manufacture of electrical
products for automobiles. The major products manufactured by the company are steering-
column combination switches for four-wheelers/trucks, parking lamps, headlamps, blinker
lamps (indicators), wiper-speed controls, hazard-warning switches and central locking systems.
All the major players in the automotive industry in India are VTL’s customers. In addition, VTL
markets its products in the after-sales segment as spares.
Hence, the total manufacturing lead time for production of switch assembly for a batch of 600
switches = 33.5 + 50.0 + 32.5 + 8.0 = 124.0 hours.
SUGGESTED READINGS
N. A. Bakke and R. Hellberg, “The Challenges of Capacity Planning,” International Journal of Production Economics
30–31 (1993): 243–264.
R. D. Hammesfahr, J. Jack, A. Pope and A. Ardalan, “Strategic Planning for Production Capacity,” International
Journal of Operations and Production Management 13, no. 5 (1993): 41–53.
P. Jonsson and S. A. Mattson, “Use and Applicability of Capacity Planning Methods,” Production and Inventory
Management Journal 43, no. 3 and 4 (2002): 89–95.
K. J. Klassen and T. R. Rohleder, “Combining Operations and Marketing to Manage Capacity and Demand in
Services,” The Service Industries Journal 21, no. 2 (2001): 1–30.
R. T. Yu-Lee, Essentials of Capacity Management (New York: Wiley, 2002).
CHAPTER 9
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
What are the factors that significantly determine process design in organizations?
How are process choices and flow characteristics of jobs linked? Do they have any bearing on the structural aspects
of organizations?
What are the well-known process design options available?
How do volume–variety–flow dimensions influence the layout decision in an organization?
What are the different types of layouts available?
How can we assess the appropriateness of a layout design?
How can one design a product layout, a process layout, and a group technology (GT) layout?
What software packages are available for solving the layout problem?
Will new technology make process design any simpler?
Over the years, Milltec has realized the importance of addressing manufacturing process design and layout issues. Through a
process of continuous improvement, Milltec employees and management have realized that several benefits arise out of designing
an appropriate layout and workplace organization.
The design of processes essentially consists of various choices that we make with respect to the flow of parts in
a manufacturing system.
Before we understand the various process choices available to an operations manager for
configuring a manufacturing system, let us understand the factors that influence the choice of
alternative processes. Generally, three important aspects have a significant influence on the
process: volume, variety, and flow.
Volume
Volume indicates the average quantity of products produced in a manufacturing system. In
reality, organizations differ on this from one end of the spectrum to the other. Turnkey project
management organizations such as Larsen and Toubro (L&T) and Bharat Heavy Electricals
Limited (BHEL) will typically have a production volume of just one. At the other extreme,
organizations catering to the consumer non-durable and FMCG sectors have high volumes of
production necessitating continuous production. Examples include the manufacture of spark
plugs, electric bulbs, soaps, standard electrical fittings, and common household utilities. Many
chemical process industries also belong to this category. Between these two extremes, we have
mid-volume manufacturers who offer several versions of products. There are numerous examples
of mid-volume manufacturing in consumer durables, white goods, and several industrial
products. Hindustan Motors, for instance, manufactures several types of earth-moving equipment
that belong to the mid-volume category.
Three important aspects have a significant influence on the process: volume, variety, and flow.
Variety
Variety refers to the number of alternative products and variants of each product produced in a
manufacturing system. Let us consider a watch manufacturer such as Titan Industries Limited.
Titan offers more than 40,000 varieties of watches. These varieties stem not only from
alternative product lines but also from variants in each product line. Similarly, consider a car
manufacturer such as Tata Motors. The Tata Indica is available in several models. Basically,
there are the petrol and diesel versions. However, within the petrol and the diesel versions, there
are variations such as the GLS, GLX, VX, VXi, AX, and so on. While one version will have
power steering, the second will have power steering and power windows, the third may have
automatic transmission, and so on. A choice of colours adds another dimension of variety to
passenger car manufacturing.
An increase in the variety of product offerings is likely to introduce variety in manufacturing
processes. More varieties may mean alternative production resources, materials, skill of workers,
and an increase in the number of stages of production. It may call for better operations
management practices. Planning and scheduling may become more complex on account of these
added choices in the manufacturing system.
One sees a similar phenomenon in service systems too. Assume that a travel agency was
earlier engaged only in selling tickets for bus journeys from Delhi to Chandigarh. In the course
of time, if the agency added booking of bus tickets to multiple destinations, rail tickets, air
tickets for domestic travel, and air tickets for international travel, one can imagine how many
alternative choices of processes, skill sets and resources are required by the travel agency.
Operations management in the case of additional services is far more involved and complex.
Flow
Irrespective of the nature of activities involved, all manufacturing systems require certain input
material with which the processes begin. The material undergoes a conversion process from the
raw material stage to the finished goods stage. Flow indicates the nature and intensity of this
phenomenon. Knowledge of flow provides vital clues to the operations manager about
production planning and control issues that need to be addressed. It also provides useful
information about the complexities of operations management in an operations system.
Volume, variety, and flow are linked to one another. In general, volume and variety will have
an inverse relationship. When the volume of production is very high, the firm is likely to be
engaged in the manufacture of fewer varieties of products. On the contrary, if the firm caters to a
wide range of products and services, then the production volume of each of these variations is
likely to be very low. Figure 9.1 identifies typical examples belonging to three categories: (i)
high volume–low variety, (ii) mid volume–mid variety, and (iii) high variety–low volume.
Furthermore, volume and variety influence the flow patterns in a manufacturing system.
Therefore, process design choices available to an operations manager could be understood by
analysing alternative flow patterns. Using this approach, we shall understand how volume,
variety, and flow influence the process design in an operations system.
FIGURE 9.1 Relationship between volume and variety in operations systems
Process characteristics are largely determined by the flow of products in a manufacturing system.
Therefore, it is useful to first identify the nature of flow in operations and then identify the
process choices available to the operations manager for each category of flow. Generically, three
types of flows can be identified in manufacturing systems: continuous, intermittent, and jumbled.
Three types of flows exist in manufacturing systems: continuous, intermittent, and jumbled.
Process Industries
Process industries have peculiar characteristics compared to discrete manufacturing industries.
Process industries have a closed manufacturing set-up and finished goods are typically obtained
at the end of the manufacturing process. In some cases, it is possible to derive products during
the intermediate stages of the process in the form of by-products. An example of a process
industry is paper manufacturing. Figure 9.2 illustrates the paper-manufacturing process in simple
terms.
An analysis of the paper-manufacturing process will reveal several features of a process
industry. As shown in the figure, there are three stages in the paper-manufacturing process: the
preparatory stage, pulp making and papermaking. During the preparatory stage, logs of wood and
chips of various sizes are first fed into the system. The logs of wood and the chips are broken
into smaller and uniform pieces in a crusher. After the pieces are cut into uniform pieces, they
are “cooked” in a huge chamber where some chemicals are also added. During this process of
cooking, the pieces are digested and the fibres are laid in a uniform pattern. This is done through
a chemical process. After this process is complete, it is ready for making wood pulp. The pulp-
making process is nothing but further refining of the wood fibre and cleaning and bleaching of
the material. Finally, excess moisture is removed to form the pulp. At this stage, the pulp is
almost devoid of moisture and is in a near-solid state. Therefore, it is possible to press it into the
required thickness and remove further moisture. This is the last stage of the process known as
papermaking. The dried and processed paper can be rolled into reels and cut to required lengths
and shapes.
FIGURE 9.2 The paper-manufacturing process
The need to keep the production facilities continuously running in a process industry directly
translates into a need to be good in maintenance management. Therefore, process-industry firms
invest more time in developing good maintenance practices.1
Moreover, the amount of spare-parts inventory that they carry will be a significant portion of
the overall inventory in the system. Ideas at Work 9.2 illustrates how disruptions in the
production process, though unavoidable, can raise the costs incurred by process-industry firms
and increase their losses. Due to the continuous and closed-loop nature of the entire
manufacturing process, process industries tend to have very little work-in-process inventory.
Although not evident from the paper-manufacturing example, process industries benefit vastly
from the vertical integration of their operations as well as from the addition of several secondary
conversion facilities. This is because, at certain intermediate stages of the processing, by-
products are released from the manufacturing system. Having the facilities to process them into
useful products will lower the overall cost of the production system and increase revenue
generation opportunities. The other alternative will be to transport the by-products to another
point of demand at an extra cost. One of the best examples of vertical integration can be seen in
the development of various facilities by Reliance Industries Limited. Figure 9.3 provides a good
illustration of the backward integration concept for process industries.
FIGURE 9.3 Backward integration in process industries: the case of Reliance Industries
Source: Based on the author’s research
As in the case of process industries, balancing of capacity across various stages of the
manufacturing system is required to maximize throughput and productivity. Moreover, line
stoppages due to breakdown and asynchronous production practices will also be detrimental to
the working of the system. Therefore, better maintenance management practices and appropriate
design of product-based layouts are important operations management practices to be followed in
a mass-production system.
VIDEO INSIGHTS 9.2
Automobile manufacturing has always been in the forefront of bringing new innovations in manufacturing technology,
processes, and design of manufacturing systems. Therefore, one can develop a good understanding of the design of
manufacturing systems by closely observing manufacturing of passenger cars. Manufacturing of Honda Amaze at the
Tapakura plant in India and Mercedes Benz C class cars in an overseas plant demonstrates these concepts well. To know
more about these, find the video links (Video Insights) in the Instructor Resources or Student Resources under section
Downloadable Resources (http://www.pearsoned.co.in/BMahadevan/) subsections Bonus Material.
Intermittent Flow
All manufacturing firms do not have the benefit of high volume and low variety like a mass
production system. As customer preferences increase, the need for creating more variety also
increases and firms respond to this requirement by expanding the product line. Moreover, as
competition increases, organizations try to retain their market share by introducing new products.
An illustration of this is the passenger car segment in the country. The passenger car segment
was delicenced in 1993. Until then, there were very few car manufacturers. They also offered
fewer variations. However, once the competition increased and customer expectations went up,
product variety also went up.
When variety increases, the volume of production for each variation will be less compared to a
mass production system. Therefore, dedication of manufacturing resources for each variation
may not be a feasible option. Manufacturing resources will be shared by a group of products.
Since each product may have different processing requirements and sequences of operations, the
flow will become complicated. Therefore, the manufacturing process is characterized as mid-
volume, mid-variety systems and process design for this type of system must address issues
arising out of intermittent flow.
Discrete and process industries have alternative ways by which they can address the flow
complexities arising out of a mid-volume, mid-variety scenario. In the case of process industries,
one way to minimize this flow complexity is to operate the system in batches. In one batch, one
set of variations is manufactured and in the next batch, another set of variations is manufactured.
In between these two batches, the necessary set-up and changeover of resources are made to
facilitate smooth production and maximize productivity. Such an arrangement results in
intermittent flow systems.
Reliance Industries Limited manufactures more than 30 grades of polyethylene in their Hazira
plant. These include several variations used in industrial, medical, and food-grade applications.
While dedicated processing facilities are available for some variations on account of their huge
volume of production, several other grades need to share the same set of manufacturing facilities
for production. There are variations in the manufacturing processes across these products. For
example, food-grade applications require certain precautions to be taken. The nature of additives
to the process is also different. Therefore, after one variation is manufactured, the plant is
stopped and the required changes are made to the process before the next variation is produced.
Similarly, in the manufacture of drugs and various formulations in a pharmaceutical industry, the
batch-processing method is applied.
In the case of discrete manufacturing systems, the entire manufacturing set-up can be split into
units in which similar processing requirements can be combined and manufacturing resources
arranged to cater to these requirements. One batch of components may be produced in one of
these units and another batch in yet another unit. In each batch, components of different product
variations will be combined since they share the same processing requirement.
Figure 9.5 shows an illustration of the process design for intermittent flow in discrete
manufacturing systems. As shown in the figure, the manufacturing resources are organized on
the basis of similarity of manufacturing operations among several variations. Components of
various product variants flow in an intermittent fashion from pre-manufacturing activities up to
the final assembly and testing stage, taking alternative routes and timings to complete the
process.
By now, it is evident that operations management issues in intermittent flow systems are very
different from those of continuous flow systems. The notion of flow balancing is not an issue in
the case of intermittent flow systems. However, capacity planning for catering to the overall
requirements of the various varieties is important. Capacity estimation is difficult in an
intermittent flow system compared to a continuous flow system. The major issue pertaining to
the overall working of an intermittent flow system is the issue of changeover from one variety to
another. Production planning and scheduling must critically analyse the impact of alternative
changeovers and develop an optimal changeover sequence while planning for production.
FIGURE 9.5 Process design for intermittent flow in a discrete manufacturing system
Designing an appropriate structure and manufacturing layout is critical to the working of batch
manufacturing systems. Inappropriate choices in structure often lead to complicated material and
information flows. This results in complex production planning and control problems. The
tractability of the products and related information becomes difficult and may call for more
paperwork, and control mechanisms to resolve them. Typically, the level of supervision and
coordination will increase and organizations may resort to progress chasing and expediting of
late jobs. All these will finally result in long lead times in manufacture, poor delivery reliability,
excess and unwanted inventory, high overheads, and high overall cost of manufacturing.
Customer orders are typically for one-off items and organizations cannot benefit from any batching and
repetitive manufacturing practices, unlike continuous flow and intermittent flow systems.
Project organizations
Consider a turnkey project executor such as BHEL or L&T. The projects that they undertake
are typically large-scale, involve high levels of customization and have long lead times (in the
order of a few years). When the product or service offered is of a very large scale, it introduces
problems that are not encountered in normal operating systems. In such a case, multiple entities
are involved (architect, engineers, government regulatory bodies, construction workers, local
communities and society and so on). It also involves multiple stages of the process (design of
buildings, foundations, superstructures, electrical systems, heating and air-conditioning systems,
fabrication activities, acoustics, and so on). The flow not only becomes jumbled but also
complex and managing such a large set of activities requires different methods of planning and
control.
Due to a high degree of customization, the degree of uncertainty in all the above cases is likely
to be signifi cant. Planning tools that incorporate uncertainty are essential in managing such
situations. Second, the number of entities involved in accomplishing the various tasks is
numerous. Furthermore, these entities have several complex interactions amongst themselves.
The performance standards of one entity and deviations from the specifications at one part of the
system have a significant impact on the others.
Good scheduling practices could minimize the capacity-related problems in jumbled flow systems.
The operational complexity arising out of such a jumbled flow is high. The most significant
issue is one of capacity management. Considerable time is lost due to the repeated set-up of
processes. Since each customer order requires different levels of capacity, the estimation of
capacity that is already committed and that which is remaining is very hard to make. Good
scheduling practices could minimize such capacity-related problems. The other significant issue
is managing operations in the system. Due to jumbled flow, criss-crossing of jobs in the system
results in poor visibility. Problems are often hidden and a build-up of work in the process
inventory takes place. Therefore, good process inventory management practices are essential.
Cost accounting and estimation systems are crucial for job shops as there is a constant need to
quote for specific customer orders.
FIGURE 9.6 Process flow in job shops
Table 9.1 summarizes the salient features of various process flow alternatives and the key
operations management issues relevant to the above discussions. It is evident from the table that
continuous flow systems are relatively easier to plan and control. Capacity planning and
scheduling are far simpler than the other two categories. At the other extreme, jumbled flow
systems pose complex challenges to operations management. Therefore, in the past 20 years,
manufacturing organizations the world over are making concerted efforts to streamline their flow
and look for alternative methods of converting job shops. Similarly, firms are also investigating
methods of implementing just-in-time (JIT) systems as they seek to create a streamlined flow
of components and products in a manufacturing system.2
Just-in-time (JIT) systems seek to create a streamlined flow of components and products in a manufacturing
system.
So far we have discussed various aspects of the alternative process choices that an operations
manager possesses and the relationship between process choices and flow patterns in a
manufacturing system. The culmination of all these ideas is the process–product matrix. A
process–product matrix depicts all these relationships in a compact form, as shown in Figure
9.7.3
FIGURE 9.7 The process-product matrix
Source: Adapted from Hayes, R. H. and S. C. Wheelwright, “Link Manufacturing Process and Product Life Cycles,” Harvard Business Review 57(1), (1979):
133–140. Reproduced with permission.
One dimension of the matrix represents product characteristics and the other process
characteristics. Product characteristics essentially indicate the degree of customization and the
volume of production. On the other hand, process characteristics indicate the complexity and
divergence in the process. When organizations have a high volume of production, the flow will
be streamlined. Similarly, when the variety is high, the flow will be jumbled. As shown in the
figure, these two situations represent the two extremes in the figure and are not applicable. Each
combination has unique operations management issues to be addressed, as we have already seen.
Imagine yourself visiting a multi-specialty hospital for a master health check-up. What if the
radiology department was located in the second floor, the general physicians were sitting in the
ground floor at the rear side, ECG and treadmill test facilities were located in the fourth floor and
so on. Finally, imagine that you need to walk out of the main complex and go 50 metres away to
another building to have your breakfast after you give your fasting food samples and return to the
main complex to continue with the process. Such instances are not uncommon. How many times
have you felt that you were made to walk too much in a hospital when you went for a health
check-up or when you went to a financial institution asking for a loan sanction or to a
government office to pay some utilities bill or make some inquiries? What is the core problem in
these examples? How many times you have noticed that components travel excessively in a
factory before they complete their processing requirements? In simple terms, these examples
suggest that, with a better arrangement of resources, it is possible to improve operations and
provide better service to the customers.
Layout planning in manufacturing and service organizations involves the physical arrangement of various
resources available in the system to improve the performance of the operating system, thereby providing better
customer service.
The relationship between “volume– variety–flow” provides crucial inputs to the layout problem.
Over the years, operations management researchers and practitioners have evolved four major
types of layouts. These include product layout, process layout, group technology layout and fixed
position layout. The salient aspects of these layouts are described here.
As the flow becomes more cumbersome, the type of layout may significantly influence the ability of the
operations manager to effectively plan and control operations on the shop floor.
Process Layout
A process or a functional layout is an arrangement of resources on the basis of the process
characteristics of the resources. Consider a machine shop consisting of lathes (L), grinders (G),
milling machines (M), and drilling machines (D). A sample process layout for this shop is shown
in Figure 9.8. In the example, components belonging to Product A first visit a lathe, then a
drilling machine, a milling machine, and finally a grinding machine. The sequence of visits is a
function of the process plan and is available in a route card. The major implication of this design
is that when the number of components manufactured is large, there will be enormous criss-
crossing in the shop, as components need to visit machines in multiple combinations. This
increases material handling and poses challenges for production control.
FIGURE 9.8 Example of a process layout
A process layout is an arrangement of resources on the basis of the process characteristics of the resources
available.
Each department in a process layout is typically organized into functional groups. Thus, all
the lathes will be organized into a lathe department. Similarly, there will be a drilling
department, a milling department, and so on. In the fabrication area a similar arrangement would
be a welding department, a fitting department, a shearing department, and so on. All
manufacturing support areas are also arranged on a functional basis. Examples include a
maintenance department, quality control department, procurement, stores, and production
planning.
Product Layout
A product layout is an alternative design for the arrangement of resources. In this case, the order
in which the resources are placed exactly follows the process sequence dictated by a product. For
the above routing information the product layout will be as shown in Figure 9.9. In a product
layout, the required set of resources for every product is dedicated. Due to this, the resources are
arranged in the order of the machining requirements resulting in smooth component flow in the
shop. Since each product will have its own set of resources, material handling is simpler and it is
possible to invest in fixed-path material-handling systems to speed up material transfer between
successive workstations. The production control issues are much simpler in a product type of
layout compared to the process layout.
In a product layout, the resources are placed to follow exactly the process sequence dictated by a product.
Very often the final assembly in several manufacturing plants follows a product layout. The
assembly workstations are designed in such a manner that at each workstation a part of the job is
completed. The feeder stations are linked to the assembly workstations to ensure material
availability. As the product moves through the assembly, the process is completed. Testing, final
inspection, and even packing could be part of this layout, so that at the end of the line it is ready
for dispatch to the market. If you visit the Chinchwad plant of Tata Motors at Pune, you will
notice that there are nearly 180 workstations involved in the final assembly of the Tata Indica.
As the chassis moves through these stations, the car is assembled in a progressive fashion, and at
the end the car is ready for dispatch. In product layouts, the design of the layout is far more
important as it results in a certain degree of operational control. Once the design of individual
workstations is properly made, many problems are solved.
The notion of product and process layouts applies not only to manufacturing settings but also
in a service setting. A classic example is the banking system. For several years, the layout in the
banking operation was process based with the resources arranged on the basis of their function
they do. Thus, we have savings bank clerks, current account clerks, cashiers, tellers and so on.
However, in recent times, the layout is more product or customer based. In this arrangement,
manpower and resources are organized on the basis of customer segments that need to be catered
to. Figure 9.10 shows alternative layout choices between Bank A and Bank B.
In the case of Bank A, customers arrive at the waiting area and a member of the customer
relationship staff attends to all the requirements of the customer. In contrast, in the case of Bank
B, the staff is organized on the basis of functions. Depending on the nature of work involved,
customers will visit different sections in the bank and get their services done.
Both process layouts and product layouts have certain advantages and disadvantages.
Therefore, one needs to choose an appropriate layout depending on the specific circumstances.
Table 9.2 lists some of the salient advantages and disadvantages of both these types of layouts. A
product layout simplifies production planning and control problems. It also enables smooth
production and high production rate. However, since a high level of dedication of resources is
required it may increase the cost of operation and may be an unfeasible option in most cases.
Only in the case of high-volume manufacturing can an organization justify the dedication of
several resources.
FIGURE 9.9 Example of a product layout
Moreover, process layouts promote functional silos in organizations, which are considered to
be undesirable today.4
On account of these issues, another type of layout that combines the benefits of both process
and product layouts is being widely adopted by organizations today.
Group technology (GT) is a philosophy that seeks to exploit the commonality in manufacturing and uses this as
the basis for grouping components and resources. The implementation of GT is often known as cellular
manufacturing.
Group technology (GT) layout provides an alternative method for configuring resources in
organizations that have mid-volume, mid-variety product portfolios.5
GT is a philosophy that seeks to exploit commonality in manufacturing and uses this as the
basis for grouping components and resources. The implementation of GT is often known as
cellular manufacturing. In cellular manufacturing, the available components are grouped into
part families. An appropriate measure for manufacturing similarity is used to identify part
families. Corresponding to each part family, machine groups are identified and the layout is
formed accordingly. An example of a group technology layout for the process layout is shown in
Figure 9.11.
The benefits of GT are many. Once the part families and the machine groups are identified, the
layout ensures that each cell has only a certain number of components to be processed. In
essence, it is akin to breaking a monolithic structure into smaller, more manageable, and
independent units of production. Because of this, production planning and control become much
simpler. The components seldom travel outside their respective cells for processing. Therefore,
material handling becomes easier and traceability improves. Moreover, employees will be able to
relate better to their workplace and make concerted improvements in the process. The new
structure will also help to implement several other operations management practices such as
small group improvement activities, kaizen, and just-in-time manufacturing practices.
Fixed Position Layout
There are several situations in which the product manufactured is very bulky, difficult to move
and is often made in quantities of one. In such situations, the layout design ought to be very
different. One cannot employ any of the layouts described so far. Typical examples include
building very large machine tools and equipment, ship and aircraft building, and preventive and
breakdown maintenance of such large systems. Since such equipment is very large and bulky, it
dictates several choices with respect to the layout. The orientation of the equipment will dictate
the placement of specific resources required for the process. Layout planning in such cases is
often a question of a good workplace organization. By proper workplace organization it is
possible to exercise greater control on the process and remove unwanted processes and delays
from the system.
For reasons mentioned above, fixed position layouts are typically employed in large-project
type organizations. Some examples include the nuclear engineering division of BHEL at
Tiruchirapalli, the final assembly hanger of the Advanced Light Helicopter manufactured at
Hindustan Aeronautics Limited’s (HAL) helicopter division at Bangalore, and the Indian Space
Research Organisation (ISRO) launch site at Sriharikota. The emphasis in fixed-positioning
layouts is not so much on optimum positioning of resources required for the process. On the
other hand, it should enable managers gain better control of material flow (see Ideas at Work 9.4
for an example).
The layout design should result in greater efficiency of the manufacturing system. This is the
simple goal of any layout planning exercise. However, in order to assess how good the layout is,
it will be useful to employ a few measures. Such measures help a layout designer not only to
choose between alternative layout choices that he/she may have but also to set some targets for
improvement in the future. Table 9.3 has a sample set of measures and the basis for
measurement.
The most fundamental measure for layout design is the distance travelled by the jobs on the
shop floor. Often organizations use a scale model of the existing layout and strings and pins to
trace the path of the jobs to measure the total distance travelled. A good layout design must
ensure that unnecessary movements and excessive material handling are avoided. Table 9.4
shows the impact of an existing layout in a manufacturer of earth-moving equipment with four
product lines. Production and assembly of 1,080 components of Product A on average resulted in
a total travel of 375.7 km. Products C and D were added much later in the factory. The average
distance per item for all the four product lines confirmed that the company had not made any
major layout changes in the last thirty years, despite adding new product lines. As products travel
long distances, the lead time goes up, the overhead costs increase, and the investment in
inventory also goes up.
The most fundamental measure for layout design is the distance travelled by the components pertaining to a
customer order in the system.
Note: *The total distance travelled includes only those of the items manufactured on the shop floor. The number of items that were finally assembled into the
final product includes many bought-out items in addition to these.
In addition to measures to assess the impact on lead time and inventory, another useful
measure is the ease of production control. A better layout will enable an organization to have
greater visibility of operations and ease of control. Finally, recent ideas pertaining to business
process re-engineering suggest that if the jobs change fewer hands, then there are greater
advantages from pinpointing responsibilities on groups of people. Jobs with fewer
interdepartmental moves and ownership changes have a greater chance of getting completed
faster. Therefore, these measures may also be useful.
The design of a process layout involves a three-step process. The first step is to identify the
number of departments required and the space requirements for each department. The second
step is to estimate the magnitude of flow across these departments. The last step is to use the
above information to arrive at the final layout of the departments so that the cost of material
handing is minimized. These steps are explained here:
The qualitative measure links some criteria to the closeness required between a pair of departments.
Using the closeness ratings between departments, layouts can be constructed. In the above
example, Departments 1 and 3 are to be located close to each other. Similarly, Departments 1 and
6 should also be close to each other. Using this information, an initial layout can be constructed.
Accommodating all the requirements may often turn out to be infeasible. However, alternative
combinations may ensure better locations that improve the appropriateness of the layout.
Therefore, one method is to begin with an initial layout that is feasible and progressively check
the appropriateness of the new layout obtained. This may be done by selectively exchanging
parts of the layout plan through swapping of a subset of locations among each other. A pair-wise
exchange could be a systematic approach to enumerate all the possible layouts that one can get
from an initial layout.
There are software packages available for the qualitative design of process layouts. These include ALDEP and
CORELAP.
Cij denote the cost per unit of transporting a unit distance from Department i to Department j
Fij denote the interdepartmental flow between Department i and Department j
Dij denote the distance between Department i and Department j
n denote the number of departments to be laid out (9.1)
The interdepartmental flows, the distance between the blocks, and cost information are gathered
during a layout planning exercise, as indicated in the previous section. Therefore, once the
decision regarding which department will be located in which block is made, the total cost of the
plan could be computed using Eq. 9.1. The use of the quantitative method requires some logic
for arriving at the best layout. The above can be modelled as a mathematical programming
problem with the objective function of minimizing the total cost of the plan. The mathematical
programming problem is one of assigning the department to the blocks, subject to the constraint
that one block can hold only one department. Further, the number of blocks assigned should be
equal to the number of departments.6
A popular heuristic for the assignment problem forms the basis for the computerized
procedure known as computerized relative allocation of facilities (CRAFT). CRAFT utilizes
the improvement method for obtaining the best layout. In CRAFT, an initial feasible layout is
formed and a series of improvement opportunities explored through a pair-wise exchange of
departments. If there are n departments, a pair-wise comparison involves n(n − 1)/2 evaluations.
After all these evaluations, the best possible pair-wise exchange is identified. If further reduction
in the cost is possible by making this change, then the location of the two departments is
changed. For example, in a six-department example, let us assume that the cost reduction by
pair-wise exchange of departments is as shown in Table 9.7 for four pairs (we shall further
assume that all other pair-wise exchanges result in an increase in material-handling costs).
Computerized relative allocation of facilities (CRAFT) is a computerized procedure that utilizes the
improvement method for obtaining the best layout.
Therefore, the location of Departments 3 and 5 will be exchanged. If the total cost of material
handling at the beginning of this iteration was ₹20,000 the new material-handling cost will be
₹19,650. The procedure continues in this manner until no more improvement is possible.
There are three underlying approaches to the use of computer packages for designing layouts: the construction
method, the improvement method, and performance evaluation modelling.
On the other hand, in the improvement method, the starting point for a solution is a feasible
layout. The feasible layout may not be the best layout. Therefore, the solution is improved
progressively. When no more improvement is possible, the procedure stops and the available
solution is the best solution. There are two known packages available in this category. These are
computerized relative allocation of facilities (CRAFT) and computerized facilities design
(COFAD). COFAD is a modification of CRAFT that incorporates the choice of a material-
handling system in the layout design process.
The third approach to layout design is to use recent advances in object-oriented modelling
techniques and build performance evaluation models. In this approach, software is used to build
a model with 3-D objects of the layout that the designer has in mind, and to assess the impact of
the design using several parameters. These parameters could include, for example, equipment
and material-handling utilization, congestion in the material-handling network, build-up of
inventory in the shop floor, and manufacturing lead time. Often, after building these models, a
simulation analysis of the layout could be done to assess the parameters listed above. Since the
cost of developing and analyzing these models has been progressively coming down, recent
approaches to designing layouts are directed towards arriving at a configuration through
performance evaluation.
There are several examples of mass production systems in operation today. A two-wheeler
manufacturer such as Bajaj Auto Limited sold nearly 3,422,400 motorcycles in 2013–2014,
which translates into a monthly sale of 285,200. It is expected that the daily production of
vehicles is nearly 11,400. Various sub-assemblies in the Bajaj plant need to be configured to
match the production rate. Similarly, the final assembly stations also need to have the required
number of resources at each station to meet the targeted demand. In such a scenario, much of the
control and scheduling boils down to appropriately arriving at a balanced flow of components on
the shop floor. A product layout design essentially caters to this objective by estimating the exact
number and the sequence of resources required in the manufacturing system for a targeted
production level. Essentially, the layout design seeks to identify the minimum number of
resources required to meet a targeted production rate and the order in which these resources are
to be arranged. In the process it seeks to establish a balance among the resources so that
production is smooth. Therefore, the technique employed for this purpose is known as the line-
balancing technique.
Let us consider a mass production system with multiple tasks such as the manufacture of
computers. If you visit the final assembly shop of the plant, you will invariably find a line layout.
At one end of the assembly components will be fed into the system in multiple stations and the
computer will be progressively assembled. Initially the mother board will be assembled with the
required components. As it proceeds, the hard disk, optical drives, fans, power systems, serial,
parallel and USB ports, and SMPS will be added to the box before it is finally secured with the
cover plate. If you visit other mass producers like a car manufacturer such as Hyundai or an
industrial valve manufacturer such as Audco India Limited you will notice a similar structure of
manufacturing and assembly process. In all these cases, you will notice multiple tasks and
precedence relationships among them.
Line balancing is a method by which the tasks are optimally combined without violating precedence constraints
and a certain number of workstations are designed to complete the tasks.
Each task i requires a finite time, denoted by ti and may have some precedence relationships
with the other tasks. Further, the tasks require certain skills and resources such as machines and
tools. Line balancing is a method by which the tasks are optimally combined without violating
precedence constraints and a certain number of workstations are designed to complete the tasks.
If there are three workstations, A, B, and C, in which 7 tasks are performed in a manufacturing
system, then the workstation times are nothing but the summation of the task times assigned to
each workstation. Let the workstation times be denoted as WA, WB and WC. Clearly, a balanced
design is one in which the workstation times do not vary widely. In such a situation, the
resources will be uniformly utilized and the flow of material will be even. Further, there will be a
good rhythm in operations. The maximum of the three workstation times determines the interval
between the production of two successive components. If WA = 5 seconds, WB = 3 seconds and
WC = 7 seconds, one component will come out of the system every 7 seconds. This measure is
known as the cycle time.
Cycle time could be considered as the reciprocal of production rate. If in a period of 20,000
seconds a shop produced 10,000 pieces of a component, then the production rate is 1/2 per
second. Conversely, the cycle time is 2 seconds. Cycle time could be actual or desired. If we
compute on the basis of actual production, it represents the actual cycle time. On the other hand,
if we compute the cycle time on the basis of what we desire the production to be, then it is
desired cycle time. Maintaining the desired cycle time requires better management and work
practices. The expression for cycle time is as follows:
The problem of designing a balanced set of workstations suffers from the classical trade-off
issue. If we combine more tasks into fewer workstations, we may require fewer workstations but
the cycle time will be high, leading to reduced production rate. At the other extreme, if the tasks
are kept separate and as many workstations are designed, we may increase the production rate
beyond what is required at the risk of deploying more resources and workers with poor
utilization. Therefore, solving the line-balancing problem calls for counterbalancing these
opposing costs and striking the right trade-off between increased production and better utilization
of resources.
The production rate, cycle time and the number of workstations are interrelated. Moreover, the
number of workstations has a bearing on the average resource utilization. Using these measures
it is possible to design an appropriate number of workstations. These relationships are indicated
in the following expressions:
EXAMPLE 9.1
A factory working in two shifts each of 8 hours produces 24,000 electric bulbs using a set of
workstations. Using this information, compute the actual cycle time of the plant operation.
There are 8 tasks required to manufacture the bulb. The sum of all task times is equal to 12
seconds. How many workstations are required to maintain this level of production assuming
that combining of tasks into those workstations is a feasible alternative?
Solution
Available time = 2 × 8 × 60 × 60 = 57,600 seconds
Actual production = 24,000 electric bulbs.
Therefore, using Eq. 9.2 we compute the cycle time for each bulbs as
This means that the factory is producing 1 bulb every 2.4 seconds.
Therefore, the tasks are to be split among the five stations such that each workstation will
have 2.4 seconds as the sum of its task times.
EXAMPLE 9.2
A computer manufacturer needs to design assembly stations in the factory where the cabinet
housing the hard disk, mother board and other accessories is to be made. The factory
currently works for one shift of 8 hours. The tasks and their durations are given in Table 9.8.
Figure 9.14 shows the precedence relationship among the tasks.
a. If the cycle time is 80 seconds, what will the daily production of cabinets be?
b. If the desired production rate is 320 cabinets per day, what is the maximum permissible cycle time?
c. What is the maximum and minimum number of workstations required to maintain this daily production rate?
d. Design an assembly set-up with 5 workstations and 6 workstations. What are the key inferences of this exercise?
(b) Since the desired production rate is only 320 cabinets, one can obtain the maximum permissible cycle time for the
assembly stations.
The computations show that while a cycle time of 80 seconds may yield a much higher
production rate, the workstations can be designed for a cycle time of up to 90 seconds
without falling short of the desired daily production. Once the cycle time crosses 90 seconds,
the manufacturer may not be able to produce 320 cabinets every day.
We know that the maximum number of workstations required to maintain the desired daily
production rate is equal to the number of tasks that are to be performed. This is because we
can simply assign one task to one workstation. In our example, the maximum number of
workstations is eight, as there are eight tasks.
In our example, the sum of all task times is 380 seconds. Therefore, combining all the
tasks into a single workstation will result in a cycle time of 380 seconds. This is not
permissible. Therefore, the minimum number of workstations is dictated by the maximum
cycle time permissible.
(c) The minimum number of workstations is given by Eq. 9.3 as
As long as we have five workstations, our cycle time may not exceed 90 seconds and we will
be able to meet the production rate.7
(d) Design of assembly shop with five workstations (Table 9.9 presents the details)
We assign tasks to the five workstations on the basis of the following two criteria:8
i. The workstation times should not exceed the maximum permissible cycle time of 90 seconds
ii. The precedence relationships among the tasks need to be honoured
The average utilization of the resources can be computed as follows (Eq. 9.4):
(e) Design of assembly shop with six workstations (Table 9.10 presents the details):
We assign tasks to the six workstations using the same set of criteria as before.
Key Inferences
The above computations clearly show the trade-off issues in line balancing. A design with
five workstations will meet the current production rate with an average utilization of 84.4 per
cent. However, when we increase the number of workstations by one, the utilization comes
down by 4.8 per cent, but the production rate could increase by 11.1 per cent (since the daily
production rate will increase from 320 to 360 on account of cycle time coming down to 80
seconds from 90 seconds). There are cost and strategic implications of these two alternatives
and management could take an appropriate decision at the time of finalizing the design.
GT layouts are designed with the objective of subdividing a universe of machines and
components into sub-groups such that each sub-group of components forms a part family and is
endowed with a corresponding sub-group of machines known as a machine group. GT layout
design is done with the systematic analysis of a machine-component incident matrix.
The resultant structure obtained by reconfiguring a manufacturing system with GT layout is known as a cellular
manufacturing system.
This structure of the matrix is known as a block diagonalized structure. The advantage of a
block diagonalized structure is that it enables us to directly read the cells on visual inspection. In
this case, four cells are formed with part families and machine groups as shown in Table 9.11.
A vast number of methods are available for identifying an appropriate set of sub-groups. The
earliest and most popular among these is the production flow analysis (PFA) proposed by
Burbidge. Later, the use of clustering techniques, matrix manipulation methods, graph theory,
and mathematical programming were proposed by a number of researchers.9 The rank order
clustering (ROC) method uses simple matrix manipulation methods to identify the cells in an
MCIM. The ROC method rearranges the rows and columns of the MCIM in an iterative manner
that will ultimately, and in a finite number of steps, result in a matrix form in which both the
rows and columns are arranged in order of decreasing value when read as a binary word.
TABLE 9.11 The Four Cells with Part Families and Machine Groups
The earliest and most popular method for design of cellular manufacturing systems is the production flow
analysis (PFA) proposed by Burbidge.
Earlier, machines were usually operated manually. Therefore, every machine needed to be
attended by one worker. However, today, manufacturing organizations use semi-automatic
machines and fully automatic machines. By semi-automatic machines, we mean those with
automatic processing capability, but requiring a human interface for the loading and unloading of
components into the machine. On the other hand, in a fully automatic machine, the machine has
automatic processing capability and is linked with an automated material handling and material
transfer mechanism. In this case, as soon as processing in one job is complete the material-
handling system will shunt the completed job out of the machine into a material storage system
or a conveyor (or some form of material-handling system) and will load the next job into the
process.
One worker–multiple machine layouts promise significant reduction in manufacturing lead time, better space
utilization, and cost reduction due to better utilization of resources and manpower.
Figure 9.17 is an illustration of the nature of transformation and benefits that organizations
obtain by employing one worker–multiple machine layouts. This example is one of several such
cells formed in the manufacturing facility of Lucas TVS at Padi in Chennai. As is evident from
the figure, the layout promises significant reduction in manufacturing lead time, better space
utilization, cost reduction due to better utilization of resources and manpower, improvement in
quality due to reduced material handling and, finally, a reduction in the distance travelled by the
jobs before the completion of processing.
Batch manufacturing has always had inherent limitations on account of the mid-volume, mid-
variety nature of manufacturing. Work-in-process levels are generally high and machine
utilization tends to be low. Jobs spend a high proportion of time waiting for a machine to be set
up, waiting to be moved or waiting for other jobs on the machine to be completed. Batch
production often requires an army of progress chasers to keep jobs flowing through the
manufacturing facilities. In batch type manufacturing, some studies conducted revealed that only
5 per cent of the total time spent in the shop contributed to actual metal cutting operation. The
rest accounted for wastage of capacity, unnecessary waiting of jobs and so on. One way to
improve productivity is to use technology to obtain a better process.
An FMS is a manufacturing system consisting usually of numerical control (NC) machines connected by an
automated material-handling system. It is operated through central computer control.
Advancements in the field of manufacturing technology have had an impact on process design.
Versatile machines that are capable of handling the work of several traditional machines have
provided some relief in simplifying process flow. These machines also provide considerable
flexibility to address some of the issues pertaining to mid-volume, zmid-variety manufacturing
systems. Of the several developments, flexible manufacturing systems (FMS) have been most
prominent. The development of manufacturing technology has been accompanied by
developments in material-handling technology, which have also contributed to process
improvements. We shall look at these developments in this section.
A schematic representation of the structure of an FMS is given in Figure 9.18. Machine tools are
the main machines used in manufacturing while auxiliary equipment are those used for other
purposes, such as measuring quality using coordinate measuring machines. The primary MHS
transports the work pieces from one location to another in an FMS. On the other hand, the
secondary MHS transports work pieces from the machine table to the storage point in front of the
machine tool.
Two general classes of FMS technology have been identified. The most recognized form is the
multi-machine tool type, sometimes called the classical FMS. This is an automated production
system for mid-volume, mid-variety production, consisting of machine tools tied together by an
MHS. The second type of FMS is the standalone unit, usually consisting of a single machining
centre or CNC lathe. This is sometimes referred to as a flexible manufacturing cell (FMC). Such
FMCs are equipped with automatic carousels or rotary tables for loading a small variety of work
pieces. In some cases, an integral robot arm is employed for loading and unloading chores.
Figure 9.19 shows typical machines used in flexible manufacturing systems.
FIGURE 9.18 Structure of an FMS
FIGURE 9.19 Typical machines used in FMS
FMS technology results in the reduction of direct and indirect labour force. With the level of
automation that is employed in such systems, it is possible for a worker to tend to a group of
machines in the system. The following is the role of human labour in the FMS:
Loading and unloading, tool set-up, tool replacement, work piece set-up off line
Maintenance of the system, multi-task monitoring (3 to 8 machines)
Supervision of the overall system, taking decisions using the information supplied by the computer system etc.
The technologically advanced features of the FMS, in part, simplify process design and
complexities in flow in an intermittent flow system in batch manufacturing by offering several
flexibilities, some of which are as follows:
Machine flexibility: This is related to the ease of making changes required to produce a given set of part types. The
measurement of these changes includes, for example, the time taken to change tools in a tool magazine to produce a
different part family, the time taken to assemble new fixtures, and so on. Technological progress, proper operation
assignment and the technological capability of bringing part and tool together, along with other such factors, will
improve this flexibility.
Process flexibility or mix flexibility: This is the ability to produce a given set of part types, each possibly using different
materials in several ways. This, in essence, relates to the mix of jobs, which the system can process simultaneously.
Process flexibility increases as the machine set-up cost decreases, and by having multipurpose adaptable CNC machines.
Product flexibility: It is the ability to produce a new set of products economically and quickly. Product flexibility can be
measured from the time required to switch from one part mix to another.
Routing flexibility: This is the ability to handle breakdowns and to continue processing the given set of part types. This
ability exists if either a part type can be processed via several routes or each operation can be performed by more than
one machine.
Volume flexibility: This is a measure of the ability to operate an FMS profitably at different production volumes. This
flexibility can be measured by how small the volumes can be for all part types with the system still being run profitably.
Expansion flexibility: It is the capability to build a system and expand it as the need arises, easily and in a modular
fashion.
AGVs can also communicate with factory computer control systems, either by making a physical
link at each stopping point, or by infrared or digital radio transmission. This provides a means to
track and control the flow of materials and alters the instructions to the guided vehicle. The
central computer provides overall control functions such as dispatching, routing, traffic control,
and collision avoidance. The recent advancements in AGVs include SGVs (self-guided vehicles)
and free-ranging vehicles. AGVs usually complement an automated production line consisting of
conveyors or transfer systems by providing the flexibility of complex and reprogrammable
movement around the manufacturing process.
Assembly-line AGV applications are being introduced in manufacturing systems. These
guided vehicles carry major sub-assemblies such as motors or transmissions to which parts are
added in a serial assembly process. The vehicle proceeds into an assembly area and stops at an
assembly workstation. When the process is completed, it enters the next assembly area, and this
process is repeated several times. When the assembly process is complete, the finished assembly,
say an engine block or a chassis, is unloaded from the vehicle, which is then sent to the start area
for the assembly process. One such application was found at the Tata Motors plant in
Jamshedpur. It uses an AGV system for rear axle assembly. The assembly line consists of 18
stations and the AGVs progressively move through these stations.
Conveyors exist in a wide variety of forms such as overhead, monorail, carry-and-free, power-
and-free, underfloor drag-chain, floor-slot, gravity-feed, belt conveyors, and plastic link chains.
All these systems are readily available in a wide variety of sizes, load-carrying capacities and
speed capabilities. The intelligence for a conveyor system is provided by placing a large number
of sensors around the conveyor and connecting them. It is this sensor and gating or routing
technology that often complicates the implementation of a sophisticated conveyor system. These
have a considerable impact on the overall cost of the system.
The use of a robot for materials handling can be beneficial for a number of reasons. The
industrial application of robots is wide-ranging, particularly in assembly operations. In addition,
robots are predominantly used as secondary material-handling systems in FMS. They typically
perform most of the loading and unloading chores at each machining centre or buffer location. In
an effort to realize most of the advantages of robots and conveyors, the gantry loading system
has been developed. Various types of gantry loaders are available.
FMS designers, over the past few years, have developed a rail-guided transport system, which
is effectively a compromise between a conveyor system and an AGV. It is a vehicle similar in
appearance to the AGV, but is constrained to run on rails. If an FMS can be arranged in a straight
line, a rail-guided transport mechanism could well be an attractive means of load transfer
between the various locations on an assembly line. Towline systems consist of carts that are
powered either by an in-floor chain or by an overhead chain. The control systems available today
make it feasible to programme tow carts and send them from one workstation to another.
Towline systems are generally most advantageous when utilized in a relatively fixed path and in
a consistent stations situation.
From the foregoing discussion on alternative flow patterns and process design alternatives for
manufacturing systems, factors that introduce complexity in operations management can be
identified. Figure 9.20 summarizes these factors and their effects.
FIGURE 9.20 Factors affecting the complexity of operations management
First is the issue of flow pattern in a manufacturing system. As the flow pattern deteriorates
from a continuous and streamlined pattern to intermittent and jumbled, we find that the
complexity of operations management increases. In the case of project organizations, operations
management needs to deal with long lead times and a high degree of uncertainty. The other issue
is the number of stages in the production, irrespective of the process flow pattern. As the number
of stages increase, the number of options to be considered increase, along with the size of the
problem, leading to complex operations management issues. To give a simple example,
scheduling of operations in a shop with 10 machines and 20 sequences of parts is much easier to
handle than scheduling of 20 machines with 40 process sequences. Similarly, predicting the
nature of disruptions, such as machine breakdown and tool breakdown, is relatively easier in a
shop having 20 machines than in one with 40 machines.
Finally, the most significant aspect of operations management complexity arises from variety.
Variety can manifest itself in several ways in a manufacturing system. This includes variety in
product lines, product families and specific variations of the product, variety in process routing
that could be employed for manufacturing, and variety in the type of technology used for
manufacture of the product. All these add to the complexity of operations management, as seen
in this chapter.
SUMMARY
Volume, variety and flow exert significant influence on process design in organizations.
Process industries and mass production systems generally have a streamlined flow of products.
Mid-volume and mid-variety manufacturing systems have intermittent flow. Capacity estimation is difficult in such
systems compared to continuous flow systems.
Project organizations and customized manufacturing systems have jumbled flow. Capacity estimation and scheduling of
jobs are quite difficult. Therefore, operations management complexity is high in jumbled flow systems.
A process–product matrix depicts the relationship between process flow characteristics and volume of production in an
manufacturing organization.
Product layouts are useful for high volume–low variety situations. At the other extreme, fixed positions and project
layouts are useful for high-variety situations.
Product layout and process layouts are used in the discrete manufacturing industry. They have several advantages and
disadvantages.
Mid-volume and mid-variety manufacturing systems can benefit from a group technology (GT) layout.
Several software packages are available for designing process layouts. CORELAP, ALDEP, COFAD, and CRAFT are
among the popular packages in this group.
Product layout design seeks to identify the minimum number of resources required to meet a targeted production rate
and the tasks to be assigned to each of these resources, using a technique called line balancing.
GT layouts are designed with the objective of subdividing a universe of machines and components into sub-groups such
that each sub-group consists of part families and machine groups.
A number of methods are available for designing GT layouts. These include production flow analysis (PFA), rank order
clustering (ROC), mathematical programming and clustering methods.
New manufacturing technologies such as flexible manufacturing systems (FMS) have the potential to simplify flow
complexities in mid-volume and mid-variety manufacturing organizations due to increased flexibility.
No. of stages, variety and flow characteristics determine the complexity of operations management. By a careful design
of the process, some of the complexities can be minimized.
FORMULA REVIEW
The total cost of the process layout,
REVIEW QUESTIONS
1. What are the key determinants of process characteristics in operations? Give specific examples to show how they influence
process characteristics.
2. Is there any relationship between volume, variety and flow with respect to process design?
3. How is the continuous flow system different from the intermittent flow system? What are the key implications for operations
managers?
4. Give three examples for each of the following and identify two important issues of interest to an operations manager:
5. As an operations manager, which areas would you focus on if you were to manage a jumbled flow process?
6. How does the choice of the layout affect the operational performance of an organization?
7. What is the relationship between the volume, variety and flow characteristics of an operating firm and its layout?
8. What are the different types of layouts? How should an organization decide on which layout to choose?
9. Is a group technology layout any different from a product layout? Give reasons for your answer.
10. Identify an appropriate layout for each of the following situations. Justify your choice in a sentence or two:
11. Suppose you are given three alternative designs for the layout of a shop floor in a manufacturing organization. How will you
decide which of the three is the most appropriate?
12. Critically compare quantitative and qualitative approaches to designing process layouts.
13. How does a construction method of layout design differ from the improvement method? Which method is preferable and
why?
14. What do you understand from the phrase “cellular manufacturing system”?
15. What are the reasons for the growing application of group technology concept for layout design?
16. What do you understand by the term FMS? What are the characteristic features of an FMS?
17. How does an FMS help in improving the process design in the case of mid-volume, mid-variety manufacturing?
18. What are the various types of automated material-handling devices available for use in a computer integrated manufacturing
system?
19. What do you understand by the term complexity of operations management? Explain.
PROBLEMS
1. A mass producer of an industrial component currently has a daily production rate of 1,200 pieces. The company works for 16
hours a day. The production process currently employs five stations. There are 12 tasks involved in the manufacturing process
and together they need 190 seconds.
2. A company procures a few components from the market and assembles them into a standard product and supplies it to its
customers. There is enough demand for the product. If the company employs four stations for the assembly, it has 92%
average utilization of the resources. The sum of the task time for the process is 180 seconds.
3. Following data are available in the design of a product-based layout in which a certain number of work stations are deployed.
The sum of all task times is 200 seconds. The cycle time is 50 seconds and the average resource utilization is 80%. If one
more work station is added, the utilization drops to 74.07% and the new cycle time is 45 seconds.
a. How many work stations are used in the product layout design?
b. What is the daily production rate if a day consists of 450 minutes?
c. What is the maximum production possible in the current design?
4. A manufacturer of spark plugs for the mass market would like to design the final assembly shop and requires certain data for
the process. The factory works for two shifts and the total available time is 15 hours per day.
a. There are five tasks involved in the final assembly and the task times (in seconds) are 3.5, 4.5, 2.0, 3.0 and 4.0. What
are the minimum and the maximum outputs possible from the factory?
b. The daily production needs to be 4,000 plugs. What should be the cycle time of the operations at the assembly shop?
c. If the firm wants to use three workstations, what will be the best output possible? By how much will the output increase
if it adds one more workstation?
5. A fast-food restaurant that operates in Connaught Place, New Delhi, specializes in offering food during peak hours. The
restaurant operates only for two hours in the morning and promises faster turnaround for customer orders. The restaurant
needs to design the workstations at the front end of the kitchen to assemble all customer orders. The back end of the kitchen
provides the required food items. A study of the tasks, their dependencies and times taken on an average to assemble reveals
the information shown in Table 9.12:
NET-WISE EXCERCISES
1. Caterpillar is the largest maker of construction and mining equipment. To know more about the products offered by them, visit
http://india.cat.com/cda/layout?m=81920&x=7 and click on Products. Navigate through their product lines and offerings.
Haldia Petrochemicals is the one of the largest petrochemicals complexes in India. Go to
http://www.haldiapetrochemicals.com/index.jsp and click on Products on the left side of the
screen. Visit various sub-links in this section. Also click on the link About Us. There is a
subsection on manufacturing processes. Visit other parts of the Web site. Take a look at the
photos of the plant loaded onto the main page.
Based on your understanding of the two organizations, compare and contrast the product flow
and process characteristics.
a. Identify three distinct features of Caterpillar and relate them to the design of the manufacturing process for the
organization.
b. Why would you classify Haldia Petrochemicals as a process industry? Identify three characteristics that you find in
Haldia Petrochemicals that are peculiar to a process industry.
MINI PROJECT
1. There are several manufacturing and service organizations in every part of the country. Select out of these three organizations
that belong to any of the following categories: a process industry; a mass production system; a service set-up; a mid-volume,
mid-variety manufacturer.
Visit the selected manufacturing and service organizations, make a formal plant tour of the organization and collect data on
various aspects of operations management. After you have recorded all the key observations based on the visit, prepare a list
of items on which you would like to collect additional information and obtain them. Also interview the plant manager, the
process planning team, and the line supervisor. Note down their key observations about the plant’s functioning. Based on this
work, prepare a final report that contains the following information:
2. There are alternative approaches to layout design. The use of software packages for this purpose has increased manifold in
recent years. Prepare a project report to answer the following question pertaining to the use of software for layout design:
a. Scan the Internet and prepare a list of avail-able software solutions for layout design. Briefly explain the salient features
of the software.
b. Develop a classification scheme to list the software that you have identified. Can you recommend a specific set of
application domains of the software on the basis of this classification scheme?
c. Select one software package from each of your classification schemes and critically compare them. Identify the specific
advantages and disadvantages of each package.
3. There are several companies that have recently made significant changes in their layout design. Prominent in the list are
automobile and auto-components manufacturing firms. Identify one such company in your locality and perform the following
studies and submit a report of your findings:
a. What were the major drivers of the layout change decision in the organization?
b. Enumerate the salient aspects of the alternative layout design process (methodology used to arrive at the new design,
tools and techniques used, scope of the project, organizational aspects of driving the change process etc.)
c. Understand the implementation aspects of the layout design process (methodology of implementation, time taken,
chronology of events leading to completion of the process etc.)
d. What were the main benefits that accrued to the organization as a result of this layout design change?
e. How did the layout change affect various people in the system including the production workers, supervisors,
managers, customers, and suppliers?
CASE STUDY
TABLE 9.13 Sub-assemblies and Machining Requirements for the New Product
Note: *41 components from the existing project have been added to avoid omission of these during capacity calculations. These components require special
purpose machines to be assigned for producing the new product.
TABLE 9.14 Details of Modules and Cells as Per the AM Approach for Redesign
Note: * Eight machines have been separately located as Cell 13. These machines were highly valued one-off machines and it was not possible to justify
either locating them in a particular cell or procuring additional machines. About 79 components from four modules undergo processing in the cell.
# Out of the 275 exceptions, only 92 required movement across the modules. In other cases, it involved inter-cellular movement between cells belonging to
the same module.
Production flow analysis was initially employed to identify cells. Forty cells were initially
formed. However, the demand on many machines in these cells was far less. Moreover, the
number of cells was too many. Using the same machine utilization cut-off value of 30 per cent,
the 40 cells were merged and finally 8 cells were formed. Table 9.15 presents the summary
details for this design. Finally, Mr Prakash presented a comparison of the two approaches based
on several criteria (see Table 9.16 for details).
The task before the executives at AIL was to identify an appropriate configuration for the
machine shop at Unit 1.
SUGGESTED READINGS
J. M. Apple, Plant Layout and Material Handling, 3rd edition (New York: John Wiley & Sons, 1977)
S. Benjaafar, S. S. Heragu and S. A. Irani, “Next Generation Factory Layouts: Research Challenges and Recent
Progress,” Interfaces 32, no. 6 (2002): 58–76.
J. L. Burbidge, “Production Flow Analysis for Planning Group Technology,” Journal of Operations Management 10,
no.1 (1991): 5–27.
T. H. Davenport, Process Innovation: Reengineering Work Through Information Technology (Boston, MA: Harvard
Business School Press, 1993).
R. L. Francis and J. A. White, Facilities Layout and Location: An Analytical Approach (Englewood Cliffs, NJ:
Prentice-Hall, 1977).
G. V. Frazier and M. T. Spriggs, “Achieving Competitive Advantage Through Group Technology,” Business Horizons
39, no. 3 (1996): 83–90.
R. H. Hayes and S. C. Wheelright, “Link Manufacturing Process and Product Life Cycles,” Harvard Business Review
57, no.1 (1979): 133–140.
P. Y. Huang, and C. S. Chen, “Flexible Manufacturing Systems: An Overview and Bibliography,” Production and
Inventory Management (Third Quarter, 1986): 80–89.
J. R. King, “Machine-Component Grouping in Production Flow Analysis: An Approach Using a Rank Order Clustering
Algorithm,” International Journal Of Production Research 18, no. 2 (1980): 213–232.
G. A. Koff, “Automatic Guided Vehicle Systems: Applications, Controls and Planning,” Material Flow 4, nos. 1 and 2
(1987): 3–16.
L. J. Krajewski and L. P. Ritzman, Operations Management: Processes and Value Chains, 7th edition (Upper Saddle
River, NJ: Pearson Education Inc., 2004).
B. Mahadevan and G. Srinivasan, Software for Manufacturing Cell Formation: Issues and Experiences. (Columbus,
Ohio: GT/CM World Symposium, 2003).
B. Mahadevan, and S. Venkataramanaiah, “Realigning Research Objectives in Cellular Manufacturing System Design:
A Users’ Perspective,” Technology, Operations and Management (2007), 17-29.
B. Mahadevan, The New Manufacturing Architecture (New Delhi: Tata McGraw-Hill Co., 1999).
S. Nahmias, Production and Operations Analysis, 3rd edition (Englewood Cliffs, N. J.: McGraw Hill International
Editions, 1997), pp. 561–596.
P. O’Grady, “Flexible Manufacturing Systems: Present Development and Trends,” Computers in Industry 12 (1989):
241–251.
R. W. Schmenner, Plant Tours in Production/Operations Management (Chicago: Science Research Associates Inc.,
1986).
R. J. Schonberger, Building a Chain of Customers (New York: Free Press, 1990), pp. 34–64.
C. Winarchick and R. D. Caldwell, “Physical Interactive Simulation: A Hands-On Approach to Facilities
Improvements,” IIE Solutions 29, no. 5 (1997): 34–42.
CHAPTER 10
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Is process design in service systems any different from that in manufacturing systems?
What factors significantly determine the design of service systems?
What do we mean by customer contact in service organizations? How does it affect the design of service systems?
What do we mean by service positioning? What are the factors that determine service positioning?
What is the basic structure of a waiting-line model?
How can one use the principles of queueing theory to analyse capacity-related issues in service and manufacturing
firms?
How do service firms address short-term capacity planning problems?
How is service quality measured?
The design of a service system and the choice of its layout are closely related to the degree of customer contact in a service
system. Blue Ginger, the Vietnamese restaurant at Taj West End, Bangalore, is designed with teak flooring, fire torches, mood
lighting, and waterfalls, in addition to several other features. Vietnamese entertainers in a show kitchen in the dining area offer a
unique dining experience.
ideas at Work 10.1
Designing operations systems typically involves making choices with respect to the location,
technology, capacity, and layout of the system. These decisions directly influence other aspects
of the operations, such as the people and skills to be deployed and the planning methodology to
be put in place. However, designing a service system is vastly different from designing a
manufacturing system. In a manufacturing system, the customer is seldom part of the process.
He/she merely places an order and awaits the delivery of the product. Issues such as what
resources are to be deployed, how they need to be laid out, how to link the successive stages of
the conversion process, and what kind of planning and coordination elements to deploy are all
made with this basic understanding. In contrast, in most service systems, customer participation
in the conversion process is inevitable. Therefore, it poses a different challenge altogether and
requires the designer to incorporate the customer as an integral element in the design process.
Consider setting up a restaurant in the city of Hyderabad. Assume that we have been able to
secure a 40,000 square feet independent building with some spacious land around it. When we
think of a good restaurant, several things come to our mind with respect to the design of the
system. Some of these include (a) a well-laid out dining space with tasteful decoration of the
surroundings and a good ambience, (b) the nature and breadth of offerings in the menu available
to the customer, and (c) the judicious use of technology and human elements so that the system is
responsive. From a service design perspective, achieving these outcomes essentially requires us
to answer some questions regarding the design of the restaurant. There are several issues that we
need to address in this process. Let us list some of them:
Designing operations systems typically involves making choices with respect to the location, technology,
capacity, and layout of the system. However, designing a service system is vastly different from designing a
manufacturing system. It requires the designer to incorporate the customer as an integral element in the design
process.
a. Are there alternatives available for us to position the service? What factors influence service positioning and what are
the implications of this on the overall service delivery design?
b. What is the nature and level of interaction that we must design between the service provider and the customer? What are
the implications of this?
c. What is the overall level of technology that we must use? How and where should we deploy the technology?
d. What are the elements of the front office and the back office of this service delivery system? How do we identify these
elements?
e. How should we design the dining area and the kitchen area? What are the factors that will influence the choice of the
design?
f. How should we lay out the service delivery system? Which are the areas that need greater attention with respect to look
and feel?
g. What should be the capacity of the dining area and the kitchen area? How can we estimate this, given a certain
uncertainty in the arrival pattern and the demand for restaurant services?
h. How should we measure and improve the quality of the restaurant?
This list is merely suggestive. It indicates the nature of decisions that one must make in the
design of a service system. One can develop similar sets of questions with respect to the design
of other service systems such as a hospital, an automobile garage, an after-sales service outlet, a
management consulting firm, or a travel agency.
Of the several factors that may influence our choice of various design elements in a service
system, the most fundamental one is the customer contact that we want to provide in the service
delivery system. Customer contact signifies the extent to which the customer participates in the
preparation and consumption of a service as well as the nature and intensity of interaction that
the customer has with the service personnel. It also indicates the level of exposure that the
customer has to the various facets of the service system while receiving the service. The degree
of customer contact relates to varying proportions of customer exposure to all these attributes.
Some examples will help clarify these ideas.
Customer contact signifies the extent to which the customer participates in the preparation and consumption of
the service as well as the nature and intensity of interaction that the customer has with the entities and service
personnel.
Let us consider an automobile garage with two different options with which the service
delivery mechanism could be designed. In Design A, the customer is received at the front portion
of the garage in a neatly laid out office. The customer assistant interacts with the customer, hears
his/her complaints and the issues to be addressed while servicing the car, records them in the
work sheet, and takes his/her approval. The customer assistant requests him/her to come back at
a particular time to pick up the car after servicing. When the customer comes to pick up the
vehicle, he/she meets the service assistant at the same office, makes the payment, and drives the
car away after completing the formalities. In Design B, while the elements of the service are the
same, the manner in which it is laid out and the extent to which the customer is exposed to
various aspects of service delivery is different. For instance, in Design B, the customer is
allowed to visit the garage and is allowed to interact with the repairperson and to seek his
evaluation of the condition of the car. The customer will also have an opportunity to “see” the
manner in which the car is being serviced.
In the example discussed, customer contact is varying. In Design B, the degree of customer
contact is high as the customer has greater opportunities to participate in the service delivery
process and discover several moments of truth. A similar situation exists in the banking sector
where a customer can use either an ATM or a branch office to withdraw or deposit cash/cheques.
There are several situations in service systems where the process design enables the customer to
either make use of self-service with helplines and menus and technology choices, or get a service
agent to co-create the service along with the customer. This includes the use of vending
machines instead of a sales agent to deliver a service (such as issuing tickets or vending coffee
and beverages), malls and supermarkets versus traditional grocery stores, and Web-based online
after-sales service facilities as opposed to on-site visits by service personnel. It is easy to
appreciate that in all these situations, the degree of customer contact varies.
The degree of customer contact is a design choice that a service organization has, and different
choices would result in different impacts on the design and operation of the service delivery
system. At the two extremes of the degree of customer contact, we find certain distinctive
characteristics of the service system. When the degree of customer contact is very low, the
system could be highly mechanized and firms can pursue efficiency goals very well. On the other
hand, when the degree of customer contact is high, it is difficult to pursue efficiency goals.
Satisfying the customer requires clear trade-offs between the efficiency and the effectiveness of
service delivery. Based on the degree of customer contact, one can visualize three types of
designs for service processes:
Low degree of customer contact: quasi-manufacturing
Medium degree of customer contact: mixed service
High degree of customer contact: pure service
Based on the degree of customer contact, one can visualize three types of designs for service processes:
Therefore, customer contact is an important decision variable in the service process design
problem. It affects several aspects of a service system. We shall take a look at some of these
aspects.
Efficiency of operations: Customer contact has a direct bearing on the operational efficiency in service systems. When
service systems are designed with a high degree of customer contact, pursuing efficiency goals becomes difficult or
futile. In a high customer contact system, customers are likely to participate to a greater extent, along with the service
personnel in co-creating the service. Therefore, convenience and quick-response requirements need to be met. These
will come in the way of organizing the entire operations from a pure efficiency perspective.
Therefore, we can conclude that in service systems designed with a high degree of customer contact, effectiveness
goals are more important to pursue. The effectiveness of a service delivery is a function of how well, in the mind of the
customer, the service was delivered. In conclusion, low customer contact may allow a service organization to pursue
internally focused measures such as low cost and efficiency. On the other hand, high customer contact may require that
the service organization pursues externally focused measures pertaining to customer satisfaction.
Capacity choices: From this discussion, we can easily infer that a service system designed with high customer contact
may need to have a capacity in excess of average requirements. The organization will be better off with moderate levels
of utilization. We shall see later in the chapter why high utilization levels are undesirable in the case of service systems.
Therefore, one can expect higher investments of capacity in high-contact systems, which may marginally increase the
cost of the service being provided.
Location of facilities: Service organizations designed with a high degree of customer contact must address the issue of
location of facilities very carefully. Since customer participation in the service creation becomes inevitable in high
customer contact situations, locating the service delivery system in a convenient place is very crucial. Consider two
alternatives for setting up an after-sales service centre for, say, Sony television sets. In one scenario, if the customer
makes a phone call or sends an e-mail, the service personnel will come home, pick up the television set, and deliver it
back after repair. This is a low customer contact scenario. In another scenario, the customer is expected to visit the
service outlet to leave the TV and take it back after the repair is done from the outlet. Several home-appliance service
providers in the country seem to follow this design. This is a high customer contact design compared to the earlier
scenario. In this case, the location of the service outlets is very important. The same logic could be extended to home
delivery of pizzas as opposed to visiting the nearest pizza corner shop to buy pizzas.
Operational control: The nature and degree of operational control that must be put in place also depends on the degree
of customer contact. The variability of the service is likely to be more in a high customer contact system as each
customer will bring his/her idiosyncrasy. Therefore, in a service delivery system with high customer contact, operational
control is not only important but also difficult. Training and skill development of the service personnel and ensuring a
high quality of service are some of the critical aspects of a high customer contact service delivery system.
Closely related to customer contact are two other attributes in a service system. These are the
degree of complexity and divergence in a service offering. The degree of complexity refers to
the steps and sequences in the process, measured by the number and intricacy of the steps. Let us
consider two identical service settings in, say, offering air services between Mumbai and
Bangalore. Let us consider one full-fare airline such as Jet Airways and a no-frills airline such as
Spice Jet. The number of steps involved in completing all the processes in one journey between
Mumbai and Bangalore is more in Jet Airways compared to Spice Jet. Figure 10.1 is an
illustration of the ticketing and the boarding process in the case of these two airlines.
The degree of complexity of a service refers to the steps and sequences in the process, measured by the number
and intricacy of the steps.
As shown in Figure 10.1, in the case of a full-fare airline such as Jet Airways, there are a total
of 10 steps in the ticketing and the boarding process. As opposed to this, in the case of a typical
no-frills airline such as Spice Jet, there are only five steps. In Jet Airways, several fare classes,
meal preferences, and alternative seating choices are available. In addition, Jet Airways needs to
handle multiple frequent-flyer options and associated processes, in addition to addressing the
issue of offering lounge facilities. A typical travel process must address all these steps in the
ticketing and service delivery aspects. On the other hand, in the case of Spice Jet, these steps are
not required in the service delivery process. From our definition, the degree of complexity is
greater in the case of Jet Airways. Drawing cash from an ATM has a lesser degree of complexity
than using a traditional branch office. Similarly, an automated car washing facility may be a less
complex service process than a fully manual car-washing facility. In general, many automation
projects may possess greater complexity in terms of technical systems, maintenance, etc.
However, from a process design perspective, they may simplify the user interface and the
process itself.
Part of the reason for increased complexity may be due to the number of alternative options
available at each step in the service delivery process. This aspect can be understood through the
concept of process divergence. The variability of the steps and sequences represents the degree
of divergence. Process divergence occurs primarily on account of multiple options available at
several steps of the process. A standardized process will have a low degree of divergence. At the
other extreme, in a process in which every step of the process is unique for each customer, the
degree of divergence is high. A psychological counselling service provider will experience a
highly divergent process as each arriving patient requires a unique approach for counselling.
Table 10.1 illustrates the divergence in the case of air travel. The table provides a comparison
between a typical full-fare airline and a no-frills airline. As evident from the table, each step has
more options in the case of a full fare airline (such as Jet Airways or Indian). If you visit the
home page of Jet Airways (www.jetairways.com), you find that, in addition to booking air
tickets, it is possible to also book hotels and insurance. On the other hand, you may not find
these options in the case of Spice Jet. In a full-fare airline, there may be three classes of cabins—
Economy, Business, and First. Typically, a no-frills airline such as Spice Jet offers a single class
cabin (economy—low fare). In a full-fare airline, the menu on offer is wide and one can indicate
one’s preferences even at the time of booking the ticket. In contrast, in a no-frills airline, a
limited set of snacks is sold on board. From a process design perspective, degree of divergence
introduces greater challenges in the service delivery as we saw in the case of degree of
complexity.
The executional latitude or variability of the steps and sequences in a process represents the degree of
divergence.
In a service setting, customers are part of the system. Therefore, organizations need to
incorporate the nature of interaction and the scope and role of the customer in the overall service
consumption as an important element of the design. A simple service scenario of providing a cup
of standard coffee/tea in exchange for exact change from a customer can be effectively serviced
by a coffee vending machine or a roadside dhaba. On the other hand, in a Café Coffee Day
outlet, a customer can choose from over 20 different varieties of coffee/tea. In such a service set-
up, it is impossible to handle requests using a machine. Moreover, the large number of varieties
may require a certain degree of interaction with the customer. The paanwallah pre-pares
virtually an infinite variety of paans because the ingredients he adds to make each paan is a
function of an individual customer’s taste preferences. Because of these considerations, the three
parameters of customer contact, degree of divergence, and degree of complexity influence the
design of service processes significantly.
Furthermore, each of these parameters has a certain impact on the labour component in a
service system design. Customer contact directly influences the nature of interaction in a service
system as we have already seen. On the other hand, the degree of divergence signifies the
customization in the system. A high degree of divergence, as we saw, results in a high degree of
customization. As the degree of customization, the degree of interaction and the complexity of
the service increases, the labour intensity required in delivering the service will also go up.
Figure 10.2 illustrates the relationship among these factors and how they all eventually affect the
labour intensity in a service system.
FIGURE 10.2 Factors affecting design in services
Knowledge of the impact of the labour components of a service design is very important as it
helps to develop appropriate systems for service delivery. When the degree of interaction and
customization is low, the flow becomes streamlined due to fewer varieties. It is also possible to
exploit technology options to provide the service. Therefore, such systems will resemble mass
production systems. On the contrary, when the degree of customization and interaction is high,
the system resembles a jumbled flow and similar operations management problems will be faced
by the service system.
From an operational perspective, when the labour intensity is high, it has certain implications
for the service delivery process. Establishing good procedures and practices and focusing on
employee welfare are important aspects of managing such service systems. This is because there
is likely to be greater interaction between the service provider and the customer in such systems.
Starting up of new units and managing growth are challenging issues. An example will clarify
these aspects of such service systems. Take the case of Aravind Eye Care, Madurai or any other
example in the case of health care. Starting another unit cannot happen by franchising the brand
name as a fast-food provider such as McDonald’s may do. Another sector is the hospitality
industry. In both these cases, you will find that labour intensity is greater. Therefore, these issues
are important in the service design process.
On the other hand, in the case of low interaction and customization, the issues of process
design are somewhat different. Since there will be fewer opportunities for interaction, attention
to physical surroundings is important. The development of good standard operating procedures,
deployment of relevant technological aids, and managing rigid hierarchies are other elements of
process design that makes these systems successful. Such firms may also have to spend more on
marketing as means of developing loyalty as repeat purchases through personalized service
offerings are minimal. An Internet-and ATM-based banking system will be a good example to
understand this. If you bank with ICICI Bank you will experience many of these. ATM
transactions, phone banking, and Internet banking channels have reduced the customer
interaction by substituting standard procedures and technological gadgets in the place of service
personnel. They have also invested in greater marketing efforts and better physical surroundings
to deliver the service. Another example to think about is the low-budget hotels such as Ginger
Hotels. They have also put in place technological gadgets and standard operating procedures in
place of service personnel.
FIGURE 10.3 An illustration of service positioning
The culmination of these issues pertaining to service process design is the issue of service
positioning. The three parameters—contact, divergence, and complexity—pertaining to the
service than an organization offers, determine service process and delivery system design.
Service positioning is the strategic choice a firm makes on these three parameters. An
illustration of service positioning is given in Figure 10.3. Let us look at two examples in the
figure. Service positioning in the banking sector has emerged in two different directions. While
technology-assisted banking solutions such as Internet banking and phone banking have created
a new service positioning with lesser labour intensity and interaction compared to the traditional
banking operations, we also witness personal banking solutions that target high–net worth
individuals in which the service system is designed with greater labour intensity and interaction.
Similarly, we see alternatives emerging in the restaurant business as well. Maiyas, the
Bangalore-based restaurant chain, opened a multi-storey service offering in the heart of the city
in which customized solutions and menu offerings are made available on one floor of the
restaurant. This increases both the interaction and labour intensity of the service offering
compared to a typical restaurant business. On the other hand, in a fast-food joint, the service
positioning moves towards lesser labour intensity and interaction as we have witnessed. As
shown in the figure, the technology solutions help the service designer push the service towards
lesser interaction and labour intensity by replacing human elements with technology interfaces
through a variety of gadgets and the Internet.
Service positioning is the strategic choice a firm makes on the degree of customer contact, degree of divergence,
and degree of complexity pertaining to the service.
Several choices of dishes are offered in Blue Ginger. Two of the sample menus offered are
given in Table 10.2. Although this may increase the diversity and the complexity of the
service offering, it provides unique advantages for the service provider.
This example brings out the importance of process design in services systems and the
alternatives from which one could choose to design a service system such as hotels and
restaurants. It also illustrates the complexity of such design choices and the implications for
competitive advantage.
Shrimp paste with crispy rice coating Crispy fried soft shell crab
Grilled chicken in Saigon marinade Fresh rice paper roll with shrimp and chicken
Fresh rice paper roll with tofu and vegetables Tofu and vegetable with dill and turmeric
Crispy deep fried babycorn Fried vegetable rice paper roll
Sautéed lobster in salt pepper and garlic Sautéed lobster in salt pepper and garlic
Braised lamb shanks with lemon leaves Stir fried vegetables in chilly bean sauce
Stir fried chicken with mushroom in sate sauce Stir fried lamb with snow peas
Sautéed spinach and greens with garlic Sautéed spinach and greens with garlic
Sautéed vegetables in chilly plum sauce Spiced fish cari in clay pot
Braised mushroom in green peppercorns Spicy vegetable red cari
Soft noodle with mix vegetables Soft noodle with mix vegetables
Egg and crab meat fried rice/vegetable fried rice Vegetable fried rice
Baguette bread Baguette bread
Crème coconut custard with choice of ice cream Crème coconut custard with choice of ice cream
Fresh fruit platter
Layout design is a process by which decisions are made about locating various resources in an
organization such that the performance of the system is maximized. In a manufacturing
organization, layout planning enables the operations manager to conveniently locate the
resources in a shop floor so that jobs travel smaller distances, are not handled too frequently and
unnecessarily, and spend less time in the system before it is completed. In the case of a service
system, these requirements are equally important. For instance, if we visit the local branch of the
Life Insurance Corporation (LIC) of India Limited to process a new policy, similar requirements
need to be met. We should not be travelling up and down across various departments to complete
the task. We may not also like our requirements to be handled multiple times by multiple
personnel. For instance, when we make a telephonic contact with a service provider, we find that
the call is often transferred to other departments and each time we start narrating the whole story
from the beginning. This is reflective of a bad process and system design. Furthermore, we
would like the service to be delivered fast. All these point to the possibility of using some of the
concepts and tools that manufacturing systems have developed to address layout planning.
However, the major point of departure is with respect to the role of the customer. Since in a
service system, the customer often “co-creates” the service along with the service provider, other
challenges crop up and need to be addressed over and above the issues mentioned so far. Some
of these are as follows:
In a service system, the customer directly interacts with the service provider. Therefore, appropriate structural
mechanisms must be in place to make the process simple, easy to use, and comfortable for the customer.
The appearance of the place where the service is provided in itself influences the impressions and the service quality
aspects. Therefore, paying careful attention to the aesthetics and ambience of the system is very important while doing
layout planning.
Organizations have alternative choices with respect to customer contact and service complexity. These choices reflect in
the service positioning. Once the choices are made, it has a bearing on the layout. Certain aspects of the service need not
be exposed to the customer; therefore, layout planning for such elements of the service delivery could be very different
from those in which there is direct customer contact. An example of this is the restaurant. In a restaurant, the layout for
the kitchen and the stores is vastly different from that of the dining area.
The culmination of this ongoing discussion is that the starting point in the layout design of a
service system is to map various activities pertaining to the service delivery process and identify
specifically how each step of the service delivery process will make an impact with respect to the
service positioning strategy. Once this is done, the steps that require direct involvement and
interaction of the customers could be isolated for one kind of layout planning and the rest with
no direct customer interaction could be considered for an alternative layout planning. The tool
used to do this is known as service blueprinting. Service blueprinting is a method by which the
service delivery process is broken down into individual elements through a step-by-step mapping
of the process. Service blueprinting helps organizations design, monitor, control, and improve
processes and the service delivery system on an ongoing basis. While this is very similar to a
process mapping done in manufacturing systems to analyse processes, we superimpose a few
more elements to this to make it relevant for service systems. We shall define these new aspects
related to service blueprinting before using an example to understand this.
Service blueprinting is a method by which the service delivery process is broken down into individual elements
through a step-by-step mapping of the process.
All aspects of a service delivery system that are in the line of visibility of a customer will be part of the front
office. The front office provides the customer with ample opportunities to discover his/her moments of truth
about the service delivery process.
Line of interaction: In a service layout planning exercise we need to draw the line of interaction. This line separates all
the activities in which the customer has to have a direct interaction with the service provider as part of the service
creation and delivery process.
Line of visibility: This line includes all the structural, planning, and interactional elements of a service delivery process
that fall within the eyesight of the customer. For example, in a restaurant, the kitchen is not in the line of visibility.
However, the parking lots, the washroom, the eating space, the cash counter, the front lawns, and the food delivery
counter are all in the line of visibility.
Front office: All aspects of a service delivery system that are in the line of visibility of a customer will be part of the
front office. Essentially, the front office provides the customer with ample opportunities to discover his/her moments of
truth about the service delivery process, the quality, the people, and a host of other issues.
Back office: All aspects of a service delivery process that are behind (or beyond) the line of visibility constitute the back
office. Customers normally do not have access to these aspects of the service delivery process as these are meant for
internal interactions among the service delivery personnel. The back office may operate as a support arm for the front-
line staff by providing them with data, information, planning tools, and processing support on several aspects of a
service.
All aspects of a service delivery process that are behind (or beyond) the line of visibility constitute the back
office.
Let us try to understand these concepts using the example of a hospital. Suppose you want to
visit a consultant in the Brain & Spine Department of Fortis Hospitals, Bangalore. First, you call
the Brain & Spine Department and ask for an appointment. You follow this up with a visit to the
hospital at the appointed time. The receptionist first collects the consulting fee and extracts the
file from the records room. If you are a first-time visitor, you also go through a registration
process. You wait for your turn to meet the doctor. The doctor interacts with you and gets
additional tests done, if required, before prescribing appropriate medicines. Finally, you will visit
the pharmacy, buy the prescribed medicines and leave the hospital. This process has been
mapped using the concept of service blueprinting. See Figure 10.4 for details. In Figure 10.4, the
line of visibility and the line of interaction are shown by dotted lines. Also, zones indicating the
front office and back office have been shown.
All structural elements in the line of visibility require a careful approach when it comes to
layout planning. Cleanliness, comfort, convenience, proper lighting of the place, provision of
adequate spaces, clear signage, and unambiguous instructions are certain aspects of layout
design. Also, the employees operating in the front office require superior communication skills,
the right attitude and temperament in dealing with the customers. On the other hand, employees
in the back office require greater planning and technical skills than interpersonal skills.
FIGURE 10.4 Service blueprinting: An illustration
Consider the capacity planning problem in the case of the computerized passenger reservation
facility of Indian Railways. In simple terms, the question boils down to deciding the number of
booking counters to be made available to the public. It is obvious that if there are fewer booking
counters, the queue is likely to build and customers may end up spending more time in the
system before they get their tickets booked. We have similar experiences in a banking system or
at BSNL’s bill payment counters. Capacity decisions in service systems are often made on the
basis of the impact on the customers. In service systems, waiting time is an important operational
measure that determines service quality. Similar examples exist in manufacturing systems as
well. A resource that is fewer in number and highly utilized is likely to be a bottleneck and
increase the waiting time of the jobs ahead of it. Due to this, manufacturing lead time will
increase and work in progress will pile up in the factory. This will have a cascading effect in
terms of missing delivery commitments and shipping delays.
Waiting-line models make use of queueing theory fundamentals such as queue length, waiting
time and utilization of resources, to analyse the impact of alternative capacity choices on
important operational measures in operating systems. Therefore, the capacity planning problem
could be analysed using queueing theory fundamentals. We will now study the basic elements of
a queueing system and the alternative scenarios that can be analysed using the waiting-line
models developed.
Waiting-line models make use of queueing theory fundamentals to analyse the impact of alternative capacity
choices on important operational measures in operating systems such as queue length, waiting time, and
utilization of resources.
The important difference between an infinite source and a finite one is the manner in which arrival rates are
estimated.
Clearly, every arrival from a calling population decreases the probability of the arrival of the
remaining machines at the maintenance shop.
Arrival parameters
Customers arrive from the calling population and place a demand on the resources. Therefore,
understanding the arrival parameters is critical to capacity analysis using waiting-line models.
Two parameters are of paramount importance in understanding the impact of arrivals on the
system. The first is the rate at which customers arrive in a service system or work orders arrive in
a manufacturing shop. The arrival rate could be characterized using a Markovian distribution
such as Poisson. Alternatively, the arrival rate could be on the basis of any general stochastic
distribution. The third possibility is a deterministic arrival rate. No matter what the exact
distribution pertaining to arrivals, the fact remains that they significantly influence the amount of
demand placed on the resources and have a bearing on the extent to which capacity should be
deployed in the system.
The second parameter pertaining to arrival is the arrival pattern. Arrivals can be single or in a
batch. People arriving at a restaurant to have dinner or to watch a movie in a theatre, for
example, seldom arrive singly. Typically, a number of people arrive in groups. Many
manufacturing and service systems exhibit such group arrivals. The other issue is whether the
group enjoys any special privileges. Such a group may enter the system and join the queue ahead
of other waiting members. Such differentiation of arriving customers or work orders is very
common in practice and may have certain implications for operational performance measures
and the capacity planning premise. The management of computing and IT facilities is done with
different service-level agreements. A platinum-level service may call for a system restoration
within minutes as opposed to a silver-level service that may specify restoration within the same
business day. These are some examples of the special groups.
Queue parameters
In any operating system, jobs wait on account of resources not being available immediately,
inevitably leading to queue build-up. Therefore, waiting-line analysis requires knowledge of
exactly how the waiting jobs are handled. Furthermore, in a servicing environment, arriving
customers may exhibit different behaviour while waiting in the queue. Some may choose to
leave the queue and exit the system, anticipating delays; some may jump from one queue to the
other anticipating faster service and so on. All these have a bearing on capacity utilization and
the operational performance measures pertaining to the system.
System structure
System structure refers to the manner in which resources are organized in the operating system.
It specifies the number of units of resources available for use in the system, as well as the
capacity available in each unit of resource. It also specifies how the resources are organized and
the order in which the resources will be used by the arriving jobs. Although several variations
exist in practice, it is useful to understand them on the basis of the four parameters (see Figure
10.6 for details). The resources can be organized as single sever or multiple servers. Further, in
order to complete the requirements, jobs may visit more than one stage. It is also possible that
successive stages are visited in a serial or a networked fashion. All these result in varying system
configurations and affect the system performance differently. Figure 10.7 graphically depicts all
these alternatives.
In this chapter, we will not deal with all the alternatives, as some of them are mathematically
too complex to analyse. We will confine ourselves to single-stage queueing systems only. In the
case of multiple servers, the number of servers is denoted by S.
System structure refers to the manner in which the resources are organized in the operating system.
Service parameters
The arrival parameters determine the rate at which demand is placed on the resources by the
arriving jobs. Service parameters, on the other hand, determine how resources are likely to be
consumed in servicing the arriving jobs. Therefore, the arrival and the service parameters
together significantly influence the operational performance measures of the system. The
servicing parameters specify the service time, which could follow a Markovian distribution such
as exponential distribution, or a general stochastic distribution. In some cases, the service time
could be deterministic without any variation (as in the case of a completely automated system).
Service parameters determine the amount of resources likely to be consumed to service the arriving jobs.
The most commonly employed distributions for modelling arrivals in a waiting-line problem
are Poisson and exponential distributions. The number of arrivals per unit time is often found to
follow a Poisson distribution. On the other hand, the inter-arrival time (between successive
arrivals) follows an exponential distribution function. The parameters of these distributions are
as follows:2
Poisson distribution. The probability of the number of arrivals (n), during T, is given by:
Here, λ is the mean number of arrivals per unit time.
Exponential distribution. The probability of an arrival within a specified time can be computed
using the probability density function and is given by:
The mean and variance of a Poisson distribution is the same and is equal to λ. On the other hand,
the mean of an exponential distribution is 1/λ and the variance is 1/λ2.
When service times are random, it can often be approximated to an exponential distribution.
Therefore, we can use Eq. 10.2 to compute service completions if we know the mean number of
services completed per unit time (service rate). The service rate is denoted by μ. If the mean
servicing time is found to be three minutes in an operating system, then μ = 20 per hour.
Similarly, if customers arrive at the system on an average of once in four minutes, then λ = 15
per hour. The arrival rate (λ) and the service rate (μ) have a significant bearing on the operational
performance of the system, as we shall soon see.
Performance metrics
Waiting-line models analyse the adequacy of the capacity to serve the arriving jobs on the basis
of certain operational performance measures, as discussed earlier. These include waiting time
distribution and queue statistics such as the length of the queue and resource utilization
measures. It is also possible to assess the queueing system using some cost parameters. For
example, the length of the queue in a manufacturing system corresponds to the work-in-progress
(WIP) inventory. Similarly, the cost of the resources can be included in the analysis and some
form of cost–benefit analysis could be done to determine the usefulness of adding more capacity
in the system.
The following measures are very useful in assessing the adequacy of capacity in waiting-line
models:
Ls = Average number of customers in the system
Lq = Average number of customers in the waiting line
Ws = Average time a customer spends in the system
Wq = Average time a customer spends waiting in line
EXAMPLE 10.1
Consider a banking system in which customers arrive for service. After data collection about
the number of arrivals in the bank, it has been found that customers arrive at the bank once
every three minutes. What is the probability of exactly 10 customers arriving at the bank
within a time period of half an hour?
Solution
Let us assume the unit time to be an hour.
Therefore, the arrival rate for the bank (λ) = 20. Now, n = 10 and T = 0.5 for the problem.
Substituting these in Eq. 10.1, we can compute the probability of exactly 10 arrivals in the
bank as follows:
There is a 12.5 per cent probability that exactly 10 customers arrive at the bank during a half-
hour interval.
The system measures include statistics pertaining to waiting in the queue as well as while getting
served at the servers. There is a generalized relationship between these performance measures,
which is known as Little’s formula. It can be stated using the following set of equations:
Further, we also know that the time spent in the system is nothing but the sum of the time spent
in waiting in the queue and the service time. Therefore, we can further state:
Multiplying both sides by λ and using Eq. 10.3, we derive the following equation:
Another useful measure is the utilization of servers in a queueing system. The utilization of
servers is denoted by r. The relationship between arrival and service rates and the number of
servers in a single stage queueing system is depicted by the following equations:
Little’s formula provides a generalized relationship between performance measures in a queueing system.
Using Eqs. 10.3–10.5, one can calculate all other measures of interest.
When the utilization of the server approaches 100 per cent, the number waiting in the queue increases
exponentially and approaches infinity asymptotically.
Table 10.3 shows relevant formulae for computing Lq for other types of queueing systems that
we may like to analyse. Once Lq is known, other measures of interest could be computed using
Little’s formula (Eqs. 10.3–10.5).
It is evident from the formulae in Table 10.3 that operational performance measures (Lq in this
case) deteriorate on account of several factors:
When the utilization of the server approaches 100 per cent, the numbers waiting in the queue increase
exponentially and approach infinity asymptotically.
EXAMPLE 10.2
The teller facility of a bank has a one-man operation at present. Customers arrive at the bank
at the rate of one every four minutes to use the teller facility. The service time varies
randomly across customers on account of some parameters. However, based on the
observations in the past, it has been found that the teller takes on an average three minutes to
serve an arriving customer. The arrivals follow Poisson distribution and the service times
follow exponential distribution.
a. What is the probability that there are three customers in front of the teller counter at most?
b. Assess the various operational performance measures for the teller facility.
c. Of late, the bank officials notice that the arrival rate has increased to one every three and a half minutes. What is
the impact of this change in the arrival rate? Do you have any observation to make?
Solution
Key Inferences
The computations point to an interesting trend in the operational performance measures and
show the linkages between these measures and the arrival and service rates. We note that
when the utilization of the teller facility increased from 75 per cent to 86 per cent, there was a
marked increase in the waiting time. Further, the number of customers in the system also
doubled.
Arriving at an optimum level of utilization is an important design issue, especially when capacity planning
decisions are made.
This has important implications for operations managers. A low level of utilization of
resources results in lesser waiting time and faster turnaround for customers. Therefore, customers
may perceive better service quality in the system. Furthermore, operations managers will have
greater flexibility in scheduling operations as excess capacity is available. They will be in a
position to handle sudden surges in demand and have greater flexibility in operations. On the
other hand, due to low utilization, the operating costs of the system will go up. In contrast,
operating the system at high levels of utilization will help the firm economize the cost of
operations but will result in greater dissatisfaction among customers. Therefore, arriving at an
optimum level of utilization is an important design issue, especially when capacity planning
decisions are made. In practice, it has been found that manufacturing systems can work with
relatively higher levels of utilization as compared to service systems. In service systems,
utilization levels more than 85 per cent may not be desirable. These are only simple rules of
thumb that could be used at the time of addressing the capacity requirements of the system in
question.
The waiting-line models discussed earlier underscore the need to address capacity planning
issues in a service system carefully. Any improper or inadequate build-up of capacity in the
system may lead to poor operational performance measures. In a manufacturing organization,
customers do not experience delays directly. It is the work order, which is substituted by
customers, that is delayed on account of bad capacity management. On the other hand, in a
service system, customers may not want to wait for a long time in the system to get their turn for
service. Therefore, service firms need to address capacity issues more seriously than
manufacturing firms, failing which it will directly affect the profitability of the firm and even its
long-term viability.
Two issues merit attention in the case of capacity planning in a service firm. The first is the
responsiveness of the service system. Capacity planning must ensure that service firms deliver
reasonable responsiveness to waiting customers. It does not make sense to make a hungry
customer wait at a restaurant for over 30 minutes to serve food. From the earlier discussion on
waiting-line models, we can infer that service systems cannot work with high utilization.
Building excess capacity is more of a business requirement than a norm in service firms.
Capacity planning must ensure that service firms deliver reasonable responsiveness to waiting customers.
The second issue is one of high demand volatility, characteristic of service firms. Since in
several service firms inventorying services is not possible, capacity planning in services should
also be able to handle fluctuations in demand. Typically, peak-hour and non-peak hour
conditions occur in several service firms and capacity planning must take into account the
complexities arising out of this. It is not prudent to simply build capacity that will meet the peak
hour requirement. Such an approach will result in very poor utilization of the available resources.
Instead of that, the capacity planning exercise should provide alternatives that can handle issues
peculiar to service firms.
It is possible to use several alternatives to handle peak versus non-peak requirements while
planning for capacity in service firms. Some of these pertain to operational decisions and the rest
pertain to the resource usage in the system. Table 10.5 summarizes some of these alternatives. In
several service firms, any of the problems pertaining to capacity are solved through varying the
strategy of operations. During the non-peak hour period, capacity constraints are often non-
existent or very minimal. Therefore, the strategy is to offer customized services. Moreover, in
several service firms involving some inventory (such as food items in a restaurant) some
processes are completed during the non-peak hour by making use of the available slack capacity.
However, when the peak hour approaches, the operations strategy changes to meet the demand.
First, variety is cut to a minimum. This is very helpful because as we introduce variety some
capacity loss becomes inevitable due to changeovers. The second strategy is to offer a “configure
to order” proposition instead of “provide to order”. This is viable in several cases because some
processes would have already been completed in the non-peak hours.
It is evident from this discussion that during peak hours, the service offerings will be narrow;
and during non-peak hours, the service offering will be sufficiently wide. These principles can be
seen in operation at fast-food joints in any of the metros. It is not uncommon that during the
early morning hours, South Indian fast food joints typically offer only idly, vada, and one or two
more items. After the peak hour is over, more varieties are made available in the fast-food joint.
The other approach to addressing capacity issues is to manage the demand itself. Capacity
reservation methods are provided during the peak hours to minimize disappointments. During
non-peak hours, demand could be induced by providing alternative tariff plans and discounts.
The two sets of strategies are complementary to each other, working towards shifting peak-hour
demand to non-peak hours and providing a reasonably level load on the existing capacity.3
Differences in resource management between peak hours and non-peak hours is another aspect
of capacity planning in service organizations. During non-peak hours, the workforce focus is on
a single skill and the performance of dedicated tasks assigned to them. The slack in capacity is
also utilized for multi-skilling, providing training and development of the workforce in certain
cases. In the long run, a multi-skilled workforce is a source of additional capacity in the system.
During peak hours, the workforce is not focused on a single task. Due to higher levels of multi-
skilling, workers perform a variety of tasks and thereby make up for some additional capacity.
Moreover, additional capacity is added to the system in the form of semi-skilled and low-skilled
labour as these inputs are available in plenty and rather easily. During non-peak hours, the
workforce could be trimmed by laying off these temporary workers.
Differences in resources management between peak hours and non-peak hours is another aspect of capacity
planning in service organizations.
Because of these inherent qualities of services, it is difficult to set up precise parameters and
measuring performance on their basis. Clearly, the measure of service quality involves the
perceptions of the customers, the service provider (employee in contact with the customer), and
the management. In this sense, it is a measure of how well the delivered service matches with
expectations. One popular model that takes several of these into consideration is the
SERVQUAL model.
The evaluation of service quality is not made solely on the outcome of the service. It also involves evaluation of
the process of service delivery.
Figure 10.9 shows the SERVQUAL model developed by Parasuraman, Zeithhaml, and Berry
for measuring service quality.4 According to them, five potential gaps exist in any service
system, which influence the service quality. The first gap occurs when the customers’ perception
about the quality of the service that he/she has received is different from his/her expectations.
This gap occurs partly on account of the other four gaps. When the management’s understanding
of its customers’ expectations is at variance with the reality, another gap is created. Moreover,
with this imperfect understanding, the management proceeds with the process of arriving at the
specifications for the service that it proposes to deliver, designs the service delivery system,
trains its employees, sends out communication about the nature of the service provided and so
on. As shown in the figure, at every stage, potential mismatches happen, increasing the gap, and
thereby significantly influencing the final quality of the service offered.
FIGURE 10.9 The SERVQUAL model for service quality
Source: Adapted from A. Parasuraman, V. A. Zeithhaml and L. L. Berry, “A Conceptual Model of Service Quality and its Implications for Future Research,”
Journal of Marketing, no. 49 (Fall 1985): 41–50. Reproduced with permission.
Understanding why these gaps exist in service quality is the first step in addressing them. Some
of the reasons for the existence of gaps are as follows:
According to SERVQUAL, five potential gaps exist in any service system that influence the service quality.
Firms need to address these issues in a sustained manner to improve quality. Obtaining customer
feedback frequently may help the service provider address Gap 1. In every room in a five-star
hotel, guest feedback forms are always kept so that management can tap this information and act
on certain areas to reduce Gap 1. We see similar practices in air travel, hospitals, and other
service outlets as well. Gap 4 is a common problem that we witness. Service firms prepare
brochures without thinking about these issues and present a rosy picture of the service offering.
Since brochures play a crucial role in shaping the expectations of the customers of the service,
more care needs to be exercised to avoid or reduce the gap. Similarly, employee training
programmes that cover not only technical skills, but also behavioural and communication skills,
need to be periodically offered to customer facing service personnel. This may to a large extent
address Gap 3.
Improving Service Quality Through Better Communication: The Case of Aravind Eye Hospitals
Service quality is much harder to measure and improve as services are experiences and not
objects as in the case of a manufacturing organization. However it does not mean that it is
impossible. The SERVQUAL model has identified five gaps between perceived and actual
quality. One of the frequently cited reasons for poor service quality is that customers’
perceptions of quality are different from what they experience while consuming the service.
One way to address this problem is the use of brochures that provide useful information
about the service so that expectations could be set at reasonable and realistic levels a priori.
Aravind Eye Hospitals has utilized this approach to educate their patients about the steps
in the process that they are likely to go through and the likely time for each of these steps. A
brochure entitled “Your Guide to Quality Eye Care,” printed both in English and the
vernacular language, is given to the patients so that they are aware of the steps and the time
taken. According to this brochure, it may take up to two hours to complete the process. The
various processes and the time taken for each of them is listed:
1. Registration: 5 minutes
2. Vision test: 10 minutes
3. Refraction check: 10 minutes
4. Preliminary examination: 20 minutes
5. Blood pressure/eye tension/sugar test: 30 minutes
6. Final examination: 5 minutes
The brochure makes an attempt to briefly describe the process in each of the listed stages and
explains why a particular test is done in that stage of the process. Further, it provides a
schematic block diagram of the process flow and the location of various facilities including
the pharmacy, the information centre and the cash counter. The brochure also has a detailed
layout plan of the building. Through this information, the brochure ensures that an arriving
patient will have a fair idea of what to expect and where to go.
The Aravind Eye Hospitals brochure educates us about some important aspects of
designing a brochure for a service offering. It reminds us that care must be taken while
designing a brochure. Committing to more than one can deliver or painting a rosy picture of
the reality in the brochure might backfire. Instead of improving the service quality, it may
indeed deteriorate it by creating false expectations when the operations cannot deliver the
same.
Source: Based on “Your Guide to Quality Eye Care,” the brochure of Aravind Eye Hospitals & Post Graduate Institute of Ophthalmology, Madurai.
SUMMARY
Process design of service systems differs vastly from that of manufacturing systems.
The degree of customer contact, the degree of complexity and the degree of diversity of service offerings have a
significant bearing on process design in service systems.
Service positioning is the strategic choice a firm makes on the three parameters of customer contact, degree of
complexity and degree of diversity in the offerings.
Service blueprinting helps organizations to design, monitor, control and improve processes and the service delivery
system on an ongoing basis.
Waiting-line models make use of queueing theory to analyse the impact of alternative capacity choice on operational
performance measures such as waiting time, resource utilization and queue length.
Waiting-line models indicate that as resource utilization approaches 100 per cent, queue length and waiting time
become very large. At low levels of utilization, costs of operations shoot up. Therefore, organizations face a critical
trade-off between cost and utilization.
Service systems tend to have plans for lesser resource utilization compared to manufacturing systems. Moreover,
service systems use alternative strategies for managing capacity during peak and non-peak hours.
Service quality is hard to measure. Various gaps exist between the delivered service and the expectations of the service
receiver. This provides a measure of the service quality.
FORMULA REVIEW
Alos,
Server utilization in the case of multiple servers, In the case of a single server queue,
a. Travel agency
b. DVD rental services
c. Courier services
d. Computer facilities management services
e. Investment consulting services
f. Sweets and savouries shop
8. Are capacity management issues in a service system different from those in a manufacturing system?
9. The manager of an automobile garage in the heart of the city of Delhi has built such a good reputation for himself that more
and more customers throng to his garage for getting their cars serviced. The utilization of the facility is now about 88 per cent.
He has been getting more and more requests for automobile servicing. What will be your advice to the manager? Prepare a
one-page report for him.
10. Compare the issues pertaining to measuring service quality with that of product quality. Which one is easier to assess? Why?
11. Consider any one service system which you are familiar with. Can you identify one instance for each of the five gaps as per
the SERVQUAL model?
12. What are the challenges in assessing the quality of services? How can an organization mitigate some of these challenges?
PROBLEMS
1. Consider a single-server queue with Poisson arrivals and negative exponential service times. The arrival rate is 10 per hour
and the service rate is 15 per hour. Compute the following:
2. In a single-server system with Poisson arrival rate and exponential service time distribution, the arrival and service parameters
are as follows:
Arrival rate: 10 per hour; Service time: 5 minutes
Compute the following:
a. How long will the customers wait in the queue to get their turn for the service?
b. If the waiting time in the queue needs to be brought down by 10 per cent, by how much should the service be speeded
up?
c. If the arrival rate comes down by 10 per cent, what will be the impact on the waiting time in the system if the service
characteristics remain unchanged in the system?
Use Little’s formula and compute other relevant performance measures for the system in each
of the given scenarios.
4. Consider a single-server queue with Poisson arrivals and exponential service time distributions. The arrival rate is 5 per hour
and the service time is 8 minutes. Compute the following operational measures of performance for the system:
What will be the threshold arrival rate so that the total time spent in the system is greater than or equal to 15 minutes?
5. Consider a queuing system with a poisson arrival and negative exponential service time distributions. If the arrival rate is 15
per hour and the service time is 3 minutes, compute the following:
6. There is a repair shop in a residential locality attending to TV repairs. Failed TVs arrive at the repair shop in a Poisson fashion
at the rate of 7.5 per day. It takes on an average about 45 minutes to restore the TVs. Service times are exponentially
distributed. The TV shop is open daily between 10.30 am and 6 pm.
7. A university campus has a students’ coffee shop, which the students normally visit during their short breaks to have coffee and
other beverages. Arrivals at the coffee shop follow a poisson distribution on an average at the rate of 3 per minute. The service
counter takes about 18 seconds on an average to attend to a student. The service times are found to be exponentially
distributed.
9. In a telephone booth, the arrivals of customers wanting to use the phone follows Poisson distribution and the duration of the
calls placed follows exponential distribution. On an average, 15 customers arrive in an hour and end up using the phone for
180 seconds. The charges are ₹2 per minute. The booth owner has invested ₹27,000 on the whole set-up and pays a monthly
tariff depending on the usage. Every call leaves him with a 10 per cent margin to cover his initial cost after paying for tariffs
and meeting other expenses of running the booth. He keeps the booth open for 10 hours every day from Monday to Friday.
Compute the following:
a. How many waiting spaces should he provide for the arriving customers?
b. How long will the customers spend ever since he/she arrived to complete his/her call?
c. What is the utilization of his facility?
d. When can he recover the cost of his investment?
10. The ATM centre for ICICI Bank in a certain locality of Lucknow currently has one ATM. Customers arrive at the ATM
centre at the rate of one in every two minutes. Although the nature of transactions may vary with every customer, it has been
found that on average, the service time for a customer is 90 seconds. The arrivals follow Poisson distribution and the service
time follows negative exponential distribution. The bank officials would like to know the impact of adding another ATM
machine in the same premises. Prepare relevant data for the bank officials.
11. Customers arrive at a car washing facility at the rate of one in every 6 minutes. The car washing facility is at present a
manual process that takes a variable time, whose mean and standard deviation are estimated to be 4 minutes and 2 minutes
respectively. A company has offered to install an automated system, which will eliminate the variability of the facility.
Furthermore, it will bring down the service time to 2.5 minutes. The system will cost ₹500,000. The company expects that if
the waiting time for the customers comes down by at least 15 per cent, the number of customers arriving at the station will
increase by 50 per cent of the current level. Servicing each customer will mean a contribution of ₹100 to the firm. The facility
works for 8 hours per day for 25 days a month. Should the firm go for the new system? What will be the payback for the
investment?
12. Prolonged observation of a multi-server facility in a supermarket resulted in the collection of some data with respect to
customer arrivals at the supermarket and the service time at the check out counter. Table 10.6 has the data based on 100
observations made in the last two months. The data is representative of the situation prevailing in the supermarket. Currently,
there are five checkout counters in the supermarket. The supermarket wants to know if reducing the variability of the service
process by 25 per cent is better than adding one more counter at the shop. Use the data given in Table 10.6, compute the
relevant operational performance measures for the supermarket, and advise.
NET-WISE EXCERCISES
1. ITC forayed into the Hotel business to support the national priority of developing new avenues of foreign exchange earnings
and boosting tourism. Beginning with the Sheraton Chola, Chennai in 1975, ITC’s journey in this business has redefined the
face of Indian hospitality. ITC Hotels is one of largest hotel chains in India with over 90 hotels in more than 70 destinations.
ITC Hotels pioneered the concepts of branded accommodation, branded cuisine, environment, and guest safety. ITC Hotels is
an exemplar in sustainable hospitality with all its premium hotels being LEED® Platinum certified. Visit the URL
http://www.itchotels.in/Index.aspx to know more about the services offered.
The Indian Hotels Company Limited (IHCL) and its subsidiaries are collectively known as Taj Hotels Resorts and Palaces. Taj
Hotels Resorts and Palaces comprises 93 hotels in 55 locations across India with an additional 16 international hotels in
Maldives, Malaysia, Australia, UK, USA, Bhutan, Sri Lanka, Africa, and the Middle East. Visit the URL
http://www.tajhotels.com to know more about the services offered.
a. Select one hotel each from the list available in the URLs and pursue it by visiting all relevant links. Assess various
elements of service design, service complexity, and divergence. Also try to understand the layout of different areas of
the hotel. Visit the photo galleries and take virtual tours wherever available to get a feel of the place.
b. Prepare a report highlighting the various aspects of process design as discussed in this chapter. Prepare a subjective
assessment of the complexity of the service design and its implications for the hotels. Also make a comparative analysis
of the two hotels.
2. The Housing Development Finance Corporation Limited (HDFC) was amongst the first to receive an ‘in principle’ approval
from the Reserve Bank of India (RBI) to set up a bank in the private sector, as part of the RBI’s liberalization of the Indian
Banking Industry in 1994. HDFC Bank is headquartered in Mumbai. HDFC Bank has a nationwide network of 3,403
Branches and 11,256 ATM’s in 2,171 Indian towns and cities. Moreover, HDFC Bank’s ATM network can be accessed by all
domestic and international Visa/MasterCard, Visa Electron/Maestro, Plus/Cirrus, and American Express Credit/Charge
cardholders.
Visit the URL of the bank at http://www.hdfcbank.com. Pursue the link related to Personal Banking and go through the links in
that page. Study in detail the Imperia/Classic/Preferred, and Private Banking options by clicking the menu ‘Products’ and then
‘Premium Banking’ links there under.
After going through the pages, prepare a report on the service offerings of HDFC bank from the perspective of service process
design and its implications. In the chapter, there is a discussion on service positioning. Use the framework given in that
section (see Figure 10.3 in the chapter also) and show how it applies in the case of HDFC Bank.
MINI PROJECTS
1. Visit the local branch of a bank and perform the tasks listed here. Based on your analysis and major findings, prepare a final
report of recommendations to the bank manager.
a. Study the working of the bank, collect data on working times, types of customers, nature of transactions, waiting times,
capacity of resources, waiting space etc. Understand from the manager the key operational performance measures of
interest to the bank branch, elicit the problems that need to be addressed.
b. Estimate the key parameters of the waiting-line model to be constructed to assess the current working of the branch.
The parameters include number of arrivals (or mean-inter arrival times), mean service times and the number of servers.
c. Build a waiting-line model and compute the measures of performance. Is there adequate capacity in the system to serve
the arriving customers? If not, what are the alternatives the bank branch has to improve the capacity (and thereby the
performance)?
d. If there is adequate capacity at present, identify methods to improve the performance by 10 per cent over the next 6
months.
2. Perform a similar exercise as described in Mini Project 1 by studying the student hostel mess in your college and prepare a
report for the mess manager.
CASE STUDY
Waiting Times at Post Offices
The customers at the post office in a busy locality of Bangalore were getting restless. Most of
them were unanimous in saying, “Do something about the agents, it’s because of them that we
have such long queues.” A few others had come to the post office multiple times for their work
as a particular “software wasn’t working.” Another common grouse of the people was that
different kinds of transactions had been combined in the same queue and due to this, one had to
go through a long wait to get even a small transaction such as a passbook entry done. However,
there were a few people, who seemed to be “seasoned customers,” who would manage to jump
the queue and discreetly carry out small transactions. When it was suggested to the customers
that they could avoid facing long queues by going through agents, a few said that they had
already tried that route, but gave up after the agents made a large number of mistakes. The issue
was about reducing the waiting time of the customers and thereby improving the overall service
quality.
Let us concentrate on the services pertaining to savings bank accounts, monthly income
schemes, savings certificates (such as Kisan Vikas Patra) and public provident funds, since these
have been having problems with long queue lengths in several post offices in the country.
It is generally observed that the queue is more on Saturdays and Mondays than on other days.
Another problem is that the monthly income interest for senior-citizen schemes accrues on a certain day in the month
and there is generally a huge crowd on the following few days.
A large number of people who transact at the counters related to savings bank and monthly income schemes are agents
representing other people. These agents get a small proportion of the money involved in the transaction as “incentive”.
They indirectly operate as salespeople for the post office’s savings and monthly income facilities.
Agents
The Government of India, in association with India Post, has formed a network of authorized
agents in order to promote small savings through the National Savings Certificate, National
Savings Scheme, Indira Vikas Patra and Kisan Vikas Patra schemes. These agents are like
“marketing executives” for the post office and are responsible for getting new customers for the
different schemes.
One can identify agents as they usually carry large bags with papers of a large number of
clients. They also seem to perform multiple transactions at counters. They do not get any
preferential treatment at the post offices. They have to stand in the common queue along with
the general public. However, given that they are familiar with the people behind the counter,
they can push in multiple applications at once. They perform a variety of transactions that are
available at the post office. They need to deliver a full range of services (post office, bank
deposits, etc.) to their clients in order to maintain the relationship. Most of the clients stay with
their agents for a longer term. The agents need to provide proper incentives to their clients and
maintain a high quality of service.
There is quite a bit of queue jumping that happens. Agents stand in a particular queue for a
couple of minutes, tell the person behind them that they will be right back, and then go and stand
in another queue. This way, they manage to “parallel process” two queues and manage to get
their job done faster. In addition, since the people behind the counters are familiar with the
agents, no one (except for the odd customer) seems to mind.
As told by one postmaster, there is a large number of agents. These people come with huge
sheets of papers to be processed, spend a long time at the counters and repeatedly stand in
queues till all their clients’ work is done. There is not enough space (physically) in the post
offices in order to form the queues. Hence, queues tend to form horizontally in front of the
counters and end up obstructing other counters.
A few counters seem to be doing “batch processing”. They collect papers from a group of
people (all of whom are crowding the counter and also the neighbouring counters), process all of
them together, and then finally complete all the processes together.
TABLE 10.8 Total Residence Time in the System (Minutes) for Alternative Proposals
TABLE 10.9 Utilization and Average Service Times for Proposal 3
2. What is your recommendation to the postmaster for solving the problem? Would you recommend any of the three proposals?
Justify your recommendations with suitable arguments.
3. What are the other steps that the postmaster can take to solve the problem at CPO?
SUGGESTED READINGS
B. Berman, (2005),“How to delight your customers”, California Management Review, 48 (1), 129 – 151.
R. B. Chase and D. A. Tansik, “The Customer Contact Model for Organisation Design,” Management Science 29, no. 9
(1983): 1037–1050.
D. P. Cook, C. H. Goh and C. H. Chung, “Service Typologies: A State of the Art Survey,” Production and Operations
Management 8, no. 3 (1999): 318–338.
L. J. Krajewski and L. P. Ritzman, Operations Management: Processes and Value Chains, 7th edition (Upper Saddle
River, NJ: Pearson Education Inc, 2004).
R. W. Schmenner, “How Can Service Businesses Survive and Prosper?” Sloan Management Review (Spring 1986): 21–
32.
J. K. Sharma, and R. Narang, (2011), “Quality of HealthcareServices in Rural India: The User Perspective”, Vikalpa 36
(1), 51 – 60.
G. L. Shostack, “Service Positioning Through Structural Change,” Journal of Marketing 51, no. 1 (1987): 34–43.
R. L. Silvestro, F. R. Johnston and C. Voss, “Toward a Classification of Service Processes,” International Journal of
Service Industry Management 3 (1992): 62–75.
S. Thomke, (2012). “Mumbai’s Models of Service Excellence.” Harvard Business Review, 90 (11), 121–126.
T. M. Tripp, and Y. Grégoire, (2011), “When Unhappy Customers Strike Back on the Internet”, MIT Sloan
Management Review, 52 (3), 37 – 44.
U. Wemmerlov, “A Taxonomy of Service Processes and Its Implications for System Design,” International Journal of
Service Industry Management 1, no. 1 (1990): 20–40.
SUPPLEMENT 10A: SIMULATION MODELLING
We shall understand these aspects of building a simulation model and using the model to
obtain data as simulation proceeds in order to gain some insights about the system being studied.
All simulation models use the inbuilt random number generator available with software
applications such as Microsoft Excel to model the arrival pattern. In order to do that, the
procedure followed is to sample a random number [in the case of a computer software, the
random number generator is invoked by using the RAND() function] and map it to the
corresponding cumulative probability value in the table and read the corresponding inter-arrival
time. If the first random number that we obtain is .09, from the cumulative probability value in
Table 10A, we will sample an inter-arrival time of three minutes. Similarly, if the second
random number obtained is 0.61 then the corresponding inter-arrival time will be seven minutes.
This is demonstrated for a set of 20 arrivals in Table 10B.
Consider the problem that a newspaper vendor faces on a day-to-day basis. The newspaper
vendor buys a copy at the rate of ₹0.33 and sells it at ₹0.50. If the vendor is unable to sell the
paper, it fetches a salvage value of ₹0.05. The decision the newspaper boy faces is the
number of papers that he needs to buy daily. The following random events complicate his
decision making:
On 20 per cent of the occasions, the demand is likely to be low. There is, however, a chance that the demand is
likely to be very good on 35 per cent of the occasions. At other times, the demand is likely to be moderate.
Likely demand on a particular day depends on whether it is low, moderate, or good. The pattern of demand
observed in the past is shown in Table 10D.
TABLE 10D The Pattern of Demand Observed in the Past
Build a simulation model for the above problem, run it for 20 days, and provide a framework
for the newspaper vendor to decide on the number of papers that he needs to order on a daily
basis.
Solution
In this example, we notice two random events affecting the decision. So, let us first prepare
the cumulative probability distribution for these random events. Based on the given
information, Tables 10E and 10F depict the cumulative probability distribution for the
respective random events.
Simulation models belong to the category of descriptive modelling in which the impact of
a particular decision is analysed, given the conditions under which the model behaves.
Therefore, in this example, we need to decide on the alternative for which we would like to
study the behaviour of the system. The alternative to be studied is nothing but the decision
variable for the newspaper vendor, that is, the number of papers that he needs to order daily.
The steps in the simulation are as follows:
1. Fix the decision variable whose effect on the system needs to be studied. In Table 10G, the effect of buying 60
newspapers every day is being studied.
2. Sample a random number from the random number generator and use it for deciding the type of the day using their
cumulative distribution table for the type of the day (Table 10E).
3. Sample another random number from the random number generator and use it for deciding the demand expected using the
cumulative distribution of demand for the type of the day (Table 10F).
4. Since the number of copies bought as well as the demand is known, calculate the revenue for the day, lost profit (if any),
salvage (if any) and daily profit (if any).
5. This completes one day of simulation. Repeat the above steps until the terminating condition for the simulation is
achieved. In our example, the above steps are performed 20 times and the statistics is used to assess their impact on the
system.
Proceeding in this manner, the relevant data for the system has been captured in Table 10G.
The table shows the behaviour of the system with respect to a decision variable of the vendor
buying 60 newspapers every day.
TABLE 10F Cumulative Distribution of the Demand Related to the Type of Day
TABLE 10G Simulation of the Newspaper Vendor Problem for a Daily Purchase of 60 Papers
The results indicate that the news vendor would have made an average daily revenue of
₹29 and an average daily profit of ₹7.52. In a similar fashion, one can compute the average
daily profit for alternate values of the number of newspapers bought daily and use this
scenario analysis to arrive at an optimum decision.
TABLE 10H The Simulation Results for Various Values of the Number of Newspapers Bought Daily
Number of Newspapers Average Daily Revenue (₹) Average Daily Profit (₹)
Bought
40 20.00 1.96
50 25.00 5.36
60 29.00 7.52
70 32.25 8.75
80 34.00 8.12
90 34.25 5.63
100 34.25 2.83
Table 10H shows the simulation results for various values of the number of newspapers
bought daily. From the table and the chart beginning for that, by buying 70 news-papers
daily, the news vendor can maximize his/her daily profit. On the other hand, by buying 90 or
more number of newspapers, that daily revenue flattens at ₹34.25 while the daily profit
deteriorates. Figure 10A pictorially demonstrates this.
FIGURE 10A Simulation analysis of the new vendor problem
Figure 10B outlines the steps in a simulation experiment before results are drawn from the
simulation model and useful interpretations are made for management decision making.
FIGURE 10B Steps in a simulation experiment before results and interpretation are drawn
Using the software, a modeller can quickly replicate any real operation system and use the
model for analysing the performance and the behaviour of the system under study. Owing to
its popularity, a number of computer packages are available for simulation study of
operations systems. The popular among them include ARENA, PROMODEL, @RISK,
WITNESS, and GPSS/H The recent versions of the software provided 3-D animation
facilities, debugging facilities, and alternatives for interfacing the input and the output with
other software. With these improvements in the development of simulation software, it is
possible to link a simulation study with several other aspects of management decision
making today. For example, when the design of a new factory is being attempted, it is
possible to link the 3-D CAD/CAM software to the 3-D animation simulation software and
simultaneously study not only the physical and geometric properties related to locating
machines on the shop floor and their relationship with material handling systems and other
system components, but also its impact on waiting time, congestion, and inventory build-up
using a simulation software.
Applications of Simulation
Simulation modelling finds a host of applications in operations management. Some of them
are listed here:
Capacity analysis: Capacity decisions are often expensive and require some detailed analysis before investments
are made. Further, the use of capacity in operations often involves complex networks of queues and several random
behaviours of system entities. Therefore, capacity analysis is often done using a simulation model.
Layout design: Designing of manufacturing systems with respect to new product lines requires some analysis about
the number of resources to be deployed, utilisation of the resources and their impact on waiting time, and inventory
build-up and congestion in the shop floor. It also requires some understanding of the material-handling
complexities arising out of material flow patterns on the shop floor. Such situations are often analysed using a
simulation model.
Operational performance of service systems: Service systems are characterized by high levels of heterogeneity and
complexity due to the presence of the human element in the process. Service systems often present a complex
network of queues. Moreover, they are transient in nature, making mathematical modelling of such systems almost
impossible. Several operational details with respect to the number of servers to be provided, the waiting space to be
provided, the scheduling practices to be adopted, and the investments to be made in various other secondary
resources are to be analyzed in greater detail in a service system. Simulation modelling is a valuable tool to address
these issues in a service system.
Traffic management systems: Study of traffic management systems often involve real-time and dynamic aspects of
congestion, queuing and waiting time. These aspects are easily studied using a simulation model. Therefore,
modelling of large-scale public systems such as public transport, railway networks and airline systems could be
done using simulation modelling.
Supply chain management: Similar to the traffic system, several aspects of supply chains also require an
understanding of dynamic situations in order to address fleet management and logistics and distribution systems.
Inventory management systems involving random demand and multiple demand points are difficult to study
mathematically. Under these conditions, a simulation model would be very useful.
SUMMARY
Simulation is a mechanism in which a numerical model mimics reality. A simulation model is often resorted to when
the system to be studied does not lend itself to any readily available mathematical model or when a complex set of
random events tend to affect the behaviour of the system.
Simulation models belong to the category of descriptive modelling in which the impact of a particular decision is
analysed, given the conditions under which the model behaves.
To build a simulation model, use of the following is inevitable: a clock to keep track of the passage of time in which the
behavior of the system is observed, a random number generator to sample random numbers, and mechanisms to collect
data of interest as the clock advances from time zero to the end time.
Central to the simulation methodology is the use of random numbers to model the stochastic behaviour that we face in
real life in several operations systems.
The variance introduced on account of using one set of random number needs to be eliminated before the results of a
simulation study are used for meaningful interpretation. This is known as tactical planning of a simulation experiment.
Owing to its popularity, a number of computer packages are available for simulation study of operations systems. The
popular ones among them include ARENA, PROMODEL, @RISK, WITNESS and GPSS/H.
Simulation modelling finds a host of applications in operations management. Some of them are capacity analysis,
layout design, operational performance of service systems, traffic management systems and supply chain management.
REVIEW QUESTIONS
1. What is the relevance of simulation as a modelling tool in operations management?
2. Suppose a factory manager approaches you with a dilemma. He is not sure if he must use simulation for a problem he faces on
the shop floor. On what basis will you advise him to use the simulation model or some other mathematical model to solve his
problem?
3. Identify three situations in each of the following places where simulation could be a useful tool to study the problem:
a. A five-star hotel
b. The Bangalore Dairy (that processes and packages milk and milk products)
c. Chennai Central Railway Station
d. The student recreation centre in your college
e. The final assembly plant of Suzuki Motors Limited
f. A multi-specialty hospital
4. How are random behaviours studied in a simulation modelling exercise? Take the example of the number of students arriving
at the hostel mess for lunch as an example to illustrate this.
5. What are the steps in carrying out a simulation experiment? Which of these steps are critical and why?
PROBLEMS
1. An ophthalmologic clinic has patients arriving at the clinic in a certain pattern. Based on a one-month observation, the
following data (see Table 10I) is obtained with respect to arrival:
a. Simulate the system for 20 arrivals. When did the last patient arrive if the clinic opened at 8.00 a.m.?
b. Simulate the system for 3 hours. How many patients have arrived so far?
c. Compute the average waiting time of the patients under the following conditions:
Each arrival spends approximately 9 minutes for service.
The service time is distributed in the following fashion: 20 per cent spend 13 minutes, 45 per cent spend 10
minutes, and the rest spend 6 minutes.
When will the clinic complete the servicing of the last arrival if the doors are closed after three
hours?
2. A bakery owner sells three types of specialty cakes: honeycomb, butterfly, and whole nut. The demand for these is random
and based on his experience in the past, the bakery owner found that 45 per cent prefer honeycomb, 37 per cent prefer
butterfly, and the rest prefer whole-nut. The other uncertainty is the order size. In 25 per cent of the cases, the order size is for
half a dozen, and in 47 per cent cases, the order is for a dozen. The remaining orders are for 2 dozens. The bakery owner
wants to know what should be the optimum daily production run that he must schedule for the three varieties. Due to cost
sharing, technology constraints, and other considerations, the production is always to be maintained in the ratio 5:3:2 for the
three varieties. The cost, price, and salvage value of unsold cakes per dozen cakes are given in Table 10J. Simulate the
problem.
TABLE 10J The Cost, Price, and Salvage Value of Unsold Cakes per Dozen
SUGGESTED READINGS
J. Banks, J. S. Carson II, B. L. Nelson, and D. M. Nicol, Discrete Event System Simulation (Upper Saddle River, NJ:
Prentice Hall, 2001).
A. David, (2001), “Models Implementation: A State of the Art,” European Journal of Operational Research 134, pp.
459–480.
G. Gordon, System Simulation, 2nd edition (New Delhi: Prentice Hall India, 1987).
C. Harrell, B. K. Ghosh, and R. Bowden, “Simulation Using PROMODEL”, (McGraw Hill, 2000).
B. Mahadevan and T. T. Narendran, “Buffer Levels and Choice of Material Handling Device in Flexible Manufacturing
Systems,” European Journal of Operational Research 69, pp. 166–176.
B. Mahadevan, D. F. Pyke, and M. Fleischmann, “Periodic Review Push Inventory Policies for Remanufacturing,”
European Journal of Operational Research 151, pp. 536–551.
R. E. Nance and R. G. Sargent, “Perspectives on the Evolution of Simulation,” Operations Research, 50 (1) (2002), pp.
161–172.
R. E. Shannon, Systems Simulation: The Art and Science (Upper Saddle River, NJ: Prentice Hall Inc., 1975).
J. G. Shantikumar and R. G. Sargent, “A Unifying View of Hybrid Simulation/Analytic Models and Modeling,”,
Operations Research 31 (6), pp. 1030–1052.
CHAPTER 11
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
The Airblade Tap hand dryer is Dyson’s newest creation, taking what the company has learned about touchless hand drying,
and packing it into a sleek faucet you can install in your own home. The Dyson Airblade Tap hand dryer won the gold in the Best
in Biz Awards 2013.
Source: http://airblade.dyson.com/hand-dryers/airblade-tap/ airblade-tap/gallery.aspx
India is fast emerging as the global research and development (R&D) hub of the world. In 2006
alone, for example, 100 of the world’s top R&D companies employed more than 15,000
scientists in India. Many multinational corporations are setting up their R&D labs in India. For
instance, Sony Ericsson has established an R&D unit for mobile phones in Chennai. Dell has
established its biggest research and development centre outside the United States in Bangalore.
British aerospace major Rolls-Royce has tied up with the Indian Institute of Science (IISc),
Bangalore, and Imperial College, London to work together on a new research project to develop
alloys for use in “greener” aircraft engines.
The DaimlerChrysler Research Centre in Bangalore is involved with fundamental and applied
research in avionics, simulation, and software development. Boeing is working with HCL
Technologies to co-develop software for everything from navigation systems and landing gear to
the cockpit controls for its 7E7 Dreamliner jet. Microsoft has built its largest development centre
outside the United States in India. The University of Oxford is to set up its first research centre
outside the United Kingdom in association with the Confederation of Indian Industry (CII) with a
focus on issues related to India’s rapidly expanding economy, such as infrastructure, education,
social entrepreneurship and business taxation.
The reasons for this interest in India by global majors are obvious: its rich talent pool of
technical know-how comes at a very competitive cost. India produces a huge pool of graduates—
2.5 million per year in IT, engineering, and the life sciences. With 4,000 doctorates and 35,000
postgraduate degrees awarded by 162 universities, India has the third largest scientific and
technical human-resource base in the world.
India’s R&D capability spans a wide spectrum of industries: agriculture, biotechnology,
energy, nanotechnology, information technology, space, defence, automobiles, aviation,
pharmaceuticals, theoretical physics, and statistics, among others. India is the only developing
country and the sixth country in the world to manufacture and launch its own satellites in a geo-
stationary orbit. Moreover, it has also launched satellites for foreign customers such as Germany
and Korea. In October 2008, India successfully embarked on its moon mission when the Indian
Space Research Organisation (ISRO) launched Chandrayaan I in a bid to expand knowledge
about the moon.
India is one of the few countries that have developed stem cell lines as part of the stem cell
network worldwide (at Reliance Life Sciences and the National Centre for Biological Sciences).
In the pharmaceutical sector, India has the largest number of FDA-approved plants after the
United States. Indian Institute of Technology (IIT) Madras has developed the world’s first
nanomaterial-based water purifier. With initiatives in so many fields, India’s role in research and
development is undeniable.
To be successful, businesses need to engage constantly in value creation for shareholders as well
as customers. Innovation has been recognized as a potential lever for value creation. Innovation
can enhance the value and applicability of the organization’s existing competence. Organizations
in a sector of an industry that offer a homogenous set of products and services have very little
opportunity for value creation. The notion of creating value through innovation has been well
articulated by researchers and demonstrated by successful companies. It is based on the premise
that innovation creates disequilibrium in the market, paving the way for new value-creation
opportunities. The heterogeneity in the offerings of an organization is likely to delay the
appropriation of profits in competitive markets by the competitors.
Operations management addresses the issue of innovation by enabling organizations to bring
in some distinctiveness in their offerings. The distinctiveness may be on account of the product
or service it offers, the technology and channel it employs, or the processes it uses to provide the
product/service to the customer. For example, when Amazon offered 1.5 million titles of books
through a virtual bookstore in 1995, customers experienced a high degree of differentiation.
Similarly, Dell’s direct selling model had a high degree of differentiation. The product
development process provides a broad set of tools, techniques, and concepts that enable an
organization to provide distinctiveness in its offerings by rolling out new products and services
faster and at a lower cost.
In recent years, we have witnessed a rise in customer expectations with respect to the products
and services offered. Organizations can benefit in this scenario, either by offering highly
differentiated products and services or by offering very cost-effective products. Furthermore,
organizations can also benefit by bringing these products and services much faster than the
competition and gain from early-mover advantages. In order to achieve this, organizations need
to have a robust mechanism to understand customers’ expectations. They must also have the
capability to react faster once they understand these expectations. In the 1950s and 1960s,
Hindustan Motors introduced variations of its Ambassador (the Mark 2, the Mark 3, and so on)
roughly once in ten years. Today, no automobile manufacturer can afford to take that much time
to introduce new products and variations of the existing ones. A good product development
process addresses these issues and provides a set of tools, techniques and concepts that an
organization needs to bring products into the market more quickly and cheaply, in order to
realize the associated gains.
Organizations have experienced several tangible benefits from a good product development
process. For instance, the International Motor Vehicles Programme showed that while Japanese
manufacturers such as Honda and Toyota introduced as many as 85 models between 1982 and
1989, their American counterparts were able to introduce only 49 models. This significantly
affected the competitive positioning of these organizations.1 Another study analysed the market
impact of new-product introduction and showed that, by introducing products six months ahead
of the competitors, an organization can gain as much as three times the cumulative profit earned
over the life of the product.2
It is clear that product development is an important aspect of the operations management
function in every organization, be it services or manufacturing. An organization armed with a
good product development process will be in a better position to bring new products and
services to the market ahead of the competition and will be able to retain customers and its
market share in the sector. Repeat performances in the future will reward the organization with
an increasing share of the market in the long run. Before we discuss the product development
process in detail, it will be helpful to understand the expected outcomes of a good development
process. Table 11.1 lists the outcomes that one can normally expect from a good product
development process.
The product development process provides a broad set of tools, techniques, and concepts that enable an
organization to provide distinctiveness in its offerings by rolling out new products and services faster and at a
lower cost.
As we see from the table, four sets of benefits are likely to accrue to an organization from a
sound product development process. The first set of benefits—customer dimensions—relate to
attributes that have a positive and significant impact on the customer for whom the
products/services are designed. These include perceived quality, better fit with requirements, and
a sense of novelty. The second set of benefits pertains to performance over a sustained period.
Addressing environmental issues, reduced life-cycle costs, and user-friendliness belong to this
category. The third set of benefits accrues to the organization through operational advantages
gained in the product development process. Typically, ease of manufacture and assembly, along
with minimization of the need for going back to the drawing board are some examples of this
category. All these benefits will lead to strategic improvements in cost, flexibility, and the
development of new and unique capabilities.
An organization armed with a good product development process will be in a better position to bring new
products and services to the market ahead of the competition.
The product development process consists of a structured and orderly set of activities. These
activities are repeated at periodic intervals when the organization feels the need for bringing out
new products. Let us take the example of the development of the Tata Nano to understand the
process. The car initially had three concepts. The design team picked the one that they thought
looked best, and from that they made a full-scale model. In mid-2005, one model was completed,
which then went through a stage of refinement. At this stage, the initial volume of the car was
defined. The design team then worked on details like lamps and doors. From that point onwards,
the design team actually developed another model, which was a second-stage model. Eventually,
IDEA, the design house involved in this project, came up with the final refined model. The style
for the Nano was frozen in mid-2006—exactly a year after the first model was completed.
Understandably, this stage was important since it had to meet the cost targets. Once the style was
frozen, efforts were directed towards engineering development.3
Concept generation
The first stage in a product development cycle is concept generation. New products and services
are the outcome of the need to close the gap that exists in the market with respect to the
customers’ needs. Therefore, understanding customer needs, how exactly the existing portfolio
of products and services fulfils this, and the areas that need greater application and attention are
critical. Once there is some clarity on this aspect, the next step is to translate the need into
products and services. This activity requires good innovation skills to develop unique products
and services. Several alternative methods of fulfilling the identified need can be identified, but
the feasibility and novelty of each of these approaches need to be carefully established.
Feasibility includes several dimensions: the technological feasibility to manufacture and
deliver products according to the design marketing feasibility to create a market, distribute, and
sustain maintenance; cost feasibility; and the feasibility of customer acceptance, to name a few.
At this stage, feasibility is not evaluated in a strict and detailed manner. Instead, expert
judgement is used to validate the specifications, numbers and assumptions behind the concept.
This is a crucial step in the product development process.
Research & Development (R&D) involves several steps. Among these, the preliminary concept generation is a crucial
step as it shapes all the future activities with respect to the new product that is being brought out into the market. Maruti
Suzuki Limited’s video brings out the various steps in the early stages in the making of a new model of a car. To know
more about this, find the video links (Video Insights) in the Instructor Resources or Student Resources under section
Downloadable Resources (http://www.pearsoned.co.in/BMahadevan/) subsections Bonus Material.
Design
Once product feasibility is established, it is likely that the choice among several promising ideas
will be narrowed down to a few potential ones. The next step in the process is to incorporate
more details into the concept. Potentially useful ideas need to be analysed from the perspective
of physical attributes such as shapes, sizes, and material. Moreover, costs, manufacturing
processes, and specifications are the other aspects of the concept that require detailed
consideration before the decision to go ahead can be taken. The design phase of the product
development process focuses on these aspects.
During the design stage, detailed specifications are first drawn about the product. For instance,
if we are designing a new pen, it is important to fix the weight, the material, the surface finish,
and the tolerances required. Once these details are arrived at, they provide the necessary inputs to
explore how exactly these requirements could be met using the available production resources in
the organization. The process planning activity provides these details of the product being
designed. Product specifications and process planning decisions are required for cost estimation
of the proposed design of the process. At the end of these three steps, it is possible to assess the
feasibility of the design.
An initial design may not be feasible for several reasons. For instance, the capability of the
existing resources may not be enough to manufacture the product as per the required
specifications. In some cases, although capability exists, the cost of the design could appear to be
prohibitive, making it infeasible to find a good market. It is also possible that even when the cost
and the process feasibility are there, the product will turn out to be uninteresting from a customer
perspective. Finally, the availability of raw material or skilled labour may be doubtful in certain
cases. Therefore, during the design phase, the product goes through several iterations before the
design is approved. Several entities are involved in the decision making at this stage. These
include the design department, the finance department, the production planning department, the
marketing department, and the procurement department.
At the end of this exercise, a few ideas having potential are thoroughly analysed with a greater
level of detail provided by the design team, and one alternative is selected for commercial
production. In some cases, a certain amount of flexibility is still incorporated in the design, in
which case two or three versions of the product are developed further before the final choice is
made.
Development
The third stage of the product development process is the physical development of the product
itself. During this stage, the details arrived at on the drawing board are translated into reality.
Usually, a prototype is built for extensive testing and fine-tuning of design specifications.
Moreover, detailed manufacturing specifications and specific methods of manufacture, assembly,
and testing are established during this phase. Just as design specifications undergo a few
iterations before the specifications are firmed up, similar efforts are required at this stage to
iterate and firm up the manufacturing specifications. Design changes are made on the basis of
findings from multiple controlled experiments and product usage.
Development refers to the stage of the physical development of the product. During this stage, the details arrived
at on the drawing board are translated into reality.
Production
The last stage of the product development process pertains to the transfer of know-how to the
production personnel and establishing the system for volume production. Typically, during the
development phase, a few pieces of the product are manufactured for testing purposes. The
efficiency and effectiveness of the manufacturing process are not of concern at that point in time.
Moreover, dedicated equipment and assured sources of supply are not established at that stage.
Therefore, in production, the last phase, these aspects are addressed in a detailed fashion.
Moreover, systems are established to translate the initial learning-curve advantages into the
production, thereby improving productivity and bringing down the cost.
Production involves the transfer of knowhow to the production personnel and the establishment of a system for
volume production.
The product development cycle described in this section is fairly representative of a vast
majority of situations faced in practice. These steps are normally found even in the case of the
development of software products and services. In the case of the development of software
products and services, however, the specific set of activities undertaken at each stage will be
different from the ones discussed here. It is interesting to note that each stage in Figure 11.1 is
separated by a decision box. In reality, these could be considered to be the review points in the
product development cycle for the top management. Top management can measure the progress
of the product development process and take some “go–no-go” decisions based on additional
information available both within the organization and in the external environment.
While a number of creative ideas need to be generated during the initial stages of the product
development process, as the process continues, these alternative choices need to be narrowed
down to either one or a few variations of a basic version. The decision boxes shown in Figure
11.1 enable the management to achieve this objective.
An alternative representation of the product development process is the product development funnel with stages
and gates.
The entire process could be thought of as a funnel filtering ideas as time goes by. Similarly, the
completion of each stage is a milestone that could potentially trigger the management to filter the
available ideas, assess the suitability of current efforts and make necessary modifications.
Therefore, each gate represents an opportunity for review and mid-course correction.
FIGURE 11.2 A product development funnel with stages and gates
Product development is an interdisciplinary effort, as we saw from the previous section. Several
functional areas are directly involved in the process. Marketing plays a crucial role in the process
by providing information on customer needs that are otherwise unaddressed. Moreover, as the
concept is established and design and manufacturing specifications are finalized, there is a
frequent need for interacting with customers for useful feedback regarding the process. Finance,
design, and production planning functions also play a major role in translating the concept to
meaningful products that can be manufactured at an attractive cost. Profit planning over the
lifetime of the product is also essential. Therefore, there is constant interaction between the
various functional areas in an organization for product development.
Currently, most of the development does not happen in a purely in-house fashion. Identifying
component suppliers fairly early in the process, involving them in the process from the
beginning, and ramping up their capacities and capabilities are important aspects of a faster and
more efficient new-product development process. For instance, an alternator manufacturer such
as Lucas TVS procures complete information on the developments in bearing design before
designing its next generation of alternators or starter motors. Since bearings form an important
component of its product, interfacing with bearing manufacturers and getting information on new
developments makes its product development process not only faster but also superior.
Therefore, the material procurement function is also involved in the process.
In view of the interdisciplinary nature of these entities and the intensity of interaction required
among them, an appropriate organization structure is vital for a good product development
process. Traditionally, the organizational structure for the product development process was one
of functional silos. In this approach, each functional area addressed its part in the product
development process in isolation. Moreover, the functional areas addressed the product
development process one at a time. After one group completed its part, it was passed on to
another group for further processing. If there were infeasibilities, the process was reversed and
the part was sent back to the earlier department. Figure 11.3(a) depicts this arrangement.
It is obvious that such an approach is time consuming. There are excessive delays and the
process results in poor productivity. A long lead time for product development creates several
problems for organizations, especially in the context of heightened competition from global
players. Most prominent among the problems is that several assumptions about the market,
pricing, and competing products may have been wrong. For instance, if an organization takes 48
months to develop and bring a product to the market, by the time the product is introduced its
assumptions about the competition might have changed substantially. The second issue is the
cost of development. A longer process would clearly mean greater development costs. Since
these costs are sunk even before the first commercial version of the product rolls out of the
manufacturing plant, it puts enormous pressure on the marketing and production functions.
However, in recent times, significant changes have been introduced in the organizational
structure for creating a product development process that promises a dramatic improvement in
the development lead time. Figure 11.3(b) shows an alternative approach known as concurrent
engineering. The basic idea behind this is to put together a team of cross-functional
professionals and provide them with the necessary resources and mandate for the product
development process. The cross-functional team works in project mode. Once the process is
over, the team is dismantled and the members return to their parent functional units. By co-
locating various functional requirements for the product development process, organizations
could benefit in several ways:
Concurrent engineering is the process of putting together a team of cross-functional professionals and
providing them with the required resources and mandate for the product development process.
The development lead time could be more than halved in this process, enabling the organization to bring new products
into the market faster. In the long run, this would help the organization capture a greater market share.
Since various functional requirements are simultaneously available, the number of design changes after the product is
introduced into the market will come down. This is because the team members would have had more opportunities to
analyse the impact of the design from multiple perspectives before approval. Therefore, design productivity also
improves.
A concurrent approach enables organizations to involve its suppliers early on in the product development process.
Therefore, suppliers are able to cut lead time for capacity augmentation and capability building with respect to the new
product being developed.
A team structure promotes consensus-based decision making and increases the propensity for collective risk-taking.
These attributes are very valuable for a product development process, especially in the early stages. In the absence of
this, important decisions for resolving trade-offs between one design choice and another will not be made efficiently,
and the process will be time consuming.
Several tools and techniques can be used for efficient new-product development. These tools
address various stages of the product development process that we discussed earlier. In this
section, we shall look at some of the tools that are available for product designers to understand
customer needs and translate them into meaningful design and manufacturing specifications, and
also at some guidelines for incorporating the manufacturing requirements at the design stage.
Understanding Customer Needs
The first step in a product development process is to know what exactly the new product is going
to be. The concept generation phase of the product development process addresses this. There are
various methods by which organizations can arrive at an appropriate concept for the new product
that it intends to develop. Irrespective of the method used, understanding customer needs is an
important input to this process.
Market research
Organizations use a set of known tools for soliciting information. For instance, they could
commission a market research organization for this purpose. In a market research, a target group
is identified and appropriate sampling is done within the target group. Using a structured data
collection method such as questionnaire surveys and interviews, information is solicited from the
sample. This information is subjected to statistical and other kinds of analytical reasoning before
arriving at customers’ preferences and needs.
An alternative method is to use focus groups. Unlike the earlier method, in a focus group
several individuals from the target group are simultaneously met and interviewed together. The
additional advantage is the reinforcement of certain ideas and needs through collective reasoning
and group-behaviour norms. Suppose an organization is developing a new beverage and its target
market for this product is youngsters in the age group of 15–25. One way to get useful inputs for
product design is to have a detailed and continuous interaction with a group of 25 to 50
youngsters in the target age group. As various aspects of the product are being designed, this
small group of youngsters will act as a sounding board.
The other option is to conduct an in-depth qualitative interview with a selected set of potential
customers on an individual basis. In this approach, rather than collecting some quantitative data
from a large number of respondents, a considerable amount of qualitative information is
collected from a chosen few. The data collected in the above manner is subjected to extensive
analysis.
In the case of in-depth interviews and focus group meetings, it is possible to analyse the
qualitative statements elaborately for identifying possible missing links in the existing offerings
in the market. Efforts are often made to exactly reproduce the customers’ own words and
alternative ways of translating them are attempted. In the case of market research using a
questionnaire survey, several statistical tests are performed. The testing of hypotheses pertaining
to customers’ likes and dislikes and the performance of existing offerings in the market is also
done. Irrespective of the method adopted, the exercise must eventually lead to creating a bundle
of attributes that customers are looking for in the product and identifying the hierarchies, if any,
among them. This forms a critical piece of input to the product design process. Figure 11.4
shows the various steps in the process explained here.
Competitor analysis
An understanding of what the existing offerings are and how the gaps and problems identified
can be eliminated can offer valuable inputs to concept generation. This method is a proxy for
understanding customers’ needs. It is useful in situations where it is either difficult or costly to
reach the customer directly. Moreover, this process could also supplement other methods of
arriving at a good concept of the product that needs to be developed.
An understanding of what the existing offerings are and how the gaps and problems identified can be eliminated
can sometimes offer valuable inputs to concept generation.
One method of competitor analysis is reverse engineering the product. In reverse engineering,
the competitors’ products are dismantled down to the level of individual components and some
detailed studies are conducted on them. These may sometimes reveal the probable processes
utilized in their manufacture, the choice of materials and their specifications, and the relationship
between these parameters and performance. In some cases, the competitors’ products are also
subjected to alternative working conditions to understand the latitude the products provide with
respect to performance. Whether reverse engineering is ethical is not an easy question to answer,
as some critics opine that a violation of professional ethics is involved in this process. There are
arguments for and against it. In the case of services, reverse engineering is not possible in most
situations.
FIGURE 11.4 Understanding customer needs
Development of a new service has similar steps to that of development of a product. However, there are several unique
aspects that a service provider goes through before commercially launching a new service. HIFI, a new service in the
health and fitness sector by Talwalkars is a case in point. To know more about this, find the video links (Video
Insights) in the Instructor Resources or Student Resources under section Downloadable Resources
(http://www.pearsoned.co.in/BMahadevan/) subsections Bonus Material.
Reverse engineering is one crude method within the larger issue of benchmarking. In the case
of benchmarking, the competitive product offerings are chosen for a detailed analysis. Specific
parameters are chosen for the benchmarking exercise. For example, cost, features, performance,
ease of maintenance, ease of manufacture, assembly, and distribution are some of the issues on
which a comparative study may be possible. Once the parameters are identified, data collection
and analysis will reveal the positioning of one’s own product vis-à-vis the competitors’ offerings
(for an example, see Ideas at Work 12.3 in Chapter 12).
Another method for competitor analysis is to develop perceptual maps. Perceptual maps are
graphical representations of the various competitor offerings and that of the proposed
product/service. Although it is similar in approach to the benchmarking exercise, the graphical
display makes it easier for managers to understand the implication of the proposed positioning of
the product for which the development processes are undertaken. However, the major
disadvantage of this approach is the limitation on the number of dimensions on which
competitive offerings can be compared and maps generated. While a two- dimensional plot is
easy to comprehend, a three-dimensional plot is somewhat manageable. It is not possible to
visualize beyond this. However, in reality, there could be more than three dimensions on which
competitive offerings can differ significantly.5
Perceptual maps are graphical representations of the various competitor offerings and that of one’s own
proposed product/service.
Quality function deployment (QFD) is a tool that helps organizations translate customer requirements into
design, process planning, and manufacturing specifications using a four-stage process.
Quality function deployment achieves this transition in four stages. In each stage, certain
inputs are considered, and based on these inputs certain specifications are arrived at. The
specifications arrived at in the previous stage form the input to the next stage. Proceeding in this
manner, customer requirements are converted into design, process planning and manufacturing
specifications.
The first stage links customer needs to the design attributes required. In the second stage, the
design attributes form the basis for actions that the organization needs to take to achieve these
attributes. The actions identified at this stage are the basis of the third stage in arriving at the
specific decisions to be implemented. Finally, in the fourth stage, the implementation decisions
drive the process plans to be deployed. Each stage is known as the “house of quality.” Figure
11.5 is an illustration of the house of quality pertaining to the first stage. Figure 11.5(a)
illustrates its use in the case of a restaurant. Figure 11.5(b) is representative of the structure of a
house of quality and the type of data used for analysis and decision making.
Seven types of data are used in this house of quality. The first type is the customer
requirement. For example, in the case of the restaurant, the fact that the customer would like to
spend less time during peak hours is an important requirement. When several such requirements
are identified, it is important to also know if some are more important than others. The important
data is captured such that it helps the design team prioritize its efforts.
The next kinds of information required are the product characteristics and the relationship
between customer requirements and product characteristics. This information is also captured in
the house of quality [see Items 3 and 5 in Figure 11.5(b)]. For example, “less time during peak
hours” clearly relate to the number of service counters during peak hours. The more the number
of counters, the less is the time spent by the customer during peak hours. It is obvious from this
example that some of these service characteristics are inter-related. Order-processing time and
the number of service counters during peak hours are strongly and negatively correlated. In
certain other cases, the correlation can be either weak or zero. It can also be positive or negative.
These trade-offs are also captured in the house of quality [see Item 4 in Figure 11.5(b)].
Finally, before focusing on the design effort, it is also important to benchmark one’s own
product or proposed specifications with that of the competition. The right side of the house of
quality provides inputs from the benchmarking exercise. Once the above information is collected
and analysed, it is possible to fill in the lower portion of the house of quality [Item 7 of Figure
11.5(b)]. While doing this exercise, it is again useful to know how important the changes are vis-
à-vis the competition and make appropriate recommendations. The lower portion of the house of
quality is the output of this stage of the exercise and this is taken to the next stage as input.
Proceeding in this manner, all the four stages of the house of quality can be constructed.6
Value Engineering
During the early stages of design, it is important to question several assumptions pertaining to
design, process planning, and the manufacture of components in a product development process.
Since the specifications are not frozen at that stage, investigating alternative methods can
potentially yield better products at a lower cost. Value engineering is a set of activities
undertaken to investigate the design of components in a product development process strictly
from a cost–value perspective and alert the product development team to alternatives that could
either bring down the cost, or increase the value by improving the functionality and performance
without increasing the cost.
Value engineering is a set of activities undertaken to investigate the design of components in a product
development process strictly from a cost–value perspective.
The Tata Nano project had ambitious cost targets to bring to the market an ₹100,000 car.
Expectedly, the suppliers had to make use of value engineering efforts. The components in the
Nano have been studied from the points of view of functionality, cost and performance. Rane
Group, which makes a rack-and-pinion steering system, focused on reducing the weight of the
materials used, replacing the steel rod of the steering with a steel tube, which turned out to be a
major cost reducer. The product is typically made of two pieces, but it was redesigned into one to
save on machining and assembling costs. GKN Driveline India, a subsidiary of global auto parts
leader GKN, designed a shaft of a smaller diameter, which made it lighter and saved on material
costs.7
FIGURE 11.5 The house of quality
*Source: Based on J. R. Hauser and D. Clausing, “The House of Quality,” Harvard Business Review, May–June 1988.
Typically, design professionals brainstorm various options in conjunction with procurement
personnel, suppliers, and production personnel with respect to the value–cost dimensions of the
product being developed. Usually, several questions are raised, which include the following:
Can we eliminate certain features from the design?
Are there instances of over-design in certain components, increasing the cost? If so, how can these aspects be
rationalized?
Are there certain features of design that cost more than they are worth?
Is it possible to replace the proposed method of manufacture with a less costly one?
Is it possible for someone else (suppliers) to produce certain components cheaper, faster and better?
Can we eliminate some parts and replace them with more universal parts?
Are there opportunities for cost cutting by development of import substitution methods?
Through a series of such brainstorming exercises, some cost reduction opportunities are
identified and necessary steps are taken to incorporate them in the design. A good value
engineering exercise results in a cost-effective design that maximizes value for target customers.
Value engineering exercises succeed only when the top management provides a mandate for
such exercises and demands better performance. In the absence of this, the complexity and long-
drawn nature of the product development process may result in the product development team
pushing this to the background.
Design for manufacturability (DFM) is a structured approach to ensure that manufacturing requirements and
preferences are considered fairly early in the design process without the need for extensive coordination between
the two.
Reducing the variety: The following guidelines can be followed for reducing the variety:
Minimize the number of parts
Minimize sub-assemblies
Avoid separate fasteners
Use standard parts when possible
Design parts for multi-use
Develop modular design
Use repeatable and understood processes
Reducing cost: The following guidelines can be followed to reduce costs:
Analyse failures
Assess value rigorously
Considering operational convenience: The following are some guidelines for operational convenience:
Simplify operations
Eliminate adjustments
Avoid tools
Design for minimum handling
Design for top-down assembly
Design for efficient and adequate testing
Mass customization provides a structured set of ideas and tools to provide high levels of customization without
increasing the complexity of planning and control of operations.
The product-platform concept is built on two basic premises. The first one is the notion of
distinctiveness and the second is the degree of commonality. Let us consider a passenger car. In
their choice of one brand of car over the other, customers care for a few features. These include
aspects such as the noise level inside the car, the upholstery design, the choice of colour options
available, and ease of steering. Distinctiveness, therefore, refers to the set of differentiating
attributes between one variant and the other in a family of similar products, both within an
organization and outside (competitors). On the other hand, consider the engine compartment or
various subsystems mounted inside the front hood of the car bonnet. Even a technically inclined
customer is less likely to open the bonnet and closely inspect or touch and feel these components
before making a choice among competing variations. These are known as “chunks.”
Distinctiveness refers to the set of differentiating attributes between one variant and the other in a family of
similar products.
While differentiating attributes provide a clear advantage to a product and position it strongly
among competing products, this also calls for adding more varieties. Therefore, there will be an
increase in the design effort to create these attributes uniquely. Ultimately, it will result in an
increase in both the cost of the design and the product. In contrast, chunks influence costs
favourably and help an organization use these across a family of offerings, compensating for the
increased cost of the differentiating attributes. Therefore, the objective of a product-platform
exercise is to critically arrive at a tradeoff between the extent of usage of differentiating
attributes and chunks in the design of a family of products. Figure 11.6 graphically depicts the
trade-offs involved in this process.
Typically, organizations practicing the traditional product development process will be in
Stage 1 of the trade-off process, as depicted in Figure 11.6. In this scenario, even when
organizations have fewer common parts, the products may be less distinctive. Such a situation
shows that organizations have not been able to benefit from some of the advantages of a good
product development process. However, there are other coordinates in the trade-off map that
organizations can hope to achieve. For instance, even if the organization is not able to increase
the distinctiveness of its product offerings, it can make an effort towards increasing the number
of common parts in the existing portfolio of the products. This is indicated by the Trajectory A in
the figure. Greater emphasis on the use of variety reduction techniques may achieve this goal and
the organization will reach Stage 2 in the product development process.
FIGURE 11.6 Trade-offs in a product-platform exercise
B indicates the second trajectory in the figure. In this approach, organizations try to increase
their product distinctiveness without any increase in unique parts. Therefore, while the degree of
commonality of parts does not change, the number of offerings increases significantly. One can
achieve this by taking a conscious decision to change the design philosophy to a modular
approach. By a careful design of the module with the existing components, the organization will
be able to offer a wide range of distinctive products to its customer base. Such an approach
requires an intensive design effort, compared to the variety reduction drive. This will take the
organization to the third stage of its product development process. The final stage is the judicious
combination of both the approaches using the principle of product platform. In reality, there
could be more trajectories available to reach the fourth stage than those indicated here.
The objective of a product-platform exercise is to critically arrive at a trade-off between the extent of usage of
differentiating attributes and chunks in the design of a family of products.
From the foregoing analysis of the product development process, it is possible to enumerate a set
of measures to judge the performance of the product development process. Table 11.3 presents a
representative list of possible measures that an organization could use for this purpose. Four sets
of measures are proposed. Cost-based measures seek to relate the design effort to the cost
incurred during the design stage and the cost after the design is utilized for production. The first
two items in this list assess the cost impact during the design process, and the next two assess the
impact after the design is approved.
Another set of measures pertains to the effectiveness of the design effort. There are several
measures that an organization could use for assessing design effectiveness. Of particular interest
are the measures “time to return to normal quality” and “the number of revisions in the design”.
It is common that once the design is approved and the actual production begins, organizations
experience certain difficulties and make revisions in the product design. Usually, a document
known as the engineering-change notice (ECN) is generated for this purpose. Therefore, the
number of ECNs is indicative of the design effectiveness. Depending on the effectiveness of the
initial design effort, it takes more or less time for the ECNs to taper down to very few. Once it
reaches this stage, it is reasonable to assume that the product has returned to normal quality
requirements of an established product. It was found that in Japanese organizations such as
Toyota, the time to return to normal quality was much less compared to their American
counterparts during the early 1990s. This shows the effectiveness of the product development
process in Japan.
TABLE 11.3 Performance Measures for the Product Development Process
Cost-based measures
Target cost achievement status
Concept to market
Strategic measures are also available for assessing the performance of a product development
process. Popular among them are the “time to market” and the “number of new products
introduced in the market”. You may recall that at the beginning of the chapter we showed how
these measures are important and what benefits organizations get from these. The final set of
measures pertains to the nature of the impact a product development process can create in the
market. Product development efforts are primarily undertaken to improve an organization’s
market share and overall positioning. Therefore, using some relevant measure to assess the
product development process is a worthwhile effort.
Recent developments in the field of management accounting have helped the product
development process in more ways than one. Researchers in the management accounting domain
have shown that the final cost of products and services offered to customers is to a large extent
determined at the design stage. Therefore, they have advocated ways and means of controlling
them at that stage itself. Two concepts that management accountants have proposed are of great
relevance to the product development process. These are life-cycle costing and target costing.
Traditional cost accounting and control concepts focus on the cost of products and services at
the manufacturing stage. As opposed to this, life-cycle costing is a methodology that focuses on
the cost and control aspects of a product or service throughout its life cycle. The life-cycle
approach clearly means that the concept generation and design stage are also to be analysed for
cost control purposes.
Life-cycle costing is a methodology that focuses on the cost and control aspects of a product or a service
throughout its life cycle.
Life-cycle costing studies point to an important aspect pertaining to controlling product costs.
Figure 11.7 depicts this graphically. It is important to understand two types of costs over the life
cycle of a product. The first one is the committed cost of the product. By committed, we mean
that no matter when we incur the cost, the amount that we incur is fixed much earlier. For
instance, the moment we finalize a design for a product, we will be able to estimate how much
the cost of the product will be once manufacturing commences. This is simply because, at the
time of design we fix all the required specifications with respect to the type and amount of
material, labour, machines, and other resources to be used to manufacture one unit of the
product. The second type of cost, incurred cost, denotes the actual cash outflow with respect to
manufacturing the product. As shown in the figure, 95 per cent of the cost of the product is
committed even before the first unit rolls out of the manufacturing facility. On the other hand,
only one-third of the cost is actually incurred (in terms of the overheads pertaining to product
development) before commercial production begins. This clearly brings out the importance of
the product development process and the need to have a good monitoring and review mechanism
during the early stages of product development.
FIGURE 11.7 Life-cycle costing: committed and incurred product costs
Target costing is another useful concept for the product development process. Traditionally,
organizations have been managing the cost of the products and services with a broad philosophy
known as cost-plus. In a cost-plus approach to cost management, organizations incur certain
costs for manufacturing a product or for providing a service. They add a desired margin to this to
arrive at the price. In simple terms, Price = Cost + Margin. However, in recent times, as
competition has intensified, organizations realize that market forces increasingly dictate the price
a product or a service may fetch. Moreover, organizations require a certain margin to remain in
business and continue to invest in new technology. Therefore, the equation is reversed in this
scenario as Cost = Price − Margin. This would mean that in order to remain competitive in the
market and continue to earn some margin; an organization has to keep its costs on target, as
specified by the equation.
Target costing is a methodology to identify cost reduction methods and pursue them relentlessly so that as the
product is designed and manufactured, it is within the target set.
The process followed to design, develop, code, test, and deliver a software product, from the
inception of an idea to the delivery of the final software to the customer, is called software
product development. Typically, a software development process consists of four stages:
requirement analysis, design, development (coding), and testing.
A software development process consists of four stages: requirement analysis, design, development (coding),
and testing.
Requirements analysis is the process of specifying requirements by studying user needs and
systematically analyzing and refining the specifications. A specification, the primary result of
requirements analysis, is a concise statement of the requirements that the software must satisfy. It
has often been reported that deficient requirements analysis is a major cause of software project
failure. On the other hand, identifying the exact requirements is the most difficult part of a
software product development process. Traditionally, requirements analysis is done using a set of
techniques such as interviewing, questionnaire surveys, and document searching, to elicit
requirements from the customer.
After the requirements have been defined, the software development process involves various
levels of design. Design produces a complete and precise specification of the software to be
developed. The specifications define the principal parts of a software package, describe how
these parts interact, and specify how they are integrated to produce the final software. A common
problem in designing large software systems is the need to define the overall system structure
before the team can specify other details. A principal benefit of teams is their potentially wide
range of skills and knowledge. Another issue in the software design stage concerns the effective
use of all team members’ ideas.
During the development stage, the actual code is delivered to the customer as the running
system is developed. In this stage, actual programs are written using programming languages.
The output of this phase is an implemented and tested collection of modules. Coding is subjected
to company-wide standards, which defines the entire layout of programs, such as headers for
comments in every unit, naming conventions for variables and sub-programs, the maximum
number of lines in each component and other aspects that the company deems worthy of
standardization. Further, in the context of software development projects, there are standard
methods of development available in manuals. Some of the development codes can also be
referred from the organization’s databases.
In the final stage of the software development process, all the modules that have been
developed and tested individually are integrated and tested as a whole system. As testing is the
last stage of the software development process, the team needs to ensure that the quality of the
product is as per the customer’s requirements. Thus, during this stage, the team needs to be
focused and should have the goal of delivering a high-quality software product with minimum
errors or problems. It should also ensure that the product is delivered to the customer on time.
Various software development process models are available. This includes the linear
sequential model or waterfall model, the prototyping model, the rapid application development
(RAD) model, the incremental model, the spiral model, the component model, and the
concurrent development model.14 From the description given here, it is clear that several
similarities exist between the product development process and the software development
process. For example, the major steps in product development are concept generation, design,
development, and manufacturing (commercialization), while the steps in software development
usually include system conceptualization and project definition, requirement analysis, design,
programming, testing, and implementation. Thus, both new-product development and software
development go through similar development stages and usually involve cross-functional teams.
Moreover, similar external and internal factors influence the development process in both
domains.
SUMMARY
Operations management addresses the issue of innovation through the product development process. An
organization armed with a good product development process can bring better products and services into the
market ahead of the competition.
The product development process consists of a structured and orderly set of activities including concept generation,
design, development, and production.
Concurrent engineering is an approach to build a cross-functional team of professionals cutting across various
departments within an organization as well as the suppliers to accelerate the new-product development process.
Quality function deployment (QFD) is a process by which the qualitative attributes of customer needs are translated
into a bundle of quantitative attributes pertaining to design, process planning, and manufacturing specifications for
products in an organization.
Design for manufacturability (DFM) principles outline a set of guidelines that help organizations reduce variety
and cost and improve the operational convenience of manufacturing while designing new products.
Variety reduction through standardization, modular design and product platforms are the tools of mass
customization.
Management accounting provides tools such as target costing and life cycle costing for efficient new-product
development.
A software development process consists of four stages: requirements analysis, design, development (coding), and
testing and delivery.
REVIEW QUESTIONS
1. How important is it to have a good product development process? What do organizations gain from this?
2. How will you determine if an organization does or does not have a good product development process in place?
3. Briefly sketch the product development process.
4. Is it prudent to exercise the same type of management control and evaluation for all the stages of a product development
process? Explain and justify your answer.
5. Given below are three attributes of the management team working in a product development project. Explain which of
these are appropriate at different stages of the project:
6. Why is a product development process analogously linked to a funnel with stages and gates?
7. Does the organization structure influence the product development process any way? Explain.
8. What, in your opinion, is the most appropriate organization structure for a product development process?
9. What are the alternative methods available to a product development team for generation of the concept for a new
product?
10. What do you mean by quality function deployment? Of what use is it to the product development process?
11. How does the use of DFM guidelines help a product development process?
12. What is the relevance of mass customization principles to the product development process? What are the alternatives
available to an organization to enable itself with mass customization?
13. How can the effectiveness of the product development process be measured?
14. Does management accounting play any role in the product development process? If so, how?
15. Compare and contrast the traditional product development process with the software product development process.
NET-WISE EXERCISES
1. Headquartered in Seoul, South Korea, Hyundai operates the world’s largest integrated automobile manufacturing facility
in Ulsan, which is capable of producing 1.6 million units annually. The company employs about 75,000 people around the
world. Hyundai vehicles are sold in 193 countries through about 6,000 dealerships and showrooms worldwide.
Visit their worldwide website at http://worldwide.hyundai.com/WW/Main/index.html. In the lower part of the page
there is a header on innovation with three sub-links – Design, Technology, and Eco. Visit each of these links to get to
know the context for Hyundai for new product development.
Prepare a report on the context and business issues that Hyundai faces in the new product development process. Can
you identify some broad directions in which Hyundai is taking forward its new product development process?
2. LG Electronics was established in 1958 and has since led the way into the advanced digital era thanks to the technological
expertise acquired by manufacturing many home appliances such as radios and TVs. LG Electronics has unveiled many
new products, applied new technologies in the form of mobile devices and digital TVs in the 21st century, and continues
to reinforce its status as a global company. Visit their website at http://www.lg.com/in/design to understand their approach
in bringing new products to the customers. Also visit their page on innovation by clicking on the web page
http://www.lg.com/in/innovation.
Prepare a two-page report on the overall approach of LG in bringing new innovations into the market. What are the
various organizational mechanisms they have in place in order to fruitfully bring new products to the market? Have
they been successful in their innovation efforts?
CASE STUDY
SUGGESTED READINGS
Y. Akao, Quality Function Deployment: Integrating Customer Requirements into Product Design (New York:
Productivity Press, 1990).
R. Dhawan, “How to Develop the Best New Products,” Business Today, 22 June 1999, pp. 74–85.
J. H. Gilmore and B. J. Pine II, “The Four Faces of Mass Customization,” Harvard Business Review 75, no.1
(1997): 91–101.
J. R. Hauser and D. Clausing, “The House of Quality,” Harvard Business Review 76, no. 3 (1988): 63–73.
H. L. Hee, C. Billington and B. Carter, “Hewlett-Packard Gains Control of Inventory and Service Through Design
for Localization,” Interfaces 23, no. 4 (1993): 1–11.
R. S. Kaplan, “Management Accounting (1984– 1994): The Development of New Practice and Theory,”
Management Accounting Research 5 (1994): 247–260.
Y. Kato, “Target Costing Support Systems: Lessons from Leading Japanese Companies,” Management Accounting
Research 4, no. 1 (1993): 33–47.
J. Matthing, B. Sandén, and B. Edvardsson, (2004), ‘New service development: learning from and with customers’,
International Journal of Service Industry Management, Vol. 15 (5), pp 479–498.
L. J., Menor, M. V. Tatikonda, and S. E. Sampson, (2002), ‘New service development: areas for exploitation and
exploration’, Journal of Operations Management, 20 pp 135–157.
M. H. Meyer and A. P. Lehnerd, The Power of Product Platforms (New York: The Free Press, 1997).
B. J. Pine II, B. Victor and A. C. Boynton, “Making Mass Customization Work,” Harvard Business Review 71, no.
5 (1993): 108–117.
R. S. Pressman, Software Engineering: A Practitioner’s Approach (New York: McGraw-Hill, 1997).
D. Robertson and K. Ulrich, “Planning for Product Platforms,” Sloan Management Review 39, no. 4 (1998): 19–31.
M. V. Tatikonda, and V. A. Zeithamal, (2002). ‘Managing the New Service Development Process: Multi-
disciplinary literature synthesis and directions for future research’ Chapter 10 in Boone, T and Ganeshan, R, (Eds,),
New directions in supply chain management: Technology, Strategy and Implementation, American Management
Association, New York 2002, pp 200–233.
R. Verganti, (2011), ‘Designing breakthrough products’, Harvard Business Review, Vol. 89 (10), 114–120.
S. C. Wheelright and K. Clarke, “Competing Through Development Capability in a Manufacturing-based
Organization,” Business Horizons 35, no. 4 (1992): 29–43.
J. P. Womack, D. T. Jones and D. Roos, The Machine That Changed the World (New York: Rawson Associates,
1990).
K. H. Yeh and C. H. Chu, “Adaptive Strategies for Coping with Product Variety Decisions,” International Journal
of Operations and Production Management 11, no. 8 (1991): 35–47.
CHAPTER 12
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Who were the well-known quality gurus and what did they advocate for better quality management?
How can we define quality? Are there implications for quality management when we change our definition of
quality?
What do we mean by total quality management (TQM)? What are the major components of a TQM system?
What are the quality control tools used by organizations today?
What are the well-known quality certification organizations and awards instituted today?
What do you understand by the term “service quality”?
What are the elements of a quality assurance system?
The fishbone diagram (also known as the cause–effect diagram) is one of Karou Ishikawa’s several contributions to the field of
quality management. It is widely used in several manufacturing and service organizations worldwide for continuous quality
improvement.
ideas at Work 12.1
FIGURE 12.1 Wrong loading of the component for the second operation
Modern-day quality management is characterized by the use of several simple yet powerful tools
such as those used by Brakes India Limited. It also requires that the employees participate in the
improvement process and take initiatives that help the organization achieve superior quality in
the long run. In this chapter, we shall look at various aspects of quality management. Globally,
several organizations have adopted new quality management practices to compete effectively in
the market.
The resounding success of several manufacturing and service firms in recent times has
invariably been linked to excellent practices pertaining to quality management. If you consider
the auto-component manufacturers in India, many of them have won the Deming Award for
quality, the largest number outside Japan. Similarly, India has the largest number of CMM Level
5-certified software companies in the world. With such international recognition in quality, these
two sectors of our industry were able to compete globally. In the past, firms such as Harley
Davidson faced critical trade-offs while they made an effort to improve quality. For instance,
providing a high-quality product would have meant higher cost, or compromising on flexibility
or delivery commitments to the customer. However, successful firms like Kawasaki
demonstrated that they need not make any trade-offs with cost or other performance attributes
such as delivery and flexibility to offer better quality. On the other hand, they showed that it was
possible to make significant improvements in quality and at the same time make the products or
services available at a lower cost than that offered by the competition. Much of the success
enjoyed by Japanese manufacturers during the 1970s and the 1980s is attributed to this
phenomenon.
The resounding success of several manufacturing and service firms in recent times was invariably linked to
excellent quality management practices.
In this chapter, we shall see how traditional thinking on quality management has given way for
an alternative known as total quality management (TQM). We will also see various components
of a good TQM programme and the enabling mechanisms required for implementing a TQM
programme.
The success of Japanese manufacturing firms in the 1970s and the 1980s is largely attributed to a
quality revolution sweeping across these firms. Japanese manufacturing firms penetrated
established markets in the West by the sheer performance gaps that they created through the
quality of the products that they brought to the market. Based on a study of air-conditioner
manufacturers in both Japan and the United States during the early 1980s, David Garvin
concluded that the best air-conditioner manufacturer in Japan brought the product to the market
at a quality level 1000 times better than the worst manufacturer in the United States.1 He further
showed that the extra cost of making good-quality Japanese goods was much lower than the cost
of fixing defective products made by the American manufacturers.
The quality revolution was possible because of new thinking on what constituted good quality
and methods. Several new tools were utilized to assess the performance of an organization with
respect to quality. Over time, several other tools helped organizations make sustained
improvements in quality. At the same time, the top management realized the need for addressing
certain mindset issues and providing a leadership role. The roles of middle managers and
supervisors underwent a change from control to facilitation of the process of building quality into
products and services. Workers were responsible for not only output but also quality.
Such improvements and wide-ranging changes did not happen all of a sudden. A rich
repository of knowledge developed by several influential thinkers facilitated this process.
Alternative definitions were coined for quality and people in organizations were trained in the
new scheme of things. All these eventually led to the quality revolution.
The progress made in quality and productivity by organizations is mainly due to the teachings of
quality gurus. Between 1950 and 1980, several people suggested alternative approaches to raise
quality standards. Therefore, without knowing the salient aspects of these teachings, we may not
be able to understand the current best practices in quality management. We shall focus on the
teachings of some of the well-known quality gurus.
Deming proposed a four-step process, plan–do–check–act, popularly known as the PDCA cycle.
Joseph M. Juran
Juran visited Japan immediately after Deming and imparted to Japanese business executives the
key principles behind quality management. Juran’s approach to quality management
complements that of Deming’s. He believed that the quality problems faced by most companies
are due the constraints imposed by the top management rather than by those at the operational
level. Juran proposed a simple definition of quality—“fitness for use”. This definition suggests
that quality be viewed from both external and internal perspectives. The interdependency of
these functions emphasizes the need for company-wide quality management. Juran also insisted
that the senior management play an active role in the quality management process.
Juran emphasized the need for investing in superior quality systems, including a sound quality
planning infrastructure. His approach to quality management focuses on three important aspects,
known as Juran’s trilogy:
Quality planning: The process of preparing to meet quality goals
Quality control: The process of making quality goals during operations
Quality improvement: The process of breaking through to unprecedented levels of performance
Juran’s approach to quality management focuses on three important aspects known as Juran’s trilogy: quality
planning, quality control, and quality improvement.
Philip B. Crosby
Philip B. Crosby was the corporate vice-president for quality at International Telephone and
Telegraph (ITT) in Japan for 14 years, after working his way up from the position of a line
inspector. The essence of Crosby’s quality philosophy is embodied in what he calls the
“absolutes of quality management”. The absolutes are:
Absolute I: Definition of quality is conformance to standards
Absolute II: The system of quality is prevention
Absolute III: The performance standard is zero defects
Absolute IV: Measurement of quality is the price of non-conformance
Absolute V: There is no such thing as a quality problem
The five absolutes of quality provide a refreshingly different view on the issue of quality
management. Crosby believed that the performance standard should be zero defects and hence
popularized the phrase “do it right the first time”. “Zero defects” implies that errors should not be
expected or accepted as inevitable. Through a system of quality costing, Crosby provided a
practical application of his absolute that measurement of quality is the price of non-conformance.
Quality cost data are useful to draw the top management’s attention to the problem, to select
opportunities for corrective action, and to track quality improvement over time. We shall see in a
later section of this chapter how the concept of quality costing can help organizations improve
quality management practices.
Crosby introduced the notion of zero-defect performance and hence popularized the phrase “do it right the first
time”.
Karou Ishikawa
Karou Ishikawa was the most prominent figure in Japanese quality management until his death in
1989. Dr Ishikawa influenced the development of a participative, bottom-up approach to quality
management. He promoted greater involvement of all employees, from the top management to
the front-line staff, reducing reliance on quality professionals and quality departments. He
advocated collecting and analysing factual data using simple visual tools, statistical techniques
and teamwork as the foundation for implementing quality. Amongst his several contributions, the
fishbone diagram (also known as the cause– effect diagram) is widely used in several
manufacturing and service organizations worldwide for continuous quality improvement. In a
later section of this chapter, we shall discuss several quality control tools.
In a nutshell, Dr Ishikawa promoted the concept of company-wide quality. To help implement
the philosophy of participation and to put the quality control tools to gainful exploitation in the
hands of everyone in the organization, he developed quality control (QC) circles.
Ishikawa promoted the concept of company-wide quality through use of quality control (QC) circles. He also
proposed the fishbone diagram to analyse quality problems at the workplace.
Shigeo Shingo
Shigeo Shingo’s primary contribution to quality management is to suggest an approach for
complete elimination of errors. He developed a systematic approach to understand why errors
creep into any operation. Further, he devised a method to completely eliminate such errors over a
period of time. In this manner, Shingo provided an operational angle to achieving zero defects, as
proposed by Crosby. This method is called poka yoke or mistake-proofing. The idea is to handle
errors as they occur. Since identifying the errors is vital to the success of poka yoke, it must be
carried out rigorously and thoroughly. Once an error (or a defect) is found, a detailed analysis is
done to understand what caused the error. Based on the analysis, alternative methods are
identified to eliminate the cause once and for all.
Shigeo Shingo proposed a method called poka yoke or mistake-proofing for eliminating defects once and for all.
Genichi Taguchi
Taguchi’s focus was on achieving quality by reducing variations in processes. His approach is
based on the basic premise that variations in processes eventually manifest as variations in the
quality performance of the product or service. Therefore he advocated the need to keep the
variations in the processes under control. A different method of measuring quality, the loss
function, is central to Taguchi’s approach. The loss function measures quality. A schematic
representation of the loss function is given in Figure 12.2.
Taguchi proposed a three-stage approach to the design process: systems design, parameter design, and
tolerance design.
The quality gurus provided alternative approaches to build quality into the products and services
that a company offers. The critical differences in their approaches stem from their definition of
quality. For instance, Deming’s emphasis on process improvements and use of statistical process
control charts (SPC) clearly point to a definition of quality that is linked to specifications.4 On
the other hand, Taguchi’s loss function or Juran’s notion of “fitness for use” extends the
definition of quality to a larger domain. Therefore, it is important to understand the various
definitions of quality and the implications of these alternative definitions.
Conformance to Specifications
The easiest definition of quality is conformance to specifications. This definition focuses on the
manufacturing process and motivates the quality manager to develop mechanisms for ensuring
that specifications are met with. Every component of a manufactured product has specifications.
As long as the specifications are met, it is assumed that quality objectives are met with. Such a
definition is very crucial for a manufacturing firm as it enables the management to develop
unambiguous instructions for the worker on the shop floor to maintain quality. Just as we have
specifications for products, specifications for services can also be developed. For instance, in a
hotel, one could set a specification that check-in of arriving guests should be done in 3 minutes ±
30 seconds. Or, in a bank, the time for servicing a customer at the teller’s counter could be
specified and the process monitored and controlled to meet this specification.
An alternative method to define quality is to understand the various dimensions of quality. Garvin proposed that
quality has eight dimensions.
Irrespective of the nature of definitions, there are some important implications for quality
managers:
Quality is both qualitative and quantitative: Attributes of good quality are 100 per cent quantitative on the shop floor. A
process engineer cannot ensure quality if the specifications of the product and the processes are ambiguous. On the
contrary, on the ultimate customer side, attributes of good quality are 100 per cent qualitative. Customers usually
recognize good quality, not by a set of attributes, but by experience and intangible issues. Thus, a customer using similar
products from two competitors will find it much easier to rate one to be of better quality than the other. However, if you
ask the customer to articulate clearly the attributes that made him/her conclude why one product was better than the
other, he/she may have difficulty. Therefore, a good quality management system should enable the manager to
understand the qualitative attributes that influence a customer, and have a method of translating these into unambiguous
quantifiable parameters for design and manufacturing. The critical success factor lies in handling this challenge.
Changing definitions: The definition of quality appears to be constantly changing over time. While it reflects the need to
keep up with changing ground realities, the implication for organizations is to constantly invest in new capabilities and
skills.
Quality is a moving target: The operational aspects of the zero-defects measure of quality clearly demonstrate that
quality is a moving target. For instance in their journey towards zero-defect status, organizations may change their
measure of quality initially from percentage of defects to parts per million, and later to parts per billion. In such a
scenario, how are organizations going to keep up their employees’ enthusiasm in meeting the targeted quality standards?
The implications for organizations are that there are mindset issues to be addressed in achieving a good quality
management programme.
The culmination of the discussions in the previous sections is the notion of total quality
management (TQM). Modern quality management systems are built on the concept of TQM.
Total quality management can be defined as an organization-wide effort to develop the
systems, tools, techniques, skills, and mindset required to establish a quality assurance system
that is responsive to emerging market needs. Such an effort will provide distinctive advantages to
a firm and will enable it to compete effectively in the market. A good TQM programme consists
of the following important elements:
a. Role of the top management
b. Employee involvement for continuous improvement
c. Addressing the training requirements of employees
d. Tools and techniques for quality assurance and continuous improvement
Total quality management (TQM) can be defined as an organization-wide effort to develop the systems, tools,
techniques, skills, and mindset required to establish a quality assurance system that is responsive to emerging
market needs.
The term “total” in total quality management connotes “everyone”, “everywhere”, and “every time”.
Similarly, in the past, middle managers and top managers played the role of controllers. They
directed improvement efforts and established tight controls and supervision to ensure that quality
goals were met. On the contrary, in TQM, the focus shifts to one of facilitating the production
workers in their efforts towards producing quality goods and services. This calls for a certain
shifting of process ownership to the workers and their supervisors. The success of the TQM
programme depends on how well this transition is made.
The notion of “every time” puts a different pressure on the traditional methods of operations.
There is a greater degree of intangibility and uncertainty with respect to outcomes in the long
term. On the other hand, short-term goals and benefits are like low-hanging fruits. They are not
only easily reachable but also tangible. Managers are averse to working for long-term benefits
and tend to compromise on them in order to post impressive performances in meeting short-term
goals. If, for some reason, there is a quality problem in the production line, the manufacturing
managers will choose to ignore it and continue to produce. When quality control rejects the
components, they could be reworked later. Therefore, the top management has to play an
important role in ensuring that “every time” is followed in letter and spirit. This may call for
leadership and guidance in minimizing conflicts between short-term and the long-term goals.
Nothing is taken seriously when employees in organizations come to know that the changes do
not have the top management’s involvement and commitment. This is a significant aspect of the
management of change. The top management also needs to play a vital role in communication.
They should not miss even a single opportunity that comes their way to demonstrate their
support and involvement in the TQM implementation process. Moreover, issues pertaining to
“everyone” and “every time” require clear directions and a leadership role from the CEO of an
organization. Only then can the TQM process get a fair chance at success.
GE Healthcare
Quality and continuous improvement are part of our culture, it is embedded in our processes,
our recruitment, and our training. We take responsibility for the projects we execute, for
maintaining the operation of the products we deliver, and for the traceability required by the
authorities.
Hero Honda
Excellence in quality is the core value of Hero Honda’s philosophy. We are committed at all
levels to achieve high quality in whatever we do, particularly in our products and services
that will meet and exceed customer’s growing aspirations through:
Innovation in products, processes, and services.
Continuous improvement in our total quality management systems.
Teamwork and responsibility.
Sasken
Achieve excellence through commitment and innovation surpassing the expectations of our
customers. In this endeavor, we shall
Set up and implement world-class systems and processes
Sensitize and raise our quality consciousness
Continuously strive for improvements to make quality a way of life
Meet all applicable environmental, health, and safety standards and demonstrate corporate citizenship on an
ongoing basis
A well-defined quality policy serves to signal the seriousness of an organization and the nature
of its commitment to various stakeholders in the system. Therefore, the first step in launching a
TQM programme pertains to stating a quality policy. Many Indian organizations have recently
set forth their quality policies and have communicated it to all their stake-holders. It is not
surprising that the top management takes a major responsibility in this process and thereby steers
the TQM programme. They help establish an unambiguous quality policy and communicate it to
all employees in the organization.
Employee Involvement
Shifting the responsibility of quality management to the workers must be accompanied by the
necessary organizational changes enabling the employee to take on the responsibility. The
important change is process ownership by the employee. Process ownership is the
transformation of the role of employees from the function of merely accomplishing the assigned
tasks related to a process to one of thinking through the various aspects of the process and taking
on a conscious role in the overall management of the process.
Process ownership is the transformation of the role of the employees from a mere function of accomplishing the
assigned tasks to one of taking a conscious role in the overall management of the process.
Consider two alternative roles for employees. In the first case, the employee will be merely
focusing on producing a component using a machine. Maximizing production (and thereby the
productivity bonus) may be of interest to the employee. The employee may not be motivated to
maintain cleanliness in the working area or ensure good maintenance of the machine and other
resources. Further piling up of inventory and problems related to the quality of the component
are not his concern. There may be separate specialists to attend these problems.
In the second case, in addition to the production, the employee will have the ultimate
responsibility of maintaining all the aspects of the processss. He will be responsible and
accountable for all local issues pertaining to the process. This manner of relating to the entire
process builds a sense of ownership in the employee and makes him/her get involved more
intimately in the process. Employees eventually tend to take a greater interest in the overall
improvement of the process, leading to better quality of the product. Therefore, employee
involvement is an important aspect of TQM.
Employee involvement invariably requires employee empowerment. Employees must be
allowed to take decisions pertaining to stopping the line when quality problems surface and they
need mechanisms to work in small groups to make continuous improvements in the process.
Each employee must also be able to improve the housekeeping of the resources in his/her work
area and improve the upkeep and maintenance of machines and other resources. These require
several organizational and procedural changes in the organization.
The other aspect of empowerment pertains to equipping employees with the required skills
and tools for managing quality. Figure 12.3 shows the normal steps involved in quality
improvement through a series of projects done in small groups. In order to be problem solvers,
employees need problem-solving tools. They need to be trained in the required quality control
tools. Small group improvement projects are the basic building blocks of a TQM programme for
improving the quality over a period of time. This calls for a set of skills oriented towards
working in a team; in other words, interpersonal skills to work with fellow employees.
Moreover, employees need better communication skills to deal with one another in a group and
with their subordinates and supervisors. Therefore, addressing these training needs is an
important component of the TQM implementation process.
FIGURE 12.3 Quality improvement through small group improvement projects
Addressing Training Requirements
Successful TQM programmes demand substantial investment in communication. The top
management is first educated regarding the need, the required changes and the potential benefits.
The process then continues, until the lowest level in the organizational hierarchy is reached. The
top management should be aware of the impending changes and the benefits. Only out of such an
understanding can there be involvement and support in resolving conflicts that arise during the
implementation stage. Middle-level managers often form part of the team of change agents.
Moreover, changes may result in some of them giving up their traditional role of control over
their subordinates. They need adequate exposure and clarity in the required changes, failing
which they become insecure and try to stonewall the efforts. The workforce, the union, and the
supervisors as well as the staff and managers belonging to the manufacturing-support and non-
manufacturing areas should be appraised of the nature of changes and the potential benefits.
Communicating the need for alternative procedures, new methods of empowerment, and the
benefits that are likely to accrue to the organization is just one aspect of training. The other
aspect relates to educating all the employees in using the new quality control tools and
techniques to be adopted for quality management. At the operational level, several new quality
control tools are available. Employees need to be trained in using these tools. Supervisors require
training in problem-solving approaches and working in small group improvement teams.
Supervisors and the middle managers require several new skills for setting up a good quality
management system at work. These include the following:
Identifying problem areas
Setting targets for improvement
Participating in the improvement process
Facilitating the process owners and monitoring and guiding them over time
The greatest strength of a quality management system such as TQM is the emphasis laid on using
simple yet powerful tools for tracking various aspects related to quality management. Over the
years, several QC tools have been developed. These tools are widely applied in practice too. The
QC tools fall under five broad categories. Table 12.1 has a description of these categories and the
tools available in each category.
Let us take a look at the categories of tools discussed in the table:
Tools for highlighting quality problems: These tools bring impending quality problems in a process to the attention of
the operational personnel and alert him/her to the issue. The most commonly used tool is the process control chart. By
carefully selecting a set of important process parameters, it is possible to set up a continuous monitoring system using
statistical principles to identify when the process is likely to go out of control. An out-of-control situation points to
deteriorating quality in the process. It triggers closer observation and further analysis to nip the problem in the bud. A
detailed description of process control charts is available in Chapter 18.
Tools for identifying specific improvement opportunities: Once an impending quality problem is brought to the attention
of the concerned supervisor, manager and the employees involved in the work area, additional tools are required to zero
in on the exact problem. This could be done by collecting some additional data, systematically plotting the data and
analysing it. Several tools are available to perform this task. The most popular among them include check sheets,
histograms, and pareto diagrams. Using these tools, it is possible to locate exactly where the problem lies.
Tools for analysing problems and their root causes: The next step in quality assurance is to conduct a systematic
analysis of the problems and their root causes. Only when the root causes are known can solutions be found for the
elimination of the problem. The most popular tool used in this category is the cause and effect diagram (also known as
fishbone diagram) and a variation of it known as the cause and effect diagram with action card (CEDAC).
Quality management tools available tools fall under five broad categories:
Highlighting problems
Control charts
Identifying specific improvement
opportunities Histograms
Check sheets
Pareto diagrams
Scatter diagrams
Graphs
These categories are tools for operational control. Once a quality problem manifests, these tools
help in recognizing it and identifying the root causes. Once the causes and remedies are
identified, it is possible to rectify the problem. However, a good quality management system also
requires tools for planning purposes. Some of the improvement methods discovered while using
the quality control tools can be permanently incorporated into operations in the future. Therefore,
the use of quality tools that help prior planning minimizes the occurrence of quality problems
during operations. Two categories of planning tools are advocated in TQM.
Tools for operational planning: Operational planning tools pertain to prior planning done to minimize quality problems
in operations. For instance, if some mistake-proof methods are incorporated in the system, it results in complete
elimination of the problem.
Tools for strategic planning: The strategic tools help an organization link the quality management initiatives to the
requirements in the market place and provide overall direction for operational planning and improvement activities. Two
popular tools in this category include the quality function deployment (QFD) and quality costing.
Histograms7
The histogram is a simple method of graphically representing the frequency distribution of
multiple attributes of interest. Based on a simple count of the number of occurrences of each
attribute, a histogram can be constructed. In a manufacturer of earth moving equipment a study
revealed that poor quality resulted in various snags in the assembly plant.8 Due to a large number
of adjustment snags, the throughput time was much higher. The data and the accompanying
graph in Figure 12.4 shows the various causes for adjustment snags.
A histogram is a simple method of graphically representing the frequency distribution of multiple attributes of
interest.
The value of a histogram lies in its ability to graphically portray the various causes pertaining
to the problem as well as the magnitude of these causes. Clearly, employees would like to focus
on major problem areas and initiate further data collection to understand why problems occur in
those areas. For instance, in this example, problems due to reworking are very significant.
Therefore, employees would like to know more about the causes of such problems.
Pareto Diagrams
The process of data collection helps to understand where the problem lies and what the nature of
the problem is. Using this information, it is easy to identify which aspects need improvement.
The difficulty, however, is deciding where to start the improvement process. A pareto diagram
is a method by which clear-cut priorities are established for directing the improvement efforts.
Moreover, it helps organize the basic data in a systematic manner such that the improvement
team detects the significant aspects of improvement and initiates necessary corrective actions.
A pareto diagram is a method by which clear-cut priorities are established for directing the improvement
efforts.
Let us look at the histogram shown in Figure 12.4 in an alternative fashion by plotting the data
in decreasing order of occurrence. The revised histogram is presented in Figure 12.5. The revised
representation of the histograms clearly establishes priorities for improvement. In the case of
adjustment, snags, reworks, leakages and missing components are important elements that
require immediate attention.
The cause and effect diagram, or the fishbone diagram, is a generic methodology developed to trace the
problems to their root causes.
CEDAC
A variation of the cause and effect diagram is CEDAC.9 In this, the “AC” stands for “action
cards”. In the cause and effect diagram, there is a fishbone structure and the members of the
small group improvement activity need to discuss on the problem, causes and solutions. On the
other hand, in the case of CEDAC, this entire process is left open to all employees of the
organization by adding two sets of cards to the fishbone. One set of cards is known as the
problem cards and another set is known as the solution cards. By placing a visual board with the
fishbone structure and problem and solution cards in two separate bins alongside the board,
CEDAC enables any employee to contribute to problem solving. Normally, the problem and
solution cards are colour-coded for easy identification. Figure 12.7 shows one such CEDAC
system that was implemented as part of a study in an automobile braking system manufacturing
plant in India.
A variation of the cause and effect diagram is CEDAC. In this, “AC” stands for “action cards”.
FIGURE 12.6 A generic representation of a cause-and-effect diagram
The implementation works as follows. Any employee who thinks that he/she knows what the
problem or solution to the problem could be will pick up an appropriate card. He/she will then
write his/her suggestions and affix it at the appropriate leg of the fishbone. Figure 12.7 is a
snapshot of the status of the board after a week of its installation. Even in such a short period of
time, nearly a dozen problem cards and another dozen solution cards have already been affixed
to the board. It is clear from this example that the CEDAC system establishes an atmosphere of
continuous improvement. It enables all the employees in an organization to make full use of their
accumulated knowledge and experience. Furthermore, it helps change the organizational culture
by empowering people to fully participate in the continuous improvement process.
Poka Yoke
Poka yoke is a Japanese term for “mistake-proofing of operations”. Shingo proposed this method
while he improved the Toyota Production System. The basic principle behind this method is that
several defects that creep into an operation are usually avoidable. Errors and defects have a cause
and effect relationship. If inadvertent errors in any process are not recognized and methods are
not established to eliminate them, they will eventually manifest as defects. Only by a careful
scrutiny of the process is it possible to identify the root causes of the defects. The root causes
could then be completely eliminated by redesigning the operations and incorporating methods by
which the process will not allow the errors to happen in the future.
Poka yoke is best explained with an actual example. Consider a component manufactured by
an automobile component manufacturer that requires a through hole to be drilled using a hand-
operated drilling machine.10 If the worker operated the hand drill imperfectly, it is likely that
he/she would not have traversed the drilling tool till the very end. Consequently, a through hole
is not drilled. It may appear as though such errors are unlikely to creep into a manufacturing
system. However, in reality, when a large number of pieces are produced, workers tend to
commit such mistakes out of sheer mental and physical fatigue. In the auto-component
manufacturing firm, the defect was found at a much later stage, during assembly.
When the worker at the assembly section wanted to insert another component he found that
there was no hole (since the component is not drilled right through). Discovering this at this
stage meant tearing down the assembly to remove the component and replacing it with a good-
quality component. This resulted in considerable loss in productivity and added to the cost. One
way to avoid this simple quality problem is to undertake mistake-proofing in the drilling process
so that imperfect drilling of a component is completely avoided. Figure 12.8 shows the new
arrangement.
FIGURE 12.8 Defect prevention through operations planning
ideas at Work 12.3
In the new arrangement, there is a pair of sensors controlling the hand drilling process. While
the sensor at the top activates the forward movement of the handle, the sensor at the bottom
activates the return path for the handle. Once the worker begins the forward movement of the
spindle, there is no way he can retract the spindle, except by the activation of the lower sensor. In
this process, it is ensured that a through hole is drilled. By incorporating these features in the
process through prior planning, quality is built into the product.
Based on a set of structured information gathering, QFD helps an organization to capture the qualitative
attributes of the quality as perceived by the customers and translate them into meaningful quantitative measures.
QFD utilizes a set of four “houses of quality”. Each house of quality captures certain
information and uses a structured methodology to translate customer needs to actionable ideas
and quantifiable attributes pertaining to the product/service. Figure 12.10 is a broad
representation of the four houses. As is clear from the figure, during each stage of the process,
specific details are culled out of the basic understanding of customer needs. In the first stage,
design attributes are arrived at. During the second stage, the design attributes are linked to
actionable items. The actions identified in the second stage are the basis for the third stage in
aiming at specific decisions to be implemented. Finally, in the last stage, specific details
pertaining to each process parameter are arrived at. The four houses are sequentially linked to
one another. Therefore, the outcome of each preceding stage becomes the input for the next
stage. Each stage is known as a “house of quality” simply because it is constructed in the shape
of a house.13
Quality Costing
A useful tool for quality management is quality costing. Quality costing is a technique by which
all the costs related to the delivering of a certain level of quality are captured and appropriately
classified for the purpose of evaluating the quality management efforts in an organization. By
translating the quality performance into monetary terms, quality costing captures the attention of
the top management. Furthermore, it enables firms to prioritize various spending opportunities in
the area of quality improvement. Juran advocated the notion of quality costing and Crosby
encouraged firms to put a good quality costing system in place. According to Crosby, firms that
begin measuring quality costs will first realize that nearly 25 per cent of their sales turnover is
spent on quality management.
Quality costing is a technique by which all the costs related to the delivering of quality are captured and
classified for evaluation of the quality management efforts in an organization.
Quality costs in any organization fall under four categories (Figure 12.11). Prevention costs
include all costs incurred to prevent, reduce, or eliminate the occurrence of defects. Expenses
incurred towards training of employees in better methods, vendor development and small group
improvement projects are some of the examples in the category. On the other hand, appraisal
costs pertain to all expenses incurred in ascertaining the quality of a product, component or a
service. The third category of costs relates to costs of internal failures. An appraisal activity may
reveal defects and may call for scrapping components altogether or necessitate rework in several
situations. This results in lost productivity and added costs for rectification. All these costs are
classified as internal failure costs. The final category is the external failure costs. All costs
related to rectifying the defects after the product is shipped out of a manufacturer’s premises will
be classified as external failure costs. The costs include warranty, cost of product returns and
replacement, on-field maintenance, and any compensation for the damages incurred. It also
includes the cost of lost customer goodwill.
FIGURE 12.10 The four houses of quality
Quality Awards
Different countries have instituted quality awards to motivate business organizations towards
achieving excellent performance. Table 12.2 has a list of awards and awarding agencies. The
Deming Prize was one of the earliest awards instituted for excellence in quality management.
Although only a few companies outside Japan have bagged the award, a majority of them are
from India (Table 12.3). Although these awards differ from one another in the manner in which
winners are selected, all of them have a system of identifying key components and a scoring
system to rate the performance of each applicant against each parameter. Visits to applicant
organizations and scrutinizing an extensive number of documents submitted in support of each
component of the award form part of the process of selecting winners.
Sixteen companies belonging to the automobile and other sectors such as steel have so far won the Deming
application prize. A few Indian companies have won the Japan quality medal as well.
There two quality awards in India: the Rajiv Gandhi National Quality Award given by the
Bureau of Indian Standards, and an award for business excellence instituted by the Confederation
of Indian Industries along with the Export-Import (EXIM) Bank of India. Figure 12.12 shows the
model utilized for the CII–EXIM Business Excellence Award. The model closely follows the one
proposed by the European Foundation for Quality Management (EFQM). Out of a total 1000
points, 500 points each are awarded for enablers and results.
In India, two quality awards are given; the Rajiv Gandhi National Quality award and the CII-EXIM Business
Excellence Award.
Quality Certifications
Every country has a national standards body that is mandated by the government to establish
standards and certify quality. The Bureau of Indian Standards is the Indian body for quality
certification. However, over the years there has been a considerable convergence of alternative
standards towards the International Organization for Standardization (ISO). ISO 9000 is a
generic set of standards that enables organizations to build quality management and assurance
systems. Individual sectors of industry have added more components to the basic ISO 9000
system to obtain their own standards for quality management. For example, automobile majors
developed QS 9000. Similarly, the players in the telecommunications industry developed TL
9000. Growing concerns about environmental issues and occupational safety have necessitated
the development of newer standards such as ISO 14000 and OHSAS 18000.
ISO 9000 is a generic set of standards that enable organizations to build a quality management and assurance
systems.
Source: B. Mahadevan, “Journey to the Deming: Implications for Corporate Transformation and Competitiveness—In Conversation with Arun Bharat
Ram,” IIMB Management Review, Vol. 17, no. 2 (2005): 59–70.
The software sector has adopted a unique quality certification programme developed by the
Software Engineering Institute (SEI) at the Carnegie Mellon University, known as the
Capability Maturity Model (CMM). CMM is a model used by many organizations to identify
the best practices useful in helping them increase the maturity of their processes. In 2000, CMM
was upgraded to Capability Maturity Model Integration (CMMI). According to SEI, the goals in
developing CMMs include addressing software engineering and other disciplines that have an
effect on software development and maintenance and providing integrated process improvement
reference models. Moreover, the models will also enable efficient improvement across
disciplines relevant to software development and maintenance. Currently, six variations of CMM
are available. These include the following:
Capability Maturity Model Integration (CMMI)
People Capability Maturity Model (P-CMM)
Software Acquisition Capability Maturity Model (SA-CMM)
Capability Maturity Model for Software (SW-CMM)
Systems Engineering Capability Maturity Model (SE-CMM)
Integrated Product Development Capability Maturity Model (IPD-CMM)
The software sector has adopted a unique quality certification programme known as the Capability Maturity
Model (CMM), used by organizations to identify the best practices useful in helping them increase the maturity
of their processes.
A quality assurance system is one that has all the components that promote total quality
management in an organization. Based on the discussions in this chapter, we can clearly identify
the key elements of a robust quality assurance system. Figure 12.14 is a graphical representation
of such a system, the components of which are briefly here:
Top management commitment: One of the important lessons of TQM is that the fundamental building block of a quality
assurance system is the top management’s commitment to quality.
Employee involvement: The other pre-requisite for a robust quality assurance system is to develop a good atmosphere
for employee involvement. Appropriate choice of quality management tools could further promote group working and
inculcate a sense of ownership of the process.
Understanding customer wants and translating them into meaningful measures: Without a good understanding of what
the customer wants and what the market values, it is futile to establish a good quality management system. It is equally
important that we have methods by which this is translated into meaningful measures at the workplace. Without this, all
efforts directed towards quality will fetch products and services that customers are not interested in. Therefore, an
important element of a good quality assurance system is to resolve this issue. Tools such as QFD are very useful for this
purpose.
Identifying quality problems: A good quality assurance system requires a methodology by which quality problems are
brought to the attention of the employees and managers as quickly as possible. Control charts and other such visual
control techniques enable an organization to achieve this requirement of a quality assurance system.
Providing tools and techniques for problem solving: Lasting solutions to quality problems are found only when the
problems are traced to their root causes and corrective measures are initiated. In the absence of this, organizations
constantly face quality problems and incur high costs in rectifying them. Therefore, a quality assurance system will
provide alternative tools and techniques, such as the fishbone diagram and CEDAC, to track the problems to their root
causes. It will also promote small group improvement projects for continuous improvement.
Mistake-proofing of operations: Higher levels of performance in quality management are attained when an organization
rises to the level of avoiding quality problems by careful redesign of its operations and standard operating procedures.
Tools such as poka yoke are valuable for such initiatives.
Documentation for continuous learning and improvement: Organizations tend to benefit greatly from learning curve
effects. Revisions and changes in operating procedures that lead to better practices need to be properly documented.
Such practices build a strong knowledge base of quality management and help face a variety of situations in the future.
Quality certification systems such as ISO 9000 equip an organization to perform this task very well.
Benchmarking: The final aspect of a good quality assurance system lies in its ability to constantly point to new areas of
improvement that an organization needs to make to remain competitive in the market. In addition to a firm-initiated
internal benchmarking exercise, rating by an independent third party helps this process a great deal. Therefore, every
organization should endeavour to work towards obtaining quality awards and certifications. Therein lies the ultimate test
of the strength of its quality assurance system.
The resounding success of Japanese manufacturing firms has invariably been linked to excellent practices in quality
management in the last 30 years.
Much of the progress that organizations made in quality and productivity management is attributed to the teachings
of quality gurus such as Deming, Juran, Crosby, Taguchi, Shingo and Ishikawa.
A good quality management system must enable a manager to understand the qualitative attributes that influence a
customer and have a method of translating these into unambiguous quantifiable parameters for design and
manufacturing.
A total quality management programme consists of four components: top management commitment, employee
involvement in continuous improvement initiatives, training and skill development, and investment in robust
systems.
Several tools and techniques are available that help in highlighting problems, identifying specific improvement
opportunities, analysing problems and their root causes, and operational and strategic planning for building quality
into products and services.
Alternative certification procedures and award mechanisms are available to recognize excellent quality
management systems in organizations.
REVIEW QUESTIONS
1. What is the relevance of quality in today’s competitive scenario? Was it any different during the 1970s and the 1980s?
2. Enumerate the key benefits that an organization derives by improving the quality of its products and services.
3. Compare the teachings of Deming and Juran and make a critical report of your key inferences from them. Were they
advocating the same idea?
4. Identify three ways by which an organization can benefit from the teachings of the following gurus:
a. Philip Crosby
b. Karou Ishikawa
c. Genechi Taguchi
5. For each of the products/services mentioned here, define quality from the customers’ perspective:
a. A laptop computer
b. A weekend holiday in a beach resort in Goa
c. A motorbike
d. Legal services obtained from a lawyer
e. A day-care centre (crèche)
f. An electric bulb
6. For each of the products/services in Review Question 5, define quality from the perspective of the firm providing the
products/services. Are these definitions the same or different? Explain your observations.
7. What are the challenges that an organization faces when it attempts to define quality? Is it possible to overcome these and
if so how?
8. What is poka yoke? Can you identify one situation each in your classroom, hostel, and college administrative office that
can potentially benefit from the application of this concept?
9. What do you mean by the term total quality management? Is there anything special about “total” in TQM?
10. If an organization practices employee involvement, how is it likely to be benefited?
11. How can the top management of an organization influence the quality management process?
12. What is the use of the QC tools in quality management?
13. What are the elements of a quality costing system? How does an organization benefit from a quality costing system?
14. In an organization if the prevention and appraisal costs are very low, what, according to you, will the magnitude of the
other components of the quality costs be? Explain your answer with supporting arguments.
15. What is a quality house? What is the use of a quality house?
16. An organization does not know if it needs to go for ISO 9000 certification or ISO 14000 certification. They are also not
aware why they should go for a certification. Prepare a one-page report to resolve their confusion.
17. How does an organization benefit by working towards obtaining a quality award? Explain with reference to the CII-
EXIM Business Excellence Award.
PROBLEMS
1. A hospital buys medical consumables from several suppliers. The bad quality of consumables sometimes affects the
hospital operations critically. Therefore, the hospital follows a quality control routine. Based on a weekly analysis of the
quality of cotton and dressing material used in large quantities in the surgery section of the hospital, the following causes
of defects were found out (see Table 12.4).
2. A bank operating in a prominent residential area in the metropolitan city of New Delhi is facing tough competition from
several of its competitors. Improving the quality of services has become a necessary condition for the branch to survive.
One of the major problems is the excessive delay that customers experience in the bank. After a brainstorming exercise, a
group of bank employees came up with the possible causes for poor quality of the service. The following is the list:
Represent the problem using a fishbone diagram. What future course of action should the group of employees take with
respect to the problem?
3. The quality of the food services in a students’ hostel mess is a cause of great concern among the students. Use the
fishbone diagram and show how one should go about studying this problem.
4. A manufacturing company is keen to understand the notion of quality costing and its relevance to quality management.
Based on one month’s work by a task force, the following data have been collected from the accounts of the company,
pertaining to the last three years. Table 12.5 has various items of cost (in thousand ₹):
NET-WISE EXERCISES
1. To get an overview of the Malcolm Baldrige Quality Award program, visit the URL: http://www.nist.gov/baldrige. Click
on ‘About Us’ and read about the history of Malcom Baldrige National Quality Awards for background information about
the award. Click on the ‘FAQs’. There are several side bars in ‘FAQs’ menu. Go through them to know more about the
award.
Next, visit the URL
http://cii-iq.in/cii/sectors/BE/CII-Exim%20Bank%20Award%20for%20Excellence.html. There are details on various
aspects of the award in several links in this page. Go through these links.
a. What do you think is the basic purpose of the Malcolm Baldrige Award and the CII– EXIM Bank Award? How do
they achieve this objective?
b. Compare and contrast the two awards. Define three criteria for comparison and use them for answering this
question.
MINI PROJECTS
1. Form a small group improvement team of six students. Visit the local branch of the bank or a hospital and get their
approval to study the quality problems that they face for possible improvements and prepare a final report based on the
entire exercise. The project should involve the following important steps:
i. Develop an appropriate definition for quality pertaining to the problem taken up for the study in consultation with
the employees in the organization.
ii. Identify the metrics that needs improvement and set realistic targets for the metrics.
iii. Collect data and use the available QC tools to analyse the problems, their root causes, and potential improvements.
a. Estimate the projected improvements and the changes required in the system to improve the quality
b. Discuss your results with the organization and understand the feasibility of your recommendations. Make
further refinements of the proposed recommendations
c. Write a detailed report based on your experiences in the project
2. Consider the following three quality awards; the CIIEXIM Business Excellence Award, the Deming Award, and Malcom
Baldrige Award. Make a detailed study of these awards by visiting their respective websites and collecting information
from secondary data such as published literature and reports. Based on this study, prepare a report to answer the following
questions:
a. What are the salient elements of each of these awards? Is there an underlying philosophy that governs selection of
successful companies in each of these awards?
b. Select one award-winning firm for each of these categories of awards and study them. Briefly document the
experience of one firm each that has bagged these awards.
c. What is the nature of the efforts required from a firm to get these awards? Is there a time frame involved? How do
firms that qualify for these awards benefit from the exercise?
d. What is the status of firms that obtain these awards? How do they perform after they obtain these awards? Are
there any significant disadvantages or problems these firms go through in the process of obtaining the award?
3. Quality certification is an important aspect in quality management. Prepare a detailed report on the basis of your own
study to answer the following questions:
a. What are the emerging standards in the area of quality management? Why are they important?
b. What are the industry specific initiatives in the area of quality certification? How are they different from generic
standards?
c. What benefits accrue to a firm that obtains a quality certification? Are there any prior experiences that confirm
these findings?
CASE STUDY
Operating Unit
BGEL has both manual and mechanized operations for shirt-making. The current capacity is
for a daily production of 6500 shirts. While cutting and basic stitching are done using high-
speed machines, there are portions of shirt-making that involves skilled labour in addition to
the machines. In all, there are 65 machines and of there are 12 cutting machines, the rest
being stitching machines. BGEL employs over 400 people; of these, 290 are engaged in
production. The production workers are functionally organized into four sections; laying and
cutting of cloth, body stitching, making slacks and attaching it to the body of the shirt, and
finally, attaching the collar to the shirt. After this, it goes to the QC department for final
checking and QC clearance. After the QC clearance, it is sent for packaging and dispatch.
The production flows in this sequence, and each section employs a number of production
workers in order to meet the daily demand.
There are separate sections for routine, preventive, and breakdown maintenance of the
machines; quality control; and production planning. All these sections, as well as the
production sections, have supervisors who are responsible for output and performance of the
units. At the beginning of each shift, the production supervisor in the section assigns jobs to
each worker based on the daily requirements. The production workers receive incentive pay
on a “piece rate” basis, once they exceed the normal production in a shift.
SUGGESTED READINGS
Lean Management
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
A lean management system often involves major changes in the manufacturing architecture. It also involves other activities
such as standardizing operations, quick changeovers, levelling production and use of better scheduling and planning methods.
Source: Milltec Machinery Pvt. Ltd.
People Waste
People waste is the result of not utilizing workers’ skills to their full potential due to little or
no training, office politics, low-pay, or high turnover. People waste also happens on account
of absence of goal setting, employees working at cross-purposes and people spending a lot of
time correcting problems (re-work). Moreover, not assigning work properly or assigning
unnecessary work is a waste of people resources.
Process Waste
Process waste usually occurs on account of neglect and bad design of the process. Moreover,
the attitude of the people leading to mind-set inertia (we have always done it that way) also
significantly contributes to process waste. There are several instances of process waste in
service organizations. Some of them include:
a. Overproduction: Doing work before it is required
b. Over processing: Working on one item more than what is necessary
c. Control: Doing more monitoring than what is required
d. Too many rules leading to very complicated work and several errors that require re-work later
e. Non-adherence to standardization: Workers ignoring best practices
f. Sub-optimization: Duplicate or competing processes
g. Scheduling: Poor organization or planning
h. Uneven flow: Backlog of information and materials
i. Checking: Over-inspection and re-work
j. Error: Re-work to correct mistakes
Information Waste
Information waste leads to loss of value because of poor, outdated, duplicated, or incomplete
data. This causes costly mistakes because the corrections (re-work) take a lot of time and
effort. Some examples of information waste are:
a. Translation: exchanging bad data and formats between procedure steps and owners
b. Hand-off: transferring data/materials within a group that is not part of the procedure chain
c. Missing information leading to committing additional resources to correct key data or information
d. Irrelevant information: unnecessary data that creates confusion, slows processing, or causes errors
e. Inaccurate information: wrong data or information generating additional non-value added work to rectify the
mistakes
Asset Waste
Asset waste is not using material and/or property efficiently. This includes buildings, office
equipment, supplies, products, and services. Examples of asset waste include:
a. Excess inventory, time, supplies, and piles of unused or outdated materials
b. Work in progress (WIP) that cannot be completed because of inefficient workflow
c. Resources committed to rectify defects or mistakes created by other categories of wastes
d. Moving product/material around inefficiently; transportation of information and materials; late deliveries to
customers; concentration of consumables
Creating lean systems in any organization calls for identifying various waste generating
incidences under each category of waste. Furthermore, the waste must be eliminated
progressively using some tools and techniques. For example, some of the waste could be
eliminated by re-designing the structural arrangement of resources and some others by
implementing simpler methods for planning and operational control. We shall see these
aspects in some detail in this chapter.
Source: Based on http://www.leanexpertise.com/MTL/LeanOffice/LeanOfficeBasics.htm last accessed on 27 June 2014.
The process of manufacturing has been significantly influenced by the turn of events in history.
In the beginning of the twentieth century, Henry Ford pioneered the mass production system as a
way for manufacturing organizations to organize, plan, control, and evaluate their operations.
Several operations management tools developed during that time sought to promote the mass
production philosophy. During World War II, flexibility was a key requirement and this altered
the manner in which operations management theory and practice developed.
During the oil crisis, recession put new pressures on the manufacturing system and demanded
better methods of managing operations in manufacturing organizations. Customers demanded
better products at lower prices. They increasingly demanded more options, along with a
commitment to deliver products and services faster. In response to these changing requirements,
new principles of managing operations were required. Managing the operations efficiently,
developing alternative methods for quality management, and creating responsive organizational
structures were some of the newer requirements. Notably, Japanese manufacturers developed a
set of tools and techniques over a period of two decades that addressed many of these
requirements. These are collectively known as lean management practices.
Using the new capabilities, Japanese manufacturers threatened to alter the forces of
competition and sought to provide new value to customers. In the automobile sector, passenger-
car manufacturers such as Toyota and motorcycle manufacturers such as Kawasaki competed
effectively with established giants in the United States, such as General Motors (GM) and Harley
Davidson, respectively. While Japanese manufacturers could offer products that came at a low
cost along with high quality, American manufacturers were offering exactly the reverse, that is,
low-quality products at a high cost. The initial success of the Japanese manufacturers in the
automobile sector was soon to be replicated in other sectors of the industry. Notable among them
include electronic components such as resistors, memory chips, and transistors, and
entertainment electronics products such as cameras and musical systems.
The world was in the midst of a severe global economic slowdown after 2007, and all major
economies underwent a slowdown, shrinkage, or even severe depression during this period. If we
carefully analyse the trends during the five years preceding this recession, we will find that there
were enough signals. Organizations in some sectors of the industry made concerted efforts to
respond to the emerging scenario. Some of the trends during this period and their impact are
worth mentioning:
a. The “Big Three”, the three large automobile manufacturers in the United States, along with other global auto majors,
faced severe problems due to the economic slowdown. The inventory of vehicles was mounting and the companies had
to cover high structural costs. General Motors had lost USD 82 billion since 2004, its last profitable year. In order to
stay alive, GM announced a USD 10 billion cut in costs in July 2008 and a further USD 5 billion in November 2008.
These included a dramatic cut in its 2009 capital expenditures and additional cuts for its white-collar workforce.
b. Due to the mounting cost of crude and the competition in the civilian aerospace market, Boeing and Airbus began
working with the central theme of enhancing efficiency and cost-cutting. Their new models (the Dreamliner of Boeing
and the A380 of Airbus) emphasized these aspects in product development and production. There was increased
competition from low-cost economies.
c. In the textile sector, Indian companies suffered from longer lead times, lower productivity, and smaller plants compared
to Chinese ones. Therefore, the Indian textile industry was not able to benefit much from the multi-fibre agreement that
promotes free global trade in the garment sector.
In the light of the recent economic slowdown, it has become very important for organizations to
improve productivity and response time and cut costs so that they can stay competitive.
Manufacturing and service organizations are required to first understand how to develop better
operational systems and deliver better quality products and services using fewer resources. The
underlying requirement to be more competitive in the market is to ensure that less is more
productive. Lean management principles address these issues directly and efficiently.
Lean management is a new philosophy of operations management that provides a set of tools and
techniques to compete in an increasingly fierce marketplace. It is an organizational mechanism
for defining value and thereby creating a value stream for customers. Lean management is based
on the premise that by identifying waste in any system and removing it, it is possible to create a
value stream for the products and services that an organization offers to its customers. It enables
organizations to provide better value in their offerings by constantly improving their
operations.A good understanding of waste and value and a set of tools and techniques to
eliminate waste from the system are crucial to creating a lean enterprise. To understand lean
management better, we need to familiarize ourselves with the concepts and definitions of value-
added (VA), non-value-added (NVA), and necessary but non-value-added (NNVA) activities.
These concepts have been discussed in detail in Chapter 8. We should also understand the
concepts of waste, value stream, and lean management. These terms are defined here.
Waste: Any process or set of activities that does not add value as perceived by the customer is classified as waste.1 It
consists of both NVA and NNVA. From a short-term perspective, NVA is avoidable waste whereas NNVA is an
unavoidable waste.
Value stream: All activities that need to be performed (VA and NNVA) from the time the customer order is received to
the time the order is fulfilled make up the value stream for an organization. Depending on the nature of the business, it
may span activities from concept to launch, order to delivery, or raw material to finished product.
Lean management: Lean management is a process by which the continuous efforts of all concerned parties enable an
organization to create a channel for the value stream by eliminating waste from the system.
Any process or set of activities that does not add value as perceived by the customer is classified as waste.
Viewed from this perspective, the entire notion of lean management revolves around the issue of
waste and waste elimination. The starting point for lean management is to develop a good
understanding of waste. Once the notion of waste is understood, it is easy to create a lean
enterprise. Let us hypothetically compare two competitors. Competitor A may have 40 days of
inventory, may employ 150 people in its factory, and may have a layout in which jobs travel 6
km from the raw material stage to finished goods. They may also have an inward goods
inspection department employing three inspectors to inspect quality of the incoming lot of
materials. On the other hand, Competitor B may have 20 days of inventory, employ 120 people
for the same level of production, with jobs travelling less than 3 km on the shop floor, and may
have no inward goods department. It may be receiving direct in-line supply of materials by the
suppliers.
After a quick comparison, we infer that Competitor B is more efficient. It appears that
Competitor B is deploying fewer resources than A, has a better layout, and is carrying less
inventory. Clearly, a customer may not want to pay for many of the activities that Competitor A
does and for the inefficiencies in the system. As per our definition of waste, Competitor A seems
to be generating more waste in the system. Table 13.1 lists some common sources of waste in
manufacturing and service organizations as per the lean management definition.
TABLE 13.1 Some Common Sources of Waste
When organizations begin their journey towards creating a lean enterprise through a process of
continuous waste elimination, they make several changes in the organization. This includes
changing the structural elements in the system, simplifying operations planning and control by
introducing alternative scheduling methodologies, and improving the responsiveness of the
system to the market. Figure 13.1 illustrates various aspects of the lean management process. As
we have already seen, the basic premise of lean management is to eliminate waste. Therefore,
every activity undertaken under the banner of lean management must result in elimination of
waste.
The changes made are principally governed by two philosophies: just-in-time manufacturing
(JIT) and total quality management (TQM). JIT and TQM fundamentally aim at elimination of
waste. While JIT is an organization-wide mechanism for exposing problems (related to wasteful
activities in an organization), TQM is an organization-wide mechanism for solving problems.
Therefore, these two are the basic enabling mechanisms of any conceivable lean management
system.
The introduction of changes by adopting the JIT philosophy may require certain changes in the
manner in which resources are organized and scheduled. Furthermore, it may call for reducing
the lot size by embarking on set-up time reduction activities. Use of simplified operations
planning control systems also helps in dramatically reducing waste arising out of poor planning.
On the other hand, the introduction of TQM enables an organization to eliminate waste on an
ongoing basis. By setting on a trajectory of benchmarking and continuous improvement,
organizations can utilize several QC tools and small-group improvement activities to make
continuous waste elimination a workable reality.
FIGURE 13.1 The lean management process
Lean enterprises have realized several benefits. In their book Re-engineering the Corporation,
Hammer and Champy showed how, for a similar procurement requirement, Ford employed 500
engineers whereas Mazda employed only five.2 Similarly, a comparative study of automobile
manufacturers such as Toyota and GM showed that for similar production activity, GM had
nearly 27 times more employees in production and 17 times more employees in purchasing
compared with Toyota.3 Deming-winner SRF Limited claims that, after lean manufacturing is
implemented, parameters such as the size of international business, productivity of employees,
and employee involvement increased over time. Others, such as breakdown hours, customer
complaints, scrap and downgrade, and the number of employees in the organization, came
down.4 All these examples clearly demonstrate that the accrued benefit of practising lean
management is that less resources become more productive.
It is evident that improvement in operations fundamentally means better use of available
resources and greater productivity of resources. Such improvements include an increase in
production with no addition of capacity; elimination of manpower even when production levels
go up; the lack of appreciable increase in the cost of production even when the volume drops;
increase in the velocity of various business processes without anyaddition of new capacity or
technology; significant reduction in defects, rework and inventory investments when the
production rates go up; and an overall increase in the productivity. Clearly, all these point to the
elimination of wasteful activities in an organization. Moreover, on account of waste elimination,
the organization becomes more responsive, agile, and flexible to handle changing needs at the
marketplace.
The core logic of JIT is that of elimination of waste in a manufacturing system using a deliberate
method. In order to understand this logic, the water-flow analogy given in Figure 13.2 will be
useful. Consider a water-flow system such as a river. The river has rocky terrain and outgrowths
from its bed. If we want to sail on the river using a boat, the amount of water required is a
function of the technical specifications of the boat. The boat requires a certain clearance of water
above the rocks and outgrowth in order to sail unhindered.
The core logic of JIT is that of elimination of waste in a manufacturing system using a deliberate method.
One can use this analogy to represent a manufacturing system. In a manufacturing system, the
smooth flow of material is often hindered by several constraints in the system. Quality problems
at the manufacturing plant of the vendor may result in shortage and production disruptions. Poor
maintenance of resources may take away valuable capacity through unscheduled breakdowns.
Inadequate information and poor managerial decision making may stall production. In order to
protect the manufacturing plant from such disruptions, it is customary to have excess inventory
of material, manpower, machining, and other resources. We use terms such as buffer and safety
stock to denote these excess investments in resources.
FIGURE 13.2 Water-flow analogy of a manufacturing system
In the water-flow analogy, only a certain volume of water will ensure that the required
clearance for the smooth sailing of the boat is available. In the same manner, in a manufacturing
organization, a certain level of investment is made as a buffer to protect the production system
from disruptions. Therefore, the volume of water directly corresponds to the buffer built into a
manufacturing system. Similar examples can be found in service organisations also. Excess
capacity of resources (mainly human), tossing of information access various stages of operations
happen in service systems.
Let us now try to understand how organizations react to certain events. Let us assume that
there is frequent production disruption due to the poor quality of material supplied by a vendor.
Traditionally, how have organizations solved this problem? The materials and production
planning managers review the situation and re-estimate an appropriate level of safety stock to be
built into the system. It may require additions to the safety stock to restore production levels
back to normal rates. One can advance similar arguments for solving problems arising out of
excess absenteeism of workers and frequent breakdown of machines. These issues will call for
similar methods of increasing the buffer in the system Let us now try to understand how
organizations react to certain events. Let us assume that there is frequent production disruption
due to the poor quality of material supplied by a vendor. Traditionally, how have organizations
solved this problem? The materials and production planning managers review the situation and
re-estimate an appropriate level of safety stock to be built into the system. It may require
additions to the safety stock to restore production levels back to normal rates. One can advance
similar arguments for solving problems arising out of excess absenteeism of workers and
frequent breakdown of machines. These issues will call for similar methods of increasing the
buffer in the system to restore smooth production rates. In our water-flow analogy, the situation
corresponds to adding more water into the system because one of the rocks is surfacing above the
water level.
The philosophy of JIT is in contradiction with traditional thinking on solving such problems
encountered in manufacturing and service systems. It works in two ways. In the case of water
flow, instead of pouring more water into the system to cover the newly grown structure, efforts
will be made to “trim” the new growth and bring it back to a limit so that the boat can continue
to sail smoothly. In our manufacturing example, if there is a problem with the supplier, instead of
increasing the safety stock, collaborative efforts will be launched with the suppliers to solve the
problem.
The philosophy of JIT is in contradiction with traditional thinking on solving problems encountered in
manufacturing systems.
However, the crux of JIT philosophy is to go one step further and deliberately create some
disturbances in the system in order to uncover problem areas. Once the problems are exposed,
the organization will work towards solving the problem and restoring smooth production rates.
Returning to our water-flow example, what it means is that when the boat sails smoothly, we
pump out some water from the system and expose the tallest rock, then chisel the rock to a level
that the boat can resume sailing smoothly. After a few rounds of smooth sailing, pump out some
more water and continue the process.
In the manufacturing example, what it means is to have a method by which the buffer is
withdrawn from the system. By withdrawing the buffer, new problems are exposed. By studying
the problem, new methods will be devised to restore smooth production rates. After one cycle of
this exercise is satisfactorily completed, begin the next cycle by withdrawing the buffer once
again. Proceeding in this manner, the buffer in the system could be progressively reduced even
while smooth production rates are restored.
What makes JIT philosophy different from conventional thinking is the “deliberate” choice on
the part of management to expose hidden problems even while the production system is
operating at a certain level of equilibrium. Therefore, one can define JIT as an organization-wide
mandate to systematically expose hidden problems.
Figure 13.3 presents a schematic representation of JIT logic and the overall impact it has on
the manufacturing system. When the system has smoother production rates, deliberate
withdrawal of buffer inventories results in reduced buffer in the system. Moreover, as we
discussed, it exposes problems and the causes of these problems. To solve the problem, ideas are
required for controlling defects, improving delivery performances, and cutting lot sizes. These
ideas are implemented in the core manufacturing and quality assurance systems in the
organization. Once these ideas are satisfactorily implemented, the systemresumes its smooth
production rate. The upper half of Figure 13.4 portrays these components of JIT logic. The key
difference between the earlier state and the current state is that there will be a reduced level of
buffer inventories in the system.
FIGURE 13.3 JIT logic and its beneficial effects on a manufacturing system
Source: Reprinted with the permission of The Free Press, a Division of Simon & Schuster, Inc., from Japanese Manufacturing Techniques: Nine Hidden
Lessons in Simplicity by Richard J. Schonberger. Copyright © 1982 by Richard J. Schonberger. All rights reserved.
The benefits accruing to the system are not just reduction in inventory. They are manifold. The
lower part of Figure 13.4 captures the beneficial effects of implementing JIT in a manufacturing
organization. One significant benefit arising out of smaller lot sizes is the faster feedback on
quality. Assume that a firm was earlier working with a batch size of 1,000. After some
investment of efforts and technology let us assume that they brought down the batch size to 100.
This has significant advantages to the firm. First, if a batch has to be scrapped or reworked on
account of a defective process or tool discovered much later, then a smaller batch size of 100
would mean fewer hours of rework or fewer hours of valuable productive capacity lost on
account of scrap. Moreover, a smaller batch size means that the problem is discovered much
earlier. All these result in improved productivity and reduced product cost as depicted in the
lower part of Figure 13.4.
FIGURE 13.4 A schematic representation of the architectural change required for JIT
TABLE 13.2 JIT Logic and Overall Impact
Moreover, as the lot size is reduced, the succeeding process no longer views inventory as the
critical link in the operations. Rather, the preceding process and the succeeding process get
tightly linked as we typically see in the case of single-piece flow systems with no intermediate
inventory. In such systems, making system-wide improvements is not only feasible but also a
common reality. Table 13.2 summarizes the key features of JIT logic and the benefits of JIT.
The benefits accruing to the system are not just a reduction in inventory. They result in improved productivity
and reduced product cost.
Lean manufacturing principles were initially introduced by Toyota in their production systems. Over the years, all
manufacturing companies irrespective of the nature of manufacturing have embraced the lean philosophy. Of late,
Aerospace manufacturers have also found the need to implement the principles of lean. To know more about how
Boeing uses principle of lean in their manufacturing system, find the video links (Video Insights) in the Instructor
Resources or Student Resources under section Downloadable Resources (http://www.pearsoned.co.in/BMahadevan/)
subsections Bonus Material.
Although the logic of JIT is intuitively appealing, in reality, to practice it and reap the promised
benefits, an organization needs to have several key elements in the manufacturing system. We
have already seen that lot size reduction is one such element of JIT manufacturing. We discuss
some of the other key features in this section.5
Manufacturing architecture defines the nature of relationships between various functional units in an
organization and addresses issues arising out of structure, systems, procedures, and people.
The most significant change required to practice JIT is to create a new manufacturing
architecture. JIT emphasizes waste elimination, and in order to perform this, the manufacturing
architecture should be conducive. Figure 13.4 illustrates the transformation schematically. The
principles of making changes in the manufacturing architecture address two types of issues. The
first is the structural issue. Structural issues confine themselves to the physical aspects of
carving out the new architecture. This includes the layout of machines and other resources on the
shop floor and in the offices, and the organization structure and reporting relationships between
employees in an organization.
The second is a set of logistical issues. Issues related to systems, procedures, and people are
referred to as logistical issues. Organizations need to understand that changes in structural issues
need to be complemented by corresponding changes in logistical issues. Merely changing the
physical aspects at the shop floor and the offices does not provide the desired improvements. The
actual improvements and benefits accrued are commensurate to the corresponding changes made
in certain systems, procedures, and people-related issues. For example, the new structure will
demand that the skills of people be re-distributed in a different fashion. The predominant
customer focus of the whole design will force radical changes in the systems used for measuring
performance and rewarding people in the organization. New systems and procedures will be
required to seamlessly integrate the supplier layer with other layers internal to an organization.
Changing the layout, re-deploying the workforce, and redefining the organization structure are
not likely to be easy. They result in altered power structures, and call for a different working
style and behaviour. Effort is required to design a better workplace organization. Visual boards
can play a major motivational role in directing improvements and increasing the ownership of
product and process. They can play a leading role in linking all the functional areas of an
organization through an information network. Without such exercises, the agenda of waste
elimination using JIT may not be feasible.
The final outcome of the architectural changes made to a manufacturing organization is that a
system having a chain of customers–suppliers emerges7. Beginning with marketing, which is the
internal arm of the ultimate customer, a chain of preceding and succeeding processes is
established. Finished goods (FG) stores will have a supplier relationship with marketing.
Similarly, final assembly will become a supplier to FG stores. Manufacturing will resemble a
linked chain with several links. Each link will have a supplier–customer relationship and will
feed the material from the raw material stores right up to the final assembly. This chain structure
will greatly aid material flow, production planning, and control functions.
Lot-size Reduction
Capacity issues in JIT need to be addressed using an approach that is different from traditional
thinking. As JIT is about waste elimination, there is a greater emphasis on uncovering capacity
wasted in manufacturing systems due to large set-up times and poor maintenance. In Japanese
manufacturing management, capacity is nothing but a sum of actual production and waste. Thus,
if there is a mass manufacturer with an installed capacity for producing 25,000 items per month
using a set of resources, but averages a monthly production of only 22,000, then the balance
3,000 units lost is accounted as waste.
One major source for waste elimination in most organization is set-up time reduction. When
set-up time is reduced, it is obvious that the lot size will also be smaller. Imagine that there are
two identical manufacturing set-ups. In the first, the set-up time is 12 hours and in the second the
set-up time is 30 minutes. Clearly, in the first case, it makes very little sense to set up a machine
for 12 hours and engage in production for say 1 or 2 hours. Therefore, the lot size will tend to be
at least equal to the production for 12 hours. In contrast, in the second case, it is possible to set
up the machine frequently and produce in smaller quantities. This increases the ability of the
organization to respond to changes better. We have already seen the benefits that accrue to an
organization on account of smaller lot sizes.
Set-up time is the time required to set up a process (or a machine) before it could be used for production.
The earliest success stories in JIT have a strong component of set-up time reduction initiatives.
Shigeo Shingo developed the single-minute exchange of die (SMED) system to drive down set-
up time. The philosophy behind SMED is that there are internal and external operations involved
in any set-up of machines. Internal operations are those that require interruption of the machine
for performing the set-up operation. Therefore, it results in loss of capacity. For example,
changing a die in a press or a cutting tool requires stopping of the machine. On the other hand,
there are several operations pertaining to set-up that could be done off-line. These operations are
known as external operations and they do not result in any loss of capacity. Obvious examples
include planning the set-up operation, obtaining the set of tools and other resources required, as
well as obtaining the required authorizations. It is clear from this classification that internal
activities are to be kept to the barest minimum so that capacity losses and inventory build-up can
be eliminated.
Single-minute exchange of die (SMED) is a methodology proposed by Shingo to reduce set-up time.
SMED is a systematic method by which internal operations are progressively converted into
external operations9. Initially, there will be minimum use of technology to achieve this.
However, as the SMED process proceeds, it may call for a very close study of the set-up process
and use of technology to make modifications on the equipment.
Set-up time reduction happens in stages. In the first stage, obvious imperfections in planning
and procedural aspects are addressed. This may bring down the set-up time from several hours to
about an hour. In the second stage, some technology is introduced to alter the design of fixtures,
dies etc. This may bring down the set-up time to less than 30 minutes. In the final stage,
extensive use of advanced technologies to make significant design changes on the equipment
will be required to bring down the set-up to a “one-touch set-up” (of having a set-up time of
about 100 seconds). This results in uncovering large amounts of wasted capacity and making the
system flexible and responsive to changes. Batch sizes eventually drop close to one, permitting
manufacturers to develop single-piece flow systems.
Set-up time reduction is a powerful tool for continuous improvement in a manufacturing
organization. In a manufacturer of automotive components, the installed capacity of a certain part
of the manufacturing system was 32,000 components per month. However, the organization
never achieved a monthly production of more than 24,000 components. This was mainly
attributed to the loss of capacity due to the large set-up time required for the machines. Instead of
augmenting the capacity by adding new machines, the organization chose to launch a set-up time
reduction campaign. Table 13.3 shows the results after four rounds of iteration. Due to a
dramatic reduction in set-up times, the company was able to produce over 29,000 components.
Such examples are not isolated in manufacturing organizations. In fact, several progressive
organizations utilize SMED for such continuous improvement efforts and instantly reap the
benefits.
SMED is a systematic method by which internal operations are converted into external operations progressively.
Though the concept of SMED was initially applied only to presses, soon it was extended to all
manufacturing processes and later even to administrative processes. The value of set-up time
reduction lies in an organization’s ability to produce in smaller batch sizes and thereby respond
faster to market requirements. It enables an organization with the necessary conditions for lead
time reduction and JIT manufacturing.
JIT manufacturers utilize a concept known as kanban, which denotes a card or a visible signal.
Taichii Ohno, the father of the Toyota production system, conceived the logic of kanban as a
production control tool based on the stock replenishment process in a supermarket. In a super
market, various goods are displayed on the shelf in limited numbers. As customers “pull out”
their requirement, the inventory of items in display depletes. Therefore, at the end of the day (or
as soon as the shelf empties), the shelves are refilled to the extent of consumption. The use of
kanban for production control is very similar.
A pre-determined quantity of items is to be stacked between every pair of succeeding and
preceding processes. As the customer pulls out her requirement, the signal travels along the chain
and each link in the chain schedules production only to the extent of refilling the stocking points.
The signalling from the customer down to the raw material stores is done through kanban.
Kanban could be a card, an electrical signal, or a message flashed through the Web.
Table 13.4 outlines the guiding principles of kanban usage and its implications for the
production planning and control function. Kanban provides the production, pickup, and transport
quantity required in any production planning control function. This is achieved by ensuring that
the preceding and succeeding processes pick up or produce only the amount indicated in the
kanban. Since neither pickup nor production is possible without a kanban, it not only serves as a
production and move authorization note but also eliminates overproduction and excess transport.
By replacing the work order with a kanban, it is ensured that kanban is attached to goods at
various stocking points. Finally, waste reduction by the deliberate withdrawal of buffers from the
system could be achieved by progressively removing kanban cards circulating between a pair of
preceding and succeeding processes. As we discussed in the core logic of JIT, this is a long-term
focus for using kanbans as a production control tool.
Kanban provides the production, pickup, and transport quantity required in any production planning control
function.
The supermarket example clearly shows that the activities in the entire system are triggered the
moment a customer “pulls” out a certain quantity of finished goods. Since kanbans use a similar
system, the scheduling logic used is known as pull scheduling. In contrast, traditional approaches
to scheduling utilize “push” logic and are hence known as push scheduling. Before we look into
the detailed working of a kanban-based system, it is important to know the key differences
between pull-scheduling and push-scheduling logic.
The scheduling logic used in JIT is known as pull scheduling. A pull-type scheduling works with the
supermarket logic.
In contrast, a pull-type scheduling works with the supermarket logic. A customer order
triggers production in a pull-scheduling system. Assume that customer demand for 50 units of an
item is met with available finished goods in the stores. Since 50 units are dispatched, an order
will be placed with the final assembly shop to assemble 50 units of the product so that stocks are
replenished. In the process of assembling 50 units, the final assembly would have consumed the
required number of sub-assemblies as specified by the product structure. Therefore, respective
orders will be placed to the feeder shops to manufacture the sub-assemblies to the extent of
replenishing the stock of sub-assemblies consumed. The information passes in this manner up to
the raw material stores. At each stage of manufacturing, the shop pulls a certain quantity of input
material and the preceding shop responds to the pull.
The choice between a pull-type system and a push-type system is dictated by several factors
governing the operation of a manufacturing system. Pull-type scheduling cannot be implemented
in all manufacturing systems. It requires that the manufacturing system have sufficient flexibility
to respond swiftly to changes. Moreover, it implies that the organization is customer-focused and
has low demand variations. On the other hand, if the manufacturing system works with long lead
time and a greater emphasis on utilization-based measures of performance, the push method of
scheduling may be appropriate. However, since the actual customer demand is not directly
incorporated into the scheduling exercise, build-up of inventory may take place in a push system.
Moreover, an organization practising push-scheduling tends to be internally focused.
However, the choice of pull or push type of scheduling has greater implications for the
manufacturing system. These relate to the ability to react to changes, responsibility issues, and
the method and ease of control. In a pull-type schedule, the responsibility of monitoring always
rests with the centralized scheduler. If plans go awry, the scheduler will perform a major re-
planning exercise and try to bring the system back to order with a fresh set of directives
regarding what to manufacture and when. The people in the system merely wait for the next
planning document and react to it. On the other hand, in a pull system, the responsibility rests
with the operating personnel, as it is their responsibility to ensure that stocks are quickly
replenished as soon as the upstream process consumes a certain quantity. Therefore, in a push-
type system, the problems are hidden, whereas in a pull-type system, the problems are exposed
directly and create a sense of urgency among the operating personnel.
In a push-type scheduling system, problems are hidden, whereas in a pull-type system, the problems are exposed
directly.
Moreover, in a pull-type system, a visible indication of the problem is that production stops on
account of nonreplenishment of stocks by the downstream process. In the case of push-type
scheduling, the only way to know that there are problems in the existing system is by
recognizing the piling up of inventory (mostly unwanted and mismatched). Therefore, push-type
scheduling requires a considerable amount of paperwork to plan and control operations. On the
other hand, in a pull-type system, simple visible controls are established automatically by the
logic behind the system. Table 13.5 outlines the salient features of pull- and push-type
scheduling from the perspective of production planning and control.
In a dual-card kanban, two cards are made use of for planning and control purposes. These are production
kanban and conveyance kanban.
Figure 13.7 presents a schematic representation of the working of a dual-card kanban system.
Before we understand the working of the kanban system, it helps to acquaint ourselves with the
structural features of the manufacturing system as depicted in the figure.
TABLE 13.5 Push and Pull Scheduling: A Comparison
The working of the kanban is best explained using a serial set of actions taken at the succeeding
and preceding processes using the numbered sequence in Figure 13.7. The step-by-step
explanation corresponding to the numbers in Figure 13.7 is as follows:
Step 1: The succeeding process begins one cycle of production as soon as P-kanbans (which authorize the production)
and empty containers are available. One P-kanban is drawn from the kanban post and an empty container is picked up
from the outbound buffer area.
Step 2: In order to begin production, one full container with an attached C-kanban is moved from the inbound buffer
area to the processing area. The C-kanban is detached from the container and displayed at the kanban post. Production
of components begins.
Step 3: As production is completed, the P-kanban is attached to the full container of finished item and the container is
moved to the outbound area. Similarly, the empty container (since all components are used up for manufacturing) is
moved back to the inbound buffer area.
Since the succeeding process pulled out one full container of components from the inbound
buffer area, used it for manufacture and returned back the empty container, it will trigger action
in the preceding process to replenish what the succeeding process consumed. The remaining
steps describe this process.
Step 4: One empty container from the input buffer area of a succeeding process and a C-kanban from the kanban post of
the succeeding process will be taken to the outbound buffer area of the preceding process for replenishment.
Step 5: Swapping of kanban cards between containers takes place at the outbound buffer area of the preceding process.
What this means is that the P-kanban attached to the full container will be replaced by a C-kanban.
Step 6: As a result of this swapping operation, the full container and C-kanban will return to the inbound buffer area of
the succeeding process.
Step 7: The empty container will be placed in the outbound buffer area of the preceding process. The P-kanban will be
displayed at the kanban post of the preceding process.
It may be noted that using this 7-step process, we have ensured that empty containers and P-
kanbans are available at the preceding process. Therefore, we can return to Step 1 and perform
one more iteration of moving empty containers and P-kanbans to an earlier process. This is how
the pull effect cascades through the system.
The number of containers required is equal to the number of kanbans in the system.
The design of a kanban-based system very closely follows the fixed order quantity (FOQ)
system in traditional inventory control.11 In the FOQ system, an order is placed for an item when
the stock level reaches the reorder point. The quantity ordered is a fixed number arrived on the
basis of some considerations, which could include set-up costs carrying costs and other practical
considerations. These inputs are used for computing the container size and the number of
kanbans required in a JIT system.
Standard containers
In a JIT manufacturing system material, flow control is obtained through the use of
predetermined quantities stored in standard containers. The size of the container is the first issue
to be resolved. Generally, containers of smaller quantities are recommended. This has several
advantages. Smaller containers are easy to move between stations, require lesser space to store,
provide better methods of access and material handling and permit the build-up of lesser buffer
quantities in the system. Typically, the thumb rule is to have about 10 per cent of the daily
demand as the quantity per container. If the daily requirement of an item is 200 units, then the
size of the container is usually kept at 20. The size of the standard containers is denoted as Q.
Material flow control is obtained in a JIT manufacturing system through the use of predetermined quantities
stored in standard containers.
Number of kanbans
The number of kanbans is clearly the function of the demand during lead time, the safety factor,
and the size of each container. Assume that
Demand rate = D
Number of kanbans = K
Production lead time = P
Conveyance lead time = C
Safety factor = α
Container size = Q
Since the total lead time is the sum of production and conveyance lead times, the number of
kanbans is given by:
The production and conveyance lead times are critical parameters for computing K. Therefore,
organizations need to have a better understanding of what constitutes these lead times. Let us go
back to the working of the dual-card kanban system once again. The C-kanban initially waits at
the kanban post. It then moves to the output buffer area of the preceding process (with an empty
container) and moves back (with full container) to the input buffer area of the succeeding
process. It waits in the input buffer area until a container is pulled out for processing. During this
time, the C-kanban again reaches the kanban post. The elapsed time between these stages is the
conveyance lead time. It is possible to estimate an average conveyance lead time based on
several observations of this process.
We can map a similar process for estimating the production lead time. The P-kanban waits at
the kanban post. It then moves to the processing station (along with an empty container). After
processing is completed, the P-kanban comes back to the output buffer area (with a full
container). It waits at the output buffer area until a C-kanban from the succeeding process arrives
to swap the kanban cards of the containers. The P-kanban then reaches the kanban post again.
The time elapsed between these stages is the production lead time. The processing time will be
the major component of production lead time. It consists of set-up time, processing time and in-
process waiting time.
Equation 13.1 is a general expression for computing the number of kanbans. It can be suitably
adopted to compute the number of C-kanbans (Kc) and the number of P-kanbans (Kp) in the case
of a dual-card kanban. The expressions for Kc and Kp are as follows:
Number of C-kanbans:
Number of P-kanbans:
EXAMPLE 13.1
Consider a JIT manufacturing with a dual-card kanban system for production control. The
daily demand for a component is 19,200. The plant works on a single-shift basis. The
conveyance time and the processing time for the components are estimated to be 30 minutes
and 45 minutes. Assume a safety factor of 10 per cent. If the standard container for the item
can hold 240 items,
a. Determine the number of kanbans required.
b. What is the effective level protection provided for the conveyance and the production process in the system?
Solution
(a) Number of kanbans required
The daily demand is 19,200. Since the plant works for eight hours, the demand rate is 2,400
per hour, the number of conveyance kanbans (Kc) required can be computed using Eq. 13.2.
Substituting the values, we get,
Similarly, the number of production kanbans (Kp) can be computed using Eq. 13.3 as follows:
Substituting the values, we get,
Therefore, by rounding off the number of kanbans to a higher integer, we have provided a
safety factor of 20 per cent for both the conveyance and the production process.
Industrial customers demand that their suppliers reduce the price of the components that they
purchase on an annual basis. It is very common to negotiate and finalize annual supply contracts
with a requirement to reduce cost by 3–5 per cent. At an inflation of 5 per cent, this translates
into nearly a 10 per cent reduction in the cost over the previous year. They also demand lead
time reduction and greater variety in their offerings. Moreover, firms also set their own internal
targets to reduce the level of inventory and the cost of manufacturing in the future. All these will
mean that operations need to improve continuously on a year-to-year basis to match with the
market demands. Therefore, the continuous improvement of operations is an important element
of modern operations management practices in every manufacturing and service firm in the
country. Moreover, continuous improvement is a crucial element of lean management.
The success of Japanese firms in the early 1970s was largely attributed to their predominant
use of continuous improvement methods to bring down the cost and lead time in their
manufacturing operations. By continuous improvement, we mean a positive change in the
working conditions in an operating system leading to better performance of the operating system,
which would be evident from key performance measures on an ongoing basis. For example, in
1976, nearly 49 per cent of all set-up time activities in Jidhosha Kiki Ltd., Japan, took more than
30 minutes. However, by 1981, 62 per cent of all set-up time activities were completed within
100 seconds. In this example, we are able to conclude that a visible improvement in the operating
system leads to better performance and elimination of wasteful activities. Such an improvement
is typically done through the continuous improvement of operations.
Continuous improvement means a positive change in the working conditions in an operating system leading to
better performance evident from key performance measures on an on going basis.
With the available data, the organization set out on in its journey towards eliminating non-
value-added activities in the existing process. Detailed studies were conducted by groups of
people in various work areas to understand what caused these excessive waiting times and
movement in the shop floor. Based on these studies, recommendations were made which
included far-ranging options, including re-layout of the machines and significant reduction in
part varieties. An interesting fallout of the exercise was the classification of all the deburring
operations as adding cost. Immediate actions were initiated to modify the process plan to
eliminate the deburring operations altogether.
Source: S. Seshadri and B. S. Manjunath, “Creating Competitive Advantage Through Costing and Manufacturing,” Unplished Report submitted in the
MPT Programme, IIM Bangalore.
Continuous improvement method requires a lower capital outlay and can be practised on an
ongoing basis. Therefore, it is relatively easy for organizations to launch a continuous
improvement process. However, it requires greater stamina on the part of the workers and the
middle managers, along with closer monitoring and tacit support from the top management.
These are needed to sustain it over a long period of time and obtain fruitful results from it. The
challenges of a successful continuous improvement programme hinges on these issues.
In the early 1980s several Indian organizations made a beginning towards the launch of quality
circles. In several manufacturing firms, quality circles were formed, and improvement projects
were launched with much fanfare. However, within a period of 9–12 months, most of them did
not even survive and a majority of those still operational made practically no improvements in
the system. Instead of motivating employees, these attempts left a bad impression among the
employees and many felt that this was yet another trick to get productivity gains from the
workforce without paying for it. This was against the very spirit of continuous improvement
philosophy.
This experience clearly showed that continuous improvement does not take place in an
organization on its own, nor is it an accidental benefit accruing to an organization. On the other
hand, it calls for a structured process to establish and promote continuous improvement and
motivate the employees towards an improvement trajectory. In the absence of such an approach,
it is quite likely that organizations make a grand beginning and end up with nothing worthwhile.
In order that any continuous improvement effort is fruitful, it calls for a set of enabling
mechanisms to be put in place. Figure 13.8 illustrates this idea pictorially.
The first step in the continuous improvement process is to create a context for improvement.
Context provides a perspective to the continuous improvement activities undertaken by an
organization and enables a meaningful application of tools and techniques to improvement
projects. The top management in an organization has to create a constancy of purpose and
incentives for the middle management and the other workforce to make improvements by
establishing an appropriate context for improvement. Several alternatives are available for
creating the context. These include benchmarking exercises, TPM and TQM initiatives,
preparing the organization for international awards (such as the TPM Prize, the Shingo Prize and
other quality awards and certifications), and participating in industry-wide initiatives for overall
improvement of the firms in the sector. In all these cases, there is both motivation and peer
pressure for making improvements.
FIGURE 13.8 The continuous improvement cycle
The second step in the process is to set up a measurement methodology for assessing the
improvement efforts. For example, an organization may have to considerably improve its quality
and delivery to face up to competition in future. In order to quantify the nature of improvements
in these two attributes, concrete measures may be needed. The measures should be easily
measurable, objective enough, and relate directly to the attribute (quality and delivery in our
case). For example, a set of lead time measures may be used for a number of processes, schedule
adherence, and on-time delivery index as the possible measures relating to delivery. In the
absence of a measurement system, employees do not know what the effect of their improvement
initiatives has been. They also face considerable ambiguity in assessing if the quantum of
improvement made is substantial or not.
In the absence of a measurement system, employees may not know what the effect of their improvement
initiatives has been.
Making continuous improvement calls for the application of several simple tools and
techniques. In quality management literature, one finds several quality circle tools that aim at the
improvement of processes. For example, poka yoke, the mistake-proofing method, enables
organizations to improve the production process and eliminate defects. Similarly, there are other
tools available for making a systematic study of the processes and identifying improvement
opportunities. Employees need to be trained in the use of these tools and techniques for
continuous improvement. Furthermore, continuous improvement requires unique organizational
structures. It typically calls for a structure consisting of a large number of small groups of
employees. Depending on the nature of improvements to be made, the employees could be from
the same functional area or from a diverse set of functional skills. An appropriate organizational
structure will provide the required mandate for the groups to champion the cause for
improvement and attain the stated goals.
Finally, the top management needs to closely monitor the improvement programmes
undertaken and utilize every opportunity to signal the importance it attaches to these efforts. It
should also establish checks and balances in the system to ensure that there is role and goal
congruence among the various teams with respect to the improvement trajectories they are
pursuing.
The first principles of lean are applicable in all sectors of business as waste elimination is the crying need of all
organizations. Despite this basic understanding, it is not clear how the principle of lean will be applicable in offices.
Moreover, there is resistance to implement lean in offices. How can one address these issues in a service set-up? To
know more about this, find the video links (Video Insights) in the Instructor Resources or Student Resources under
section Downloadable Resources (http://www.pearsoned.co.in/BMahadevan/) subsections Bonus Material.
Continuous improvement is a matter of the systematic study of an existing process and the
identification promising opportunities for improvement. Improvements happen only when the
extent of usage of resources is reduced while discharging the same set of activities. Alternatively,
with the same set of resources, the level of activity must increase. To give a specific example, if
there are five people in the purchase department of a firm, after an improvement, fewer than five
should be able to complete the requirements of the department, or the existing five should be
able to handle more procurement-related activities. It will be difficult to make such
improvements without the use of some standard methods such as process mapping, non-value-
added analysis, business process re-engineering (BPR), and QC tools, and kaizen. We discussed
process mapping, non-value-added analysis and BPR in detail in Chapter 8. In Chapter 12, we
introduced TQM and discussed several QC tools. The fundamental philosophy of TQM is to
improve the quality of both the product and process and eliminate waste from the system.
Therefore, it is obvious that the QC tools are ideally suited for continuous improvement in an
organization. In this section, we shall look at how kaizen is employed for making improvements.
Kaizen is a Japanese term that literally means “change for the better”.13 In the context of
business organizations, it implies making continuous improvement involving everyone in the
organization in a structured manner. A 5S campaign is frequently an operational aspect of most
kaizen projects. 5S derives its name from five Japanese words, all starting with S.
Step 1: Seiri, or “sort”, deals with the contents of a workplace and removes all items that are not needed there.
Step 2: Seiton, or “set in order”, refers to “a place for everything, and everything in its place” to enable easy access to
needed items.
Step 3: Seiso, or “scrub”, refers to cleaning, so that employees can be proud about the way the workplace is organized
and kept in good condition.
Step 4: Seiketsu, or “standardize”, refers to having standards that everyone has to adhere to. Visual management is an
important aspect to facilitate easy understanding of these standards.
Step 5: Shitsuki, or “sustain”, refers to training all employees to ensure 5S application.
Typically, before the start of any 5S programme, the shop floor is not clean and also very
cluttered. Machines do not have any maintenance schedules and they accumulate lots of dust and
filth. Furthermore, improper operating procedures are adopted. This leads to a pile up of
inventory on the shop floor, poor utilization of machines and avoidable wastages. A 5S
programme helps keep the shop floor clean and orderly and provides a better working
atmosphere for everyone. Due to the daily management discipline that develops through a 5S
programme, better team play and improved employee morale and participation are promoted.
Furthermore, it promotes process ownership by teams of workers and helps them with proper
maintenance and cleaning schedules. Due to these changes, organizations experience
considerable waste reduction and overall process improvement.
Creating a context for continuous improvement is an important and the first step in a continuous
improvement process. Providing the necessary tools is also important. However, what provides
the final impetus is the setting up of appropriate organizational structures. In a normal sense,
every employee is part of an organizational structure, reports to a superior, and performs the
tasks assigned to him/her. However, when it comes to improvement, a similar structure is
required. Making improvements in the business processes often calls for collaborating with
people in different functional areas. Therefore, unless they have the required mandate, a clash of
priorities is inevitable. Employees may find it hard to choose between spending time between an
improvement initiative and the routine production tasks. Superiors will emphasize on production
tasks and signal the need to complete them before they spend time on other activities. Due to
these ground realities, employees may not be able to perform improvement activities in a
sustained fashion, and improvement activities in the organization will cease to exist soon.
In recognition of the importance of appropriate organizational structures for continuous
improvement processes, organizations have resorted to some methods for continuous
improvement initiatives. We shall briefly look at some of them.
Quality Circles
One of the popular methods to implement a continuous improvement process in an organization
is to make small groups of employees, known as quality circles, and provide them with the
required tools, techniques and the mandate to identify problems and solve them by making the
required changes in the business processes. Typically, a quality circle is made up of 8–12
employees, either from the same work area or from different areas of the organization. The
quality circle members meet at regular intervals, usually for about 90 minutes in a week and
discuss various problems and possible solutions to eliminate the problems. Instead of focusing on
several issues at the same time, they work with one problem at a time and after the problem is
solved they take up another. In order to signal the importance of such methods of continuous
improvement, the management of the organization normally permits the quality circle members
to meet during work hours at a pre-designated time. Furthermore, to promote excellence and
healthy competition, annual awards are presented to the best quality circle initiative.
A quality circle is made up of 8–12 employees either from the same work area or from different areas of the
organization.
Quality circles help organizations in several ways in their journey of continuous improvement.
They provide a structured methodology for making improvements in a sustained fashion over a longer time.
They provide the required organizational mandate for a small group of employees to make continuous improvement
without fear of their superiors and conflicts of purpose and motives.
They provide greater empowerment to the employees, raise their morale, and improve their job satisfaction.
They equip every employee with the tools and techniques for continuous improvement.
It is easy to come up with a big list of items that could be measured at each work area. However,
it is important to bear in mind that the number of items included on the display board should be
kept to a minimum. A large number of measures will confuse and divert the attention of the
employees when it comes to making a choice of improvement projects. There should be three
sets of measures on the display board. One set will consist of core operational measures ranging
from two to four. These measures will be permanently monitored and used for making
improvements in the core business processes performed at the workplace. Often, these may
include measures related to quality, lead time, schedule adherence, and customer complaints.
The second set will consist of measures that help the workers to improve on certain time-
bound programmes. For example, when the 5S campaign is on, it may be useful to have a few
measures that help in improving workplace organization. The third set of measures is intended to
bring out exceptional performances from employees at the workplace, either individually or in
groups. These measures often help to positively reinforce the employees and provide them with a
sense of satisfaction and pride. They will also help in establishing a healthy and competitive
spirit among the various small groups, thereby orchestrating a company-wide improvement plan.
Improvement projects in the core manufacturing areas usually begin when the members of
SGIA analyse the operational measures plotted on a weekly basis. For example, if the schedule
adherence or the quality is deteriorating, the group members brainstorm on the possible reasons.
Based on the discussions, an action plan is chalked out. At this stage, the help of other
manufacturing support functions, such as industrial engineering, design, quality and purchase, is
taken. The group works on this project and completes it by documenting the experiences and the
benefits obtained. Similarly, in the manufacturing support area, members representing costing,
design, and production planning may have discovered a problem of long lead time from order
enquiry to sending a quotation to the customer. They can make use of the data captured through
the dis-play board and improve the process by proposing certain administrative procedural
improvements.
Lean management is not rocket science. As we have already seen in the chapter, it uses simple
and already available tools in other domains of operations management such as TQM to set up a
robust system for waste elimination. Despite this, experience shows that organizations find it
difficult to successfully transform a traditional organization into a lean organization. The reasons
often pertain to mindsets, people and culture issues, and the varying role of the top management
in the transformation process. Lack of top management vision, practice–preach gap, reluctance to
empower people at the lower levels, tunnel vision at the top, and not knowing where and how to
make improvements are some of the often quoted reasons. We shall examine some of them.
Practitioners often think that JIT is applicable only to volume producers belonging to the
discrete engineering industry. However, the underlying principle of JIT is one of continuous
waste elimination and hence is universally applicable. Obviously, the manner in which JIT
principles are applied across different industries will vary. For example, in process industries,
potential applications of JIT philosophy would include implementation of TPM, vendor
development, and JIT purchasing. On the other hand, in the automobile sector, in addition to the
the ones just mentioned, factors such as cellular manufacturing, multi-skilled workforce, and
small lot sizing may also find application. A high-volume manufacturer may find kanban to be a
useful tool.
When different organizations begin to implement changes in their manufacturing and business
systems, they end up with a wide spectrum of variations in the nature of efforts and the quantum
of benefits realized. JIT implementation is no exception to this. The conceptual details of JIT,
TQM, and other such new approaches to manufacturing management are well known to many
Indian firms. However, there is a lack of internalization of these concepts, leading to varying
degrees of conviction and clarity at the time of implementation. This has resulted in wide
variations in the perception and implementation patterns among Indian firms.
Major JIT efforts succeed only when they are part of a strategic planning exercise. Such a
process will ensure top management’s active role and commitment. However, what hampers
successful JIT implementation is the lack of clarity and priority, leading to confusion and
wasteful expenditure of organizational energy.
Resistance to change
The fundamental issue challenging lean management efforts in any organization is the inertia to
change. Managers, line supervisors, and the workforce resist any change in their status quo. They
are often gripped by the fear of the unknown. Changes that alter the existing layout on the shop
floor will evoke significant resistance from the workforce because such changes alter the social
structure existing on the shop floor. Groups are disbanded, new groups are carved out, and the
power structure within each group is likely to be realigned. Moreover, workers tend to imagine
that the new mechanism is yet another method to get more work out of them at no extra cost.
The middle managers worry about their power structure being disturbed. They foresee an
impending reduction in their power and influence over each other and the workforce. If small
groups consisting of workers and machines are formed and encouraged to work in an
autonomous fashion with respect to several day-to-day production functions, managers and line
supervisors feel threatened. They instinctively stonewall the changes out of the fear of losing
control over the production workers. Functional managers may not understand why these
changes are required and line supervisors foresee the absence of their role in the new set-up.
Similarly, there could be resistance from several other quarters.
Tangibility of improvements
Organizations are likely to face tougher challenges while making improvements in non-
manufacturing areas of business. Unlike office areas, improvements made on the shop floor have
better visibility and are relatively easy to initiate. But that does not mean that they are the only
places to look for improvements. There is a limit to which these improvements can take place.
Moreover, it is important to realize that in order to make these improvements reach the
marketplace, the business processes in non-manufacturing areas also need improvement. Hence,
efforts are required for inculcating a continuous improvement culture among the white-collar
staff in organizations.
Nothing is taken seriously when employees in organizations come to know that changes have
neither the top management’s blessing nor their involvement and commitment. This is a
significant aspect in the introduction of a culture of continuous improvement. The top
management can help in many ways to ensure the smooth and successful implementation of the
changes. On the contrary, playing the role of a mute spectator can take away the spirit of the
whole exercise and cause significant damage, not only to the initiative, but also to the morale of
the workforce actively involved in the change process.
The top management also needs to play a vital role in communication. They should
demonstrate their support and involvement in the implementation process. Discussions about the
changes, the role played by the employees, and the benefits accrued so far should happen while
addressing the employees in all in-company programmes and functions. The top management
should be aware of the impending changes and the benefits. Only out of such understanding
comes involvement and support in resolving conflicts that arise during the implementation stage.
Middle-level managers often form part of the team of change agents. Moreover, the changes may
result in some of them giving up their traditional role of control over their subordinates. They
need adequate exposure and clarity regarding the required changes, failing which they become
insecure and try to stonewall the efforts.
SUMMARY
The philosophy of lean management is one of continuous waste elimination. Therefore, lean systems have an
unambiguous definition of what constitutes waste in the system.
Contrary to traditional thinking, the inventory in a JIT system is deliberately removed to expose hidden problems.
These problems are solved resulting in lower inventory and waste in the system, along with greater productivity.
The implementation of JIT requires that the manufacturing architecture is converted into a chain of internal
customers. Each element in the chain will have a customer-supplier relationship with the adjacent element.
Lot-size reduction and the use of standard containers are other elements of a JIT system.
Production planning and control in JIT systems is achieved through the use of kanban. There are prescribed
guidelines for the use of kanbans. Moreover, kanban enables waste elimination from the system by preventing
overproduction and exposing problems in processes.
Traditional operations systems are scheduled by the push approach. In contrast, JIT systems utilize pull-type
scheduling. Pull-type scheduling is very effective in providing visible control of the processes and bringing
problems to the surface rapidly.
The number of kanbans required is a function of production and conveyance lead time, the size of the standard
container, and the demand rate.
Organizations make improvements either in step-wise mode or a gradual mode. The step-mode improvement
happens on account of radical innovations. On the other hand, gradual-mode improvements happen on account of
continuous improvements.
A continuous improvement does not require any radically new technology or product. It concentrates on improving
the effectiveness of existing processes in small steps.
A process mapping exercise provides vital clues for improvement. It enables organizations to segregate value-
added activities from non-value-added ones and concentrate their efforts on eliminating the latter.
REVIEW QUESTIONS
5. Why does JIT manufacturing require changes in manufacturing architecture? What are the key changes made to a
manufacturing system for JIT manufacturing?
6. How does push scheduling differ from pull scheduling? What are the key implications of these two methods to an
operating system?
7. What are the key benefits of lot-size reduction to a manufacturing system? How can a manufacturing firm achieve lot-size
reduction?
8. Explain the working of a dual-card kanban system.
9. Under what conditions will the kanban-based scheduling systems fail to produce satisfactory results?
10. What are the perceptual differences between an organization implementing JIT and another not implementing JIT?
11. What do you mean by continuous improvement? Give two examples of continuous improvements that organizations
undertake.
12. What are the differences between continuous improvement and radical improvement? What is your recommendation to
manufacturing organizations for improving their performance?
13. Briefly describe the various steps involved in a continuous improvement process.
14. Compare and contrast quality circles and small-group improvement activities.
15. Explain the relevance of visual control aids for the continuous improvement process.
16. Identify some major stumbling blocks in the way of continuous improvement process in an organization. What are your
recommendations for avoiding them?
PROBLEMS
1. The following information is available from a manufacturer who is using a single-card kanban. Standard container size =
50 units, the daily demand = 1,600 units, working hours per day = 8, conveyance time = 20 minutes, production time = 30
minutes, safety factor = 5 per cent. Estimate the number of kanbans required.
2. A manufacturing system uses a single-card kanban system for production control. The system is operating satisfactorily
with a 10 per cent safety factor. The size of the standard container is 20 and the conveyance and production times are 20
and 10 minutes, respectively. The factory operates on a two-shift basis (of 16 hours). They currently use 12 kanban cards
for production scheduling
3. Consider a pair of processes scheduled using a single-card kanban system. The conveyance time and the processing time
for the items are 20 minutes and 40 minutes, respectively. Currently the system uses standard containers that can hold 200
units and has a demand rate of 2,000 units per hour. They are using nine kanbans to schedule the system. They have been
experiencing some problems in the system.
a. Can you identify the reasons for the problems that they have been facing?
b. Rectify the system so that it operates with a safety factor of at least 10 per cent.
4. A dual-card kanban system needs to be designed for a manufacturing process that has a demand rate of 200 per hour of a
certain item. The standard containers designed for the components can hold 25 items. The conveyance time and the
production time for the components are 30 minutes and one hour, respectively. Assume a safety factor of 10 per cent.
MINI PROJECTS
1. The TVS group of companies that produce automotive components has implemented several principles of JIT. Notable
among them are Brakes India Limited, Lucas TVS, Sundaram Fasteners, and TVS Motor Company. Similarly other
companies such as ABB, Maruti Udyog, Mahindra & Mahindra, Mico, Siemens, and Hindustan Motors have also
implemented JIT practices. This list of companies is partial and only suggestive. Select two companies and study the JIT-
related practices that the company has established by gathering data from the following sources:
Making continuous improvements is an essential requirement of all organizations, no matter how successful they are.
Choose one organization from the following list to conduct a improvement project:
First talk to the head of the organization to understand the major problems and concerns.
Based on your understanding of the problem and the consultation with the members in the organization, identify a
set of measures that needs to be improved in order to solve the problem and set some meaningful targets for the
same over the next six months.
Collect data to assess the current status of these measures, the business processes affecting its performance, and the
activities that currently constitute these business processes.
Perform the necessary analyses of the various business processes.
Write a final report detailing these steps and your recommendations to the organization
regarding the nature of efforts to be undertaken for improving the situation and solving some
of the problems listed. Also, clearly identify the challenges in the process and the nature of
precautions that the unit head should take while implementing your recommendations.
CASE STUDY
Implementation Challenges
A senior executive had the following to comment on the lean initiatives in BACL:
“We have made some progress in the last two years in the plant, some with suppliers and
the logistics. However, my personal sense is that it is not picking up fast enough. I think it is
primarily because: (a) there are varying levels of understanding of the concept of lean
management and its usefulness; (b) mounting pressures in business sometimes pushes us
back on the lean initiatives that we would have otherwise embarked on; (c) we have the
culture of focusing on short-term returns on the investments and expenditures that we make
on any new initiative. Lean management is not an exception to this norm.”
According to another senior manager, the implementation of the LPS did not fully meet
the requirements, although there was every effort from the top management. He had the
following to say:
“Changing the layout, re-deploying the workforce, and redefining organization structure
are not going be to be easy. It results in an altered power structure of the various individuals
in the organization. It calls for a different working style and behaviour. Efforts are also
required to design a better workplace organization.”
It was not clear if the lean management practices at BACL indeed focused on several
aspects of a lean system, with the result that the implementation of LPS could be at variance
with the core logic of lean management and might not deliver some of the promised benefits
of lean. One expert had the following to say:
“In lean management, God is in the details and the results are likely to accrue over a
slightly longer time. It also demands a different mindset from the top management, one in
which short-term orientation must give way to long-term orientation. Only with such
perspective and support will the process deliver improvements and develop new capabilities
in the organization.”
FIGURE 13.9 Implementations of LPS: some examples
While discussing the challenges of lean implementation with a senior manager at BACL,
he said:
“We have not done a very good job of reducing the batch size in our production. There is
still a dominant mindset for running larger batch sizes. I see two underlying reasons for this.
Many other lean management initiatives have also suffered from these underlying causes.
First and the foremost is that there is a genuine issue of lack of understanding of what
“lean philosophy” is all about. Therefore many do not know about the benefits of lean, leave
alone appreciating them. We have implemented the kanban system in some areas but it does
not work the way it ought to. This is an indication of a lack of clarity about the intended
system and its benefits. While we had good training programmes on lean management, I feel
it may not be enough. It requires multiple exposures and visits to a few other lean
implementations outside to appreciate and internalize these ideas. Only then will there be a
conviction and perseverance in pursuing them relentlessly.
The second stumbling block is the classical short-term-versus–long-term conflict. Short-
term benefits are easy to quantify. On the other hand, long-term benefits are quite difficult to
value in monetary terms. Does that simply mean that we can ignore long-term benefits and
save us the trouble and inconvenience of searching for ways and means of quantifying long-
term benefits? My sense is that in the existing scheme of things, we seem to take the easy
route and there is a tacit acceptance to this way of dealing with improvement projects.
I will give you one specific example. Let us look at the issue of standard containers. In
order to implement this, we were required to invest in new tote boxes and plastic containers.
In some cases it also required fabricating newer storage devices. In order to get this project
approved we may need to provide a cost–benefit exercise. Quantifying the benefits is not
impossible. But it is also not straightforward and requires straining one’s imagination. I don’t
think many of the managers would take this trouble. It is much easier to abandon this project
and continue with old storage devices. This is what you will see.”
SUGGESTED READINGS
D. Arfmann, and G. T. B. Federico, (2014), ‘The value of lean in the service sector: A critique of theory and
practice’, International Journal of Business and Social Science, February 2014, 5(2).
T. H. Davenport, Process Innovation: Reengineering Work through Information Technology (Boston, MA: Harvard
Business School Press, 1993).
R. Dubey, “Clustering for Competitiveness,” Business Today, 22 April 1999, 50–63.
R. Fukuda, CEDAC-A Tool for Continuous Systematic Improvement (Portland, OR: Productivity Press, 1990).
M. Ghosh, ‘Lean manufacturing performance in Indian manufacturing plants’, Journal of Manufacturing
Technology Management, 24(1), 113-122.
M. Greif, The Visual Factory: Building Participation Through Shared Information (Portland, OR: Productivity
Press, 1992).
R. Hall, Zero Inventories (Homewood, IL.: Dow Jones Irwin, 1983).
M. Hammer and J. Champy, Reengineering the Corporation (London: Nicholas Brealey Publishing Ltd, 1993).
M. Imai, Kaizen: The Key to Japan’s Competitive Success (New York: McGraw-Hill International Edition, 1991).
J.R. Immelt, (2012), ‘The CEO of General Electric on sparking an American manufacturing renewal’, Harvard
Business Review, 90 (3), March 2012, 43-46.
Japan Management Association, The Canon Production System: Creative Involvement of the Total Work Force
(New York: Productivity Press, 1987).
M. S. Jayaraman, G. Natarajan, and A. V. Rangaramanujam, Business Process Reengineering (New Delhi: Tata
McGraw-Hill, 1994).
R. Kaplan and M. Porter, (2011), ‘How to Solve the Cost Crisis in Health Care’, Harvard Business Review, 89 (9),
September 2011, 47-64.
I. Kobayashi, Keys to Workplace Improvement (Portland, OR: Productivity Press, 1990).
J. K. Liker, The Toyota Way (New Delhi: Tata McGraw-Hill, 2004).
J. Lovelle, “Mapping the Value Stream,” IIE Solutions 33, no. 2 (2001): 26–33.
B. Mahadevan, The New Manufacturing Architecture (New Delhi: Tata-McGraw-Hill, 1999).
B. Mahadevan, “Are Indian Companies Ready for Justin-Time,” IIMB Management Review 9, no. 2&3 (1997): 85–
92.
M. Mannon, (2014), ‘Lean healthcare and quality management: The experience of ThedaCare’, The Quality
Management Journal, 21(1), 7-10.
Y. Mondon, Toyota Production System, 2nd edition (Norcross: GA Institute of Industrial Engineers, 1993).
J. M. Nicholas, Competitive Manufacturing Management (New York: McGraw-Hill International Editions, 1998):
254–306.
T. Ohno, Toyota Production System: Beyond Large-Scale Production [Chennai: Productivity Press (India) Pvt. Ltd,
1992],
G. A. Pall, The Process-Centered Enterprise (Boca Raton, FL: St. Lucie Press, 2000).
M. Purch, “Are You Up to the Cycle Time Challenge?” IIE Solutions 33, no. 4 (2001): 24–29.
R. J. Schonberger, Japanese Manufacturing Techniques: Nine Hidden Lessons in Simplicity (New York: The Free
Press, 1982).
R. J. Schonberger, World Class Manufacturing: The Lessons of Simplicity Applied (New York: The Free Press,
1986).
S. Shingo, A Revolution in Manufacturing: The SMED System (Cambridge, Mass.: Productivity Press, 1985).
D. M. Stewart and S. A. Melynk, “Effective Process Improvement: Developing Poka-Yoke Processes,” Production
and Inventory Management Journal 41, no. 4 (2000): 48–55.
K. Suzaki, The New Manufacturing Challenge (New York: The Free Press, 1987).
K. Suzaki, The New Shop Floor Management (New York: The Free Press, 1993).
Part IV
Demand Forecasting
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
What is the role of forecasting in an organization? How does it help the planning process?
How does the forecasting methodology and context change with respect to the time horizon?
What are the steps involved in designing a forecasting system?
What are the sources of data for forecasting? How do organizations use these data?
What are the well-known models used for forecasting? How can we estimate the model’s parameters?
How can we assess the accuracy of forecasts obtained from a forecasting model?
What are the managerial considerations in using a forecasting system?
Every organization invariably engages in an annual planning exercise. The heads of various functional areas such as marketing,
production, materials, and finance take part in this exercise with specific objectives. The marketing function provides data on the
sales the organization should target in the coming year.
Source: Milltec Machinery Pvt. Ltd.
Every organization invariably engages in an annual planning exercise. The heads of various
functional areas such as marketing, production, materials, and finance take part in this exercise
with specific objectives. The marketing function provides data on the sales the organization
should target in the coming year. This is primarily achieved through forecasting. Based on this
input, the production function prepares an annual production plan and projects various
requirements on the basis of this plan. The materials function prepares a procurement plan to
match the requirements projected by the production function. Finally, on the basis of all these,
the finance function undertakes cash planning and funds management. Therefore, forecasting
plays a vital role in every organization. In this chapter, we deal with various aspects of
forecasting. This includes the need for forecasts, methods of making reliable forecasts, and
managerial issues in using a forecasting system.
Forecasts are estimates of the timing and magnitude of the occurrence of future events.
The timing of the occurrence of events is not only important in short-term planning, as
described; it is equally important in several other situations. Let us consider another example.
Suppose a manufacturer of household appliances decides to add one more product line—
microwave ovens—to its existing product portfolio. The decision to add microwave ovens to the
product line requires a good understanding of the nature of demand for the range of microwave
ovens proposed to be manufactured. Depending on both the timing and the magnitude of
demand, the manufacturer will schedule the building of new plants, planning of the product
launch, and the creation of the necessary infrastructure for the marketing and distribution of
microwave ovens.
Forecasting is an important tool in public policy decisions as well. For example, the
Government of India needs to have a reasonable estimate of the population growth over the next
10–20 years in order to formulate long-term plans for creating infrastructure for transportation
and the development of cities and towns. Based on the estimates of population growth, the
government may formulate detailed plans to channel investment into certain sectors of industry.
Another example is the meteorological department’s annual forecasting exercise to predict the
monsoons.
In all these examples, we see that forecasting plays a vital role. Specifically, one can identify
the following key functions of forecasting:
a. An estimation tool
b. A way of addressing the complex and uncertain environment surrounding business decision making
c. A tool for predicting events related to operations planning and control
d. A vital prerequisite for the planning process in organizations
Since forecasting activity typically precedes a planning process, one can identify specific reasons
for the use of forecasting in organizations. Organizations face a different set of issues while they
engage in planning, and in each of these, forecasting plays an important role as a tool for the
planning process. Based on the examples discussed, we can summarize the key areas of
application of forecasting:
Dynamic and complex environments: Forecasting is not required if an organization has complete control over market
forces and knows exactly what the sale of its products is going to be in the future. However, as we have already seen,
such situations seldom exist.
Short-term fluctuations in production: A good forecasting system will be able to predict the occurrence of short-term
fluctuations in demand. Therefore, from this knowledge, organizations can avoid knee-jerk reactions to the unfolding
reality. Production planning decisions could utilize this information and develop plans that minimize the costs of
adjusting the production system for short-term fluctuations.
Better resources management: Since the impending events in an organization are predicted through a forecasting
system, organizations can benefit from better materials management and ensure better resources availability. Returning
to the fast-food joint example, if the owner could predict the occurrence of peak hours in his joint, he would have
planned and ensured better material and greater availability of resources.
Rationalized manpower decisions: Organizations can minimize hiring and lay-off decisions through forecasting.
Moreover, better planning of overtime and idle time can also be done based on this information.
Strategic decisions: Our earlier example of microwave ovens suggests that forecasting plays an important role in long-
term strategic decision making. This includes planning for product line decisions, new products, augmenting capacity,
building new factories, and expanding the current level of activities. In each of these, an understanding of the unfolding
future is a key factor in decision making and forecasting plays a role in providing this information.
A good forecasting system will be able to predict the occurrence of short-term fluctuations in demand.
The use of forecasting is clearly pervasive. However, the type of data used, the nature of analysis
done, and the tools and techniques employed could vary from one situation to another.
Therefore, it is useful to understand what causes these variations and how one can sub-classify
and group them. Amongst other parameters, the time horizon for forecasting provides a sound
basis for classification. The three categories are short-term, medium-term and long-term
horizons. Table 14.1 lists the salient features of this classification.
Short-term Forecasting
Typically, short-term forecasting is employed to fine-tune an existing plan on the basis of the
new information obtained. Forecasting acts as an input for the tactical decisions that an
organization makes. For example, based on the sales in the last quarter, an organization could
develop better estimates of the demand in the next quarter and use that information for adjusting
various quarterly plans. The errors in forecasting that are sought to be corrected by an alternative
estimate are more related to random events than any long-term cyclical patterns or medium-term
seasonal patterns. The forecasting data is used in a disaggregated fashion and analysed in detail.
For example, the sales data will be analysed by region and product variety for possible short-
term impact in a particular region or for a variant of a product. Specific corrective measures
could be taken after analysis of the data. The techniques used in this case are simple
extrapolation of the immediate past data and mechanisms for the incorporation of new data and
period-by-period adjustments.
TABLE 14.1 Forecasting Time Horizon—Some Implications
Medium-term Forecasting
An organization uses forecasting as a starting point to the annual business planning exercise.
This typically constitutes medium-term forecasting. In medium-term forecasting, the planning
horizon is usually 12–18 months. During this period, some aggregation of data is done. For
example, if an organization offers 15 variations of a product, the demand for the product is
estimated at an aggregate level. Based on this information, capacity and material plans could be
made. Since the forecasting is done for a slightly longer time, cyclical and seasonal patterns will
make a significant impact and need to be incorporated in the analysis. The decisions taken using
the forecasting information vary from purely tactical decisions such as annual production
planning to somewhat strategic ones such as augmentation of capacity in specific areas of
business. Forecasting techniques should be able to handle these requirements. Use of
extrapolative methods, some subjective judgment and regression-based methodologies are often
employed in medium-term forecasting.
An organization uses forecasting as a starting point to the annual business planning exercise.
Long-term Forecasting
Long-term forecasts involve purely strategic decisions for a time period of about 5–10 years, so
the forecasting processes need to cater to these requirements. For instance, an organization may
be interested in projecting the future technology trends in their business and use it as the basis for
developing new products, production technology, and human and other resources. Strategic
decisions substantially draw subjective knowledge from the expertise of senior management
personnel in an organization involved in the decision making. Furthermore, the level of
uncertainty in the process tends to be high. Therefore, the forecasting methodology should be
able to use this data and develop reliable estimates. Some amount of detailed modelling based on
some macro-level assumptions is often required.
14.4 DESIGN OF FORECASTING SYSTEMS
Designing and using a forecasting system in an organization involves three important stages, as
shown in Figure 14.1. In Stage 1, an organization needs to address the requirements for the
forecasting system and identify an appropriate time horizon. On the basis of these, it needs to
further identify a suitable technique, collect data, construct the forecasting logic using available
tools and techniques, and test for logical and empirical validity using some past data.
Designing and using a forecasting system in an organization involves three important stages: (i) identifying an
appropriate time horizon, (ii) building and validating a forecasting model, and (iii) using the output from the
forecasting model with managerial judgment.
Once the designer is satisfied with the forecasting logic, the next stage involves developing
control mechanisms for using the forecasting system. This requires establishing some
performance measures and subjecting the forecasts obtained from the system for reliability from
time to time. If, after some time, the forecasting system is found to be consistently biased, then
the basic logic may require a review. In the third stage, the focus is on using the estimates
obtained from the forecasting system with a managerial perspective. It is not enough to merely
take the output from a forecasting system and make decisions based on these estimates. On
several occasions, managers need to understand the need for incorporating additional information
not captured by the forecasting system. We shall look at each of these three stages in some detail
in the rest of the chapter.
FIGURE 14.1 Designing and using a forecasting system: a three-stage process
Figure 14.2 describes the various steps involved in developing the forecasting logic. The first
step in the process is to have clarity on the purpose behind developing the forecasting system.
Why are we developing a forecasting model? Is the organization interested in launching new
product lines in the near future? Is it trying to estimate the resources required to meet the
production targets for the next year? Or is it analysing the impact of short-term changes in
demand with a view to adjust the capacity to meet the demand? As we have already seen, much
of the logic behind developing a forecasting system critically depends on these issues. In the first
case, we are aiming at a long-term horizon, whereas in the second case we are addressing a
medium-term requirement. In the last case we focus on short-term requirements. As we have
already seen, the type of data required for developing a system in each of these cases and the
nature of decisions to which the output from the forecasting system is put differs significantly.
Once the purpose, the time horizon, and the nature of data required are clearly identified, the
next step in the process involves choosing an appropriate model from the available set. This is
easily done by collecting sample data and analysing them using simple tools such as visual
inspection and statistical measures. This is an important step to understand the specific
requirements of the proposed forecasting system. For instance, let us assume that an organization
is interested in developing a forecasting system for estimating the yearly demand of the
equipment they manufacture for production planning purposes. Suppose a sample sales data for a
year for the equipment is collected and plotted. From the plotted graph, the organization may get
to know if there is a pronounced seasonal pattern. If there is a seasonal pattern, the model should
be able to handle seasonality. Similarly, the plot of the data may also sometimes show the
significant impact of business cycles. Understanding these patterns in the data will help in
choosing an appropriate set of models as possible candidates for consideration. On the basis of
these analyses, one can select an appropriate model for active consideration.
FIGURE 14.2 Steps in developing the forecasting logic
One can classify the various models available for forecasting into three categories:
extrapolative methods, causal methods, and subjective judgments using qualitative data.
Extrapolative methods make use of past data and essentially prepare future estimates by some
method of extrapolating the past data. For example, the demand for soft drinks in a city or a
locality could be estimated as 110 per cent of the average sales during the last three months.
Similarly, the sales of new garments during the festive season could be estimated to be a
percentage of the festive season sales during the previous year. In both these examples, we are
using past data and extrapolating it into the future on some basis.
Causal models analyse data from the viewpoint of a cause–effect relationship. For instance, in
the process of estimating the demand for new houses, the model will identify the factors that
could influence the demand for new houses and establish the relationship between these factors
and the demand. For example, the factors may include real estate prices, housing finance options,
disposable income of families, the cost of construction, and benefits derived from tax laws. Once
the relationship between these variables and the demand is established, it is possible to use it for
estimating the demand for new houses.
Causal models analyse the data from the viewpoint of a cause–effect relationship.
Another set of models consists of subjective judgments using qualitative data. In some cases, it
could be based on a set of quantitative and qualitative data. In several of these methods, special
mechanisms are incorporated to draw substantially from the expertise of a group of senior
managers using a collective decision-making framework. For instance, an organization such as
the Indian Space Research Organisation (ISRO) would use this method to forecast the
technology trends in space launch vehicles and the expertise that needs to be developed in the
organization over the next 10 years. Similarly, several corporate planning exercises for the next
ten years could often use this method for forecasting trends in their sector of operation.
The next stage involves developing the forecasting logic for the model selected. Each model
has some parameters that are to be fixed. For example, if an organization wants to use a simple
moving average technique for forecasting future demand, then the model parameter is the
average number of periods for moving average. In other models there may be more parameters.
Based on the sample data, the parameters of the selected models need to be established. Once the
model parameters are established, the organization could use the logic for forecasting.
However, before putting it to use, the adequacy of the model and the forecasting logic needs to
be tested. This could be done using historical data and comparing it with the forecasts obtained
from the logic developed. If the forecasts obtained compare consistently and favourably with the
actual data, then the model can be assumed to be adequate. In some cases, there may be
significant deviations between these two. In such cases, the forecasting logic needs to be
reexamined and some of the model parameters need to be adjusted. Alternatively, a completely
new model may have to be considered.
After a few iterative procedures, the designer may be convinced that the model selected and
the logic developed for forecasting are satisfactory. This would indicate that the organization has
completed the first stage in the design and use of a forecasting system.
Forecasting is often as good as the quantity and quality of data that is available with an
organization performing a forecasting exercise. This is particularly true of forecasting exercises
used for the purpose of planning. Therefore, it is important to know the type of data required and
the normal sources through which such data could be collected.
Forecasting is often only as good as the quantity and quality of data available.
Sales-force Estimates
For every organization, one of the most valuable sources of data is the sales force that operates in
the field. Since the sales force spans the entire geographical range of operation, they have access
to data on the actual consumption and the changing patterns in consumption. They may also have
information on the performance of competitor brands and the overall patterns in market share
and market growth. An organization will have to set up information-tracking mechanisms where
the sales force can make periodic entries of actual consumption in the immediate past and
projected trends. Using this data, organizations can make an end-use analysis to project emerging
scenarios in market demand and one’s own share of the total demand. Moreover, the data
obtained from the sales force is very valuable in short-term forecasting and mid-course
corrections in production and sales planning.
Advances in information technology have enabled organizations to capture data at the point of sale using POS
systems.
Forecasts from Supply Chain Partners
Obtaining POS data is often not easy. Distributors and retailers (also known as supply chain
partners) hesitate to share information. They often perceive a threat arising out of erosion of their
power. Hence, organizations often have to rely on their supply chain partners to obtain data on
actual sales during a period. Moreover, supply chain partners provide vital information on
market trends, competitor performance, and overall market sentiments and projections. These
estimates are crucial for accurate forecasting of future demand, particularly during annual
planning exercises.
B2B Portals/Marketplaces
Another useful source of data in the era of the World Wide Web is the existence of industry
portals and B2B marketplaces. These are the digital variations of trade/industry association
journals. Consider, for example, an agro-portal such as. http://indiaagronet.com. This site
provides a one-stop solution to the agricultural industry. It contains a vast amount of data on
world agriculture, market news, commodity news, price movements of agricultural commodities,
and some of the technology trends in the sector. In addition to these, B2B marketplaces and
powerful search engines on the Web provide a vital source of data for the purpose of long-term
forecasting.
A time series is simply a collection of data at fixed time intervals over several years. Since
extrapolative methods are estimates of future requirement on the basis of past data, the most
important requirement for extrapolative methods is the existence of past data. Hence, this method
is unsuitable for brand new products and new markets. For example, if Samsung wants to
estimate the demand for the Galaxy Note 10.1, the latest version of tablet smart phone for the
next two quarters, it is not possible to use this method. Established product lines will have
several data points of the past that could be put to use. Extrapolative methods are very useful for
short-term forecasts in an organization. This includes, for instance, predicting weekly/monthly
demand for several fast-moving items and forecasts of capacity requirements in manufacturing
and service organizations. Extrapolative models with some level of sophistication will also be
useful for medium-term forecasts.
Moving Averages
The simplest model for extrapolative forecasting is the method of simple moving averages. The
model has a single parameter, that is, the number of periods to be considered for computing the
moving average. For example, an organization may use a three-period moving average to
estimate the demand of one of its fast-moving products. Table 14.2(a) presents a simple
illustration of the methods of moving averages.
In the case of simple moving averages, as in Table 14.2(a), the forecast for the month of April
is the average of the actual sales in the preceding three months. Similarly, the forecast for the
month of September is the average of the actual sales during the months of June, July, and
August. The moving average method is very simple to set up but has some limitations. If there is
a significant change in the pattern, it reacts rather slowly. For instance, when there was a shift in
the demand in June, it could react to the change only by September. This is because it weighs all
the past periods of data equally. One way to take care of this limitation is to have a weighted
moving average of the past data [see Table 14.2(b) for an illustration].
TABLE 14.2 Illustration of the Moving Average Method of Forecasting
Therefore, the generalized formula for forecasting using the moving average method is:
where
Ft = The moving average forecast for Period t
n = The number of periods for moving average
Di = Actual demand during Period i
Wi = Weight for the ith period demand data
If different periods do not have different weights, the forecast obtained will be based on a
simple moving average model. In this case, Eq. 14.1 reduces to a simpler form:
There are pros and cons in using the simple and weighted moving average model. While the
weighted moving average method may appear to be more useful, it results in an added burden of
trying to fix appropriate values for the weights. How should one fix the weights in this example?
Is the immediate past month more important than the one preceding that or is it the other way
round? The other issue in using the moving average method is the logic behind the choice of n.
How should one decide on n? In general, it can be seen that when demand is stable, larger values
of n are appropriate. On the other hand, if the demand is not stable and has frequent tendencies to
have significant shifts, then smaller values of n and the use of the weighted moving average
model are likely to provide better results.
Then,
This equation is the simplest form of exponential smoothening. The smoothening effect does not
become evident in this form of the equation. However, by recursively substituting Ft terms in Eq.
14.3, one can obtain an alternative equation as follows:
Rearranging the terms in Eq. 14.3, we get
Ft + 1 = αDt + (1−α)Ft and Ft = αDt−1 + (1−α)Ft−1
Substituting the second equation into the first and proceeding in this fashion, one obtains
alternate expression for Ft + 1 as:
We see in this equation that the previous demand points are successively smoothened with a
factor of (1 − α).
Table 14.3 presents an illustration of forecasting using exponential smoothening. The
behaviour of the model parameter is of particular interest. A glance at Tables 14.3(a) and 14.3(b)
clearly shows the impact of α on the forecast. A lower value of α indicates that the forecast is not
responsive to the demand, whereas a higher value makes the forecast responsive to the demand.
In other words, by choosing a higher value for α, the model weighs recent demand points more.
Figure 14.3—which is a plot of the forecast and the demand for various values of α—graphically
portrays this phenomenon.
A lower value of α indicates that the forecast is not responsive to the demand.
Note: *Forecasts in this illustration are rounded to units; α = model parameter smoothening constant.
There are four components of a time series: trend, seasonality, cyclical, and random.
The third component is the cyclical (C) component. Several sectors of the industry are
influenced by business cycles that repeat over a much longer period, say, 10–20 years. The
global meltdown of the economy is a case in point. Several developed economies in the west
such as that of the United States are reeling under deep recession. Even in India, beginning from
late 2008, we have experienced a slowdown of the economy. All these result in cyclic demand
patterns for the business. Several factors influence business cycles and the business cycles in turn
influence the demand for products and services. The last component is the random (R)
component, also known as the “noise in the system”. There are several uncontrollable events
happening in the short-term that could influence demand. For instance, a sudden development of
inclement weather in the afternoon following a sunny morning could cause a slump in the
number of people visiting the beach and can consequently influence the demand for snacks and
beverages offered by a seaside restaurant. There are numerous such events that influence the
business in the short run. Figure 14.5 shows the seasonal, cyclical, and random components of a
time series. For instance, the seasonal factor for Month 6 in the figure is 1.10. This indicates that
the demand during that month will be 110 per cent of the average demand due to high
seasonality. The random component cannot be predicted, however, unlike the other three
components.
FIGURE 14.4 The trend component in a time series
FIGURE 14.5 An extract of the seasonal, cyclical, and random components of a time series
The forecast for the demand is obtained by incorporating the components of the time series. In
a multiplicative model, extrapolation is done using a multiplicative relationship among the four
components. Thus, a forecast of future demand is given by F = T × C × S × R. For example, the
trend component for each period in Figure 14.4 could be multiplied by the other components for
the same period in Figure 14.5 to obtain the demand forecast. The multiplicative model is more
often employed and representative of several time series data used in business applications. In
order to develop forecasts using time series data, an organization needs to identify the
components of time series that require estimation of values and use these values to obtain the
forecast. For example, an organization may choose to include trend and seasonal components of
the time series data and accordingly estimate values of T and S to obtain the forecast.
In a multiplicative model, the extrapolation is done using a multiplicative relationship among the four
components.
Then, a simple linear regression of the form Y = a + bX could be constructed to predict the
demand Y for any value of X using the equation, where
a = Intercept (at Period 0)
b = Slope of the line
X = Time period
Y = Forecast for the demand Period X
The coefficients of the regression equation could be computed using the method of least
squares:
during a period i and is the average demand for the corresponding period, then the seasonality
The seasonality index can be estimated by taking a ratio of the actual period demand with the average demand
for the period.
One can use a simple moving average method to obtain the average demand for every period
in a time series data and use that for computing the seasonality index. Further refinement of the
seasonality index is obtained by averaging the seasonality indices of the same period in the
successive years in a time series. Example 14.2 illustrates the method of computing the
seasonality indices for Example 14.1.
EXAMPLE 14.1
Solution
The model for forecasting using linear trend is denoted by Y = a + bX.
From Table 14.4, we compute the following:
TABLE 14.5 Computations for Trend Extraction Using the Method of Least Squares
Therefore, using the values from Table 14.5, we can compute b from Eq. 14.4 as:
For the sake of simplicity, all the values are rounded to the nearest integer in this
computation.
These methods construct a forecasting logic through a process of identifying the factors that
cause some effect on the forecast and building a functional form of the relationship between the
identified factors. In other words, a set of independent variables are identified and associated
with the dependent variable through a functional relationship. For example, let us consider the
demand in the country for a new product such as direct-to-home receivers (DTH). Since this is a
new product, we may not have adequate past data on the demand and may need other means of
establishing the potential demand. Even in the case of an existing product, the number of factors
that influence demand may be several, requiring us to understand the interactions among these.
Let us return to the example of forecasting the demand for polyethylene in the case of Reliance
Industries (see Ideas at Work 14.1). Several factors—including exchange rate fluctuation,
installed capacity in the country, new product launches, customs tariffs, and the price of raw
material at the international markets—influence the demand. Forecasting in these situations uses
causal methods.
EXAMPLE 14.2
Extract the seasonal component and adjust the forecast obtained for seasonality for Example
14.1.
Solution
For Example 14.1, we first obtain the average demand for each quarter using the four-period
moving average method. A four-period moving average value is centred between two
quarters. For example, the average of Year : Q1 to Year 1: Q4 is centred between Year 1: Q2
and Year 1: Q3 (Column 3 in Table 14.6). Therefore, we once again take an average of such
values obtained to finally centre the averages against each quarter (see Column 4 in
Table14.6). By this method, we obtain the average demand for Year 1: Q3 to be 394.875.
Taking a ratio of actual demand and this value provides the seasonality index for Year 1: Q3
(see the last column in Table 14.6). For example, the demand during Year 1: Q3 is lower than
the average demand for the period and the demand for Year 1: Q4 is higher than the average
demand.
We obtain more than one value of seasonality index for a particular quarter from the time
series data. In our example, the seasonality indices of Year 2: Q1 and Year 3: Q1 are
computed to be 0.960 and 1.007. An average of these values finally provides us the
seasonality index for each quarter. Table 14.7 has the computations for obtaining the
seasonality index for all the four quarters in this manner.
TABLE 14.6 Computation of Seasonal Index
TABLE 14.7 Computations for Obtaining the Seasonality Index in All Four Quarters
In general, let us consider the forecast for a dependent variable Y using n independent
variables X1, X2, X3, …, Xn. Then developing a forecasting logic requires establishing a
relationship as follows:
Y = f(X1, X2, X3, …, Xn)
The use of the causal method to extract the trend component in a time series is a frequent
application of the causal method. However, in the case of extracting the trend component in a
time series, simplifying assumptions (including linear relationship between time and demand) are
made. Other causal methods include econometric models, multiple regression models, and
technological forecasting techniques.
Causal methods of forecasting require a greater degree of mathematical treatment of the data
and a sound background in multiple regression techniques. Several computer packages such as
SPSS are available today to help the forecast designer in this process. However, developing a
good regression model requires considerable experience in using regression analysis and a good
knowledge of the problem on hand. The inclusion of variables having spurious correlation
among themselves and the dependent variable could result in forecasting models with poor
explanatory power.
Computer packages such as SPSS help the forecast designer in developing causal models.
EXAMPLE 14.3
A manufacturer of tricycles for children in the age group of two to four years commissioned
a market research firm to understand the factors that influenced the demand for its product.
After some detailed studies, the market research firm concluded that the demand was a
simple linear function of the number of newly married couples in the city. Based on this
assumption, build a causal model for forecasting the demand for the product using the data
given in Table 14.8 collected for a residential area in a city. Also estimate the demand for
tricycles if the number of new marriages is 150 and 250.
TABLE 14.8 Data for Example 14.3
X Y
New Marriages Demand for Tricycles
200 165
235 184
210 180
197 145
225 190
240 169
217 180
225 170
Solution
Since the causal relationship is a simple linear regression, we make use of the method of least
squares to determine the coefficients of the linear regression Y = a + bX using Eqs. 14.4 and
14.5. The computation is shown in Table 14.9.
Econometric Models
The primary use of causal methods lies in their usefulness in model building involving situations
with unknown functional forms. For example, if we throw an object from a height, we can model
the trajectory of the path and the time taken by the object to reach the ground from a certain
height because known functional forms and relationships govern this process. On the other hand,
a vast majority of business problems, including the problem of predicting the market for a new
product or service or the likely acceptance of a new technology defy known functional
relationships. Such a situation calls for collecting data from the market using pilot studies,
eliciting information from subject matter experts, analysing related areas, or observing some
patterns of behaviour and establishing the relationship. Often, these relationships can be
established using principles of multiple regression analysis.
In econometric modelling, macroeconomic performance is predicted for a variety of planning purposes using a
large number of variables.
Two popular applications of this include econometric modelling and technology forecasting.
In econometric modelling, macroeconomic performance is predicted for a variety of planning
purposes using a large number of variables. These variables are typically included in a multiple
regression model and the relationship between these variables and the dependent variable is
established. Using such a relationship, several predictions are made at the macroeconomic level,
and planning exercises are undertaken. In technological forecasting, a similar approach is taken
to identify a host of variables pertaining to the alternative technology choices in question and a
multiple regression model is developed to predict the impact and trends in using alternative
technologies in the future.
Developing such causal models is not only time consuming but also expensive. They demand
specialized skills of model building and analysis. The collection of a vast amount of data, use of
extensive field trials, and pilot studies will precede the development of the model. Moreover, it
will also require the use of powerful computing environments to handle complex and numerous
mathematical relationships and regression analyses. Due to these features of causal models, they
are mostly employed for analysing long-term forecasting and planning requirements in large
corporations and economic and public policy institutions.
So far, we have focused our attention on the various methods used for preparing forecasts.
Forecasting models are useful only as long as their predictions are close to reality. If a
forecasting model is found to be erroneous in predicting the demand for the future, managers
lose their faith in using the system. Moreover, incorrect forecasts could create several problems
in the organization as forecasting forms a key input to the planning function.
where
Y = Fertilizer use per hectare (in kg)
X1 = Percentage of gross cropped area under high yielding varieties
X2 = Deflated price of fertilizers (in ₹)
X3 = Percentage of irrigated gross cropped area
The regression analysis revealed that the value of b3 was 2.38 and was significant. This
indicated that gross cropped area had a positive effect on fertilizer use. Moreover, the area
under high-yielding varieties also had a positive coefficient, whereas price had a negative
coefficient, as expected. The negative and low price elasticity indicated that smaller price
changes may not affect fertilizer use significantly.
Source: Based on K. Vatta and K.C. Dhawan, “Trends in Fertilizer Consumption in Punjab,” Productivity 41, no. 3 (2000): 463–466.
Imagine that a forecasting model predicted the demand for a product in the next quarter to be
250,000 units. On the basis of this information, material procurement, production, inventory
build up, and distribution activities will be organized. If the actual demand turns out to be only
100,000 units, then there could be an excessive build-up of raw material and non-moving
inventory. On the other hand, if the demand was 350,000 units, severe shortages, last minute
rescheduling of production, rush purchasing, and expediting deliveries could all result in
excessive cost, compromise of quality, and severe strain in the operations system. Therefore,
obtaining reliable forecasts is very important.
Various methods are available to assess the usefulness of the forecasting model. All these rely
on alternative methods of comparing the actual demand with the forecast provided by the system
for each period. The most popular among them are discussed here.
The sum of errors (SFE) is merely the sum of errors during the period of consideration. For
example, if we estimate the forecast error using the monthly data for a period of n months, sum
of forecast errors is given by:
We know that forecast errors could be either positive or negative depending upon overestimation
or underestimation of the demand. Therefore, even when the system is not performing well, one
can expect a value close to zero for SFE. This happens when the system overestimates demand
as much as it underestimates it.
The tracking signal helps managers detect any drift in the forecasting system.
In Eq. 14.6, we see that the denominator is always positive and is a fraction of the numerator
value. But the numerator could be positive if the demand is more than forecast or negative if the
forecast is more than the demand during successive periods. Therefore, the sign and magnitude
of TS will indicate if and by how much the forecasting system is drifting away from the actual
demand.
EXAMPLE 14.4
Table 14.10 has data pertaining to the actual demand and the forecast using a forecasting
system for 18 time periods in the past. Compute the measures of forecast accuracy and
comment on the usefulness of the forecasting system.
Solution
All the required computations are available in Table 14.11.
TABLE 14.10 Actual Demand and Forecast Data for Example 14.4
TABLE 14.11 Measures of Forecast Accuracy Based on Data Given in Table 14.10
Key Inferences
The results in Table 14.11 bring out some of the salient features of the accuracy measures
and the performance of the forecasting system. We shall begin by observing the results at the
end of Period 6. The measure SFE tends to indicate that the system is performing well as the
cumulative difference between the forecast and the actual demand is only 4 units. However,
on looking at the demand, we realize that the system has deviated in all the periods from the
demand. As it was overestimating during some periods and underestimating in other periods,
we have not been able to detect this accurately. However, the cumulative absolute deviation
during the period is 54 units. Therefore, MAD brings sharper focus to the performance of the
forecasting system.
The MAPE measure does not indicate any significant change during the period of analysis.
On the other hand, MAD has increased steadily beginning from Period 2. This indicates that
the forecasting system is deteriorating in its performance over time. However, it still does not
indicate if the system is consistently underestimating or overestimating the demand. The TS
measure clearly shows that the forecasting system is consistently underestimating. This is
because only in such a situation will the TS have a positive value of increasing magnitude. A
graphical plot of the TS is available in Figure 14.6. It is evident from the graph that the
forecasting system has been performing without introducing much bias during the first eight
periods. One possibility of using the TS is to set some acceptable limits of positive and
negative values. Once the TS value exceeds this limit, alerts could be set up to draw the
attention of the management for some possible investigative and corrective measures.
An alternative perspective emerges from the MSE measure. The MSE measure clearly
amplifies the error by an order of magnitude. Therefore, it has a tendency to alert the
management much earlier on the impending poor performance of the forecasting system.
FIGURE 14.6 A graphical representation of TS
Getting Started
The issues of choosing the right model and estimating parameters depend on the cost, data
availability, and the time horizon. These issues have been addressed. However, as we have seen
already, forecasting models differ in terms of their level of complexity, and the degree of
mathematical and analytical skills required in setting up and analysing the results from each
forecasting system. Therefore, in the course of choosing an appropriate model, it is useful to
draw from earlier experiences of practitioners and researchers. By doing so, it is possible to
avoid reinventing the wheel. For instance, at the time of establishing the forecasting system, one
can utilize other users’ experiences of using alternative models. One interesting development
based on this approach is the concept of focus forecasting.
Choosing the right model and estimating the parameters for forecasting depend on the cost, data availability, and
the time horizon.
Focus Forecasting
An earlier research study involved testing of the appropriateness of several forecasting models
for short-term forecasting using 1,001 actual time series.1 The models included in the
comparison ranged from simple moving averages to mathematically more involved auto-
regressive moving averages (ARMA) and Box-Jenkins models. The study concluded that simple
models consistently outperformed complex models, especially in situations requiring short-term
forecasting.
Simple forecasting models consistently outperformed complex models, especially in situations requiring short-
term forecasting.
Encouraged by this result and the observed practice of using simple forecasting models, the
concept of focus forecasting was developed. In focus forecasting, a set of five or six simple
forecasting models is set up initially. During every period, forecasts are made using each of these
models and the one with the least error (as per the forecasting performance measure chosen, for
example, MAPE) will be used for forecasting the demand for the next period. Let us say we set
up four simple forecasting models:
Model 1: Forecasting for the next period is a moving average of the last three months’ demand
Model 2: Forecasting for the next period is 110 per cent of the actual demand during the same period last year
Model 3: Forecasting for the next period is a moving average of the last six months’ demand
Model 4: Forecasting for the next period is the average of the forecast values obtained using Models 1 to 3
Forecasts for the period that just ended will be made using these four models and the model with
the best forecast performance will be used for forecasting the demand for the next period. The
procedure is repeated during every period.
The critical issue in using a forecasting system is the ability of an organization to develop methods for
incorporating external information as it becomes available from time to time.
Users of forecasting systems need to decide whether they need a responsive system or a stable
system. Returning to our earlier example, during periods of significant happenings in the market
(such as sales promotion schemes and the arrival of a new competitor), one may like to
temporarily tune the system to be more responsive. After the effects of the external information
diminish, the system could be resorted back to stability.
Finally, a good performance measurement system to identify errors in forecasts is valuable in
helping the organization to periodically assess the need for either re-estimating the model
parameters or changing the forecasting model itself.
SUMMARY
Forecasting is an important planning tool in organizations. It helps to estimate future demand, predict certain
events pertaining to operational planning and control, and provides useful information for the strategic planning
exercise involving addition of new product lines, capacity, and technology.
The forecasting context and methodology varies with respect to the time horizon. Consequently, the type of data
required and the nature of analyses done also varies with the planning time horizon.
Designing a forecasting system involves three steps.
Developing an appropriate forecasting logic
Establishing control mechanisms to obtain reliable forecasts
Incorporating managerial considerations in using the forecasting system
Extrapolative and causal methods are the two generic classes of forecasting models available. While extrapolative
methods devise methods of identifying some patterns in the past data and extend a similar logic into the future,
causal methods identify cause–effect relationships between dependent and independent variables.
By changing the model parameters in the moving averages and exponential smoothening methods, it is possible to
create a responsive or a stable forecasting model.
A time series consists of four components: trend, seasonality, cyclical and random. Developing time series models
for forecasting involves estimating the parameter values for the first three components.
Several methods are available to assess the accuracy of the forecasts obtained from a forecasting model. The
tracking signal measure detects the impending tendency of a forecasting model to consistently overestimate or
underestimate the demand.
A sophisticated forecasting system is not enough; managers must use the forecasts obtained from the system in the
context of external information.
FORMULA REVIEW
a.
b.
c.
d.
5. Forecast error, γt = Dt − Ft
REVIEW QUESTIONS
1. For each of the following, identify an appropriate time horizon for forecasting:
2. What is the relationship between the forecasting time horizon and the choice of forecasting models?
3. Explain how forecasting helps an organization handle uncertainties.
4. Is there a relationship between the choice of data and the type of forecasting model that one wishes to use?
5. Suppose an organization wishes to forecast the demand for the next financial year to prepare its annual budget. What
sources of data are useful for this exercise?
6. Identify a suitable forecasting methodology for each of the following situations:
PROBLEMS
1. A manufacturer of television picture tubes has the following data (given in Table 14.13) pertaining to the actual demand
for picture tubes (in thousands) during the period April to September in a year. Use a three-period moving average
method to forecast the demand for October.
TABLE 14.13 Demand for Picture Tubes (in Thousands) Between April and September
Month Demand
April 125
May 110
June 95
July 130
August 110
September 100
2. Ganesh Darshini is known for its specialty item, the Mysore Masala Dosa (MMD). Estimating the demand for MMD
during the evening peak hours (6.00 pm to 8.00 pm) is crucial. Table 14.14 has data regarding the demand for MMD
during the last six weeks.
a. Use the exponential smoothening method to estimate the demand for the next week.
TABLE 14.14 The Demand for MMD for Six Weeks
Week Demand
Week 1 80
Week 2 95
Week 3 75
Week 4 110
Week 5 100
Week 6 90
b. Compute the MAD and SFE at the end of Week 6. (Hint: Assume the forecast for Week 1 to be the same as the
demand)
3. The demand for a certain raw material used in the manufacture of a piece of medical equipment varies from month to
month. An organization engaged in the manufacture of this equipment wants to place orders with the suppliers for the
next month. The inventory records for the component indicate the consumption pattern in the last six months (in metric
tons), as shown in Table 14.15.
a. Use the method of moving averages and forecast the demand for Month 7 using three periods as the model
parameter.
b. Suppose the three periods have the following weights: 0.50 for the immediate past, 0.30 for two periods before and
0.20 for three periods before. What will the forecast be?
c. Change the number of periods for moving average to 2, 4, and 5 and obtain the forecast in each of these three cases.
d. Compare the results obtained in (a), (b), and (c). What do you infer from this exercise?
4. Using the data given in Problem 3, compute the forecasts using the method of exponential smoothening when
Week Demand
Month 1 15
Month 2 30
Month 3 45
Month 4 25
Month 5 50
Month 6 35
a. α = 0.10
b. α = 0.20
c. α = 0.40
d. α = 0.60
e. α = 0.80
f. α = 0.90
g. Plot the forecasts obtained for various values of α. What do you infer from this exercise?
h. Compare the forecasts obtained in each of these cases with those obtained in Problem 3. What are your inferences
from this exercise?
5. An organization is in the process of setting up a forecasting system for short-term forecasting of the demand for their
product. They have data on the actual demand in the last six months (Table 14.16). However, they are unsure regarding
which model they should use and how to fix the parameters. Prepare a report on the various issues pertaining to the use of
exponential smoothening and the moving average method of forecasting.
Week Demand
Month 1 75
Month 2 60
Month 3 55
Month 4 65
Month 5 80
Month 6 90
(Hint: Identify the parameters for each of these models, perform a series of tests to fix the model parameter, and
prepare a report of inferences based on this exercise)
6. The sale of Bajaj two-wheelers in the last twelve months in a retail outlet in Chennai is given in Table 14.17. Assume that
the demand has a linear trend.
a. Establish the trend equation for the demand. What do you infer from the trend equation?
b. Use the trend equation for estimating the trend of the demand during the first three months of the current year.
c. Calculate MAPE and MSE at the end of December.
Month Demand
January 145
February 110
March 100
April 140
May 130
June 140
July 160
August 165
September 180
October 170
November 150
December 140
7. Sales of refrigerators in the past tend to have an element of seasonality in addition to the trend. The data in Table 14.18
shows the sale (in thousands) of refrigerators in the last three years. A manufacturer of refrigerators wants to forecast the
demand for the next year by extracting the trend and seasonal components of this time series.
TABLE 14.18 Sales of Refrigerators (In Thousands of Units) in the Last Three Years
a. Extract the trend and seasonal components of the time series.
b. Using a multiplicative model, prepare estimates for the next year.
c. What could be the reasons for the actual demand varying from the forecast that has been prepared?
8. An organization is using the exponential smoothening method for preparing short-term forecasts for its product with an
alpha value of 0.20. However, it is not clear if the forecasts obtained from the system are reliable enough. It is also not
clear what the most appropriate measure is for assessing the forecast accuracy. Perform the following exercises using
Table 14.19 and suggest a suitable course of action for the organization:
a. For the existing forecasting system compute the MAD and MAPE at the end of December.
b. Suppose you decide to change the value of α to 0.50, how will this impact forecasting accuracy?
c. On the other hand, if you replace the exponential smoothening model with a 3-period moving average model, will
the forecasting accuracy improve?
d. Assume that a forecasting model with a linear trend will suit the system. Is it worthwhile to shift to this model?
e. Based on your computations, what will you finally recommend to the organization?
f. For the forecasting model that you have recommended, compute the tracking signal and plot the curve. What are
your observations based on this plot?
Month Demand
January 65
February 80
March 90
April 100
May 110
June 80
July 75
August 55
September 50
October 60
November 50
December 45
9. A health services provider needs to estimate the demand for a certain type of service on a weekly basis. This information
is important for planning manpower allocation decisions. Table 14.20 has data on the demand for the services in the last 8
weeks.
The organization is currently using an exponential smoothening model with an alpha value of 0.25.
a. Demonstrate the usefulness of the focus forecasting technology by forecasting the demand during Weeks 2 to 9.
The focus forecasting system will consist of the following forecasting rules:
Rule 1: Exponential smoothening with α = 0.25 (existing)
Rule 2: Moving average with 3 periods
Rule 3: Forecast is 110% the average of last four period demands
Week Demand
Week 1 80
Week 2 90
Week 3 65
Week 4 110
Week 5 60
Week 6 80
Week 7 80
Week 8 100
10. A market research firm was contracted by an organization manufacturing air conditioners (ACs) to forecast demand for
their product. The market research firm conducted some studies in three prime markets in the country to understand what
factors influence the demand for ACs. The study revealed that the demand is dependent on two factors—the average
disposable income (in thousands of ₹) of the target customer segment and population (in thousands) in the city. The
market research firm believes that a linear relationship will satisfactorily explain the behaviour. Use the data in Table
14.21 and build a linear regression model for the demand of ACs.
TABLE 14.21 Demand for ACs With Respect to Population and Disposable Incomes
Demand for ACs (in Hundreds) Population (in Thousands) Disposable Income (Thousands of ₹)
115 240 12.0
125 230 18.5
110 155 14.0
120 260 11.0
160 340 9.0
90 165 11.5
120 125 15.5
125 140 17.0
90 110 16.0
105 200 11.5
11. A service provider has to make necessary arrangements for the work force to be employed in her service delivery
system, which is a function of the expected weekly demand for her services. Table 14.22 has details for the actual demand
for the last seven weeks.
a. Develop a forecasting model using exponential smoothening which is very responsive to the demand and use it for
forecasting the demand for week eight. Use week one demand as its forecast.
b. Develop a forecasting model using exponential smoothening which is stable and use it for forecasting the demand
for week eight. Use week one demand as its forecast.
c. Use MAD as a measure of forecast performance. Which of the mentioned two models will you recommend?
d. Will your recommendation change if tracking signal is the measure of forecast performance?
12. The forecasts and the actual demand for an item are given in Table 14.23. Using the data, compute all relevant measures
of forecast performance.
NET-WISE EXERCISES
a. What are the challenges in demand forecasting? How can you overcome some of them?
b. How do organizations benefit from accurate demand forecasting data?
2. Visit http://www.hfma.org/Content.aspx? id = 1944 and read the article by Hugo J. Finarelli, Jr and Tracy Johnson,
“Effective Demand Forecasting in 9 Steps: Shifts in Demand for a Hospital’s Services Can Occur Unexpectedly. Demand
Forecasting Can Help You Prepare for these Shifts and Avoid Strategic Missteps”, Healthcare Financial Management,
Nov 2004. Prepare a final report to answer the following questions:
a. What are the key elements of data required to forecast healthcare services? Where do you find this data?
b. What are the main steps involved in the demand forecasting exercise pertaining to the healthcare sector?
c. Can you relate the ideas given in this paper to the issue of healthcare services in India?
d. How can the policy makers make use of the demand forecasting done for the healthcare sector? Do hospitals and
doctors find any use for this exercise?
MINI PROJECTS
Visit the local branch of any bank and collect statistics on the demand for the teller and savings bank facilities in the
bank. Aggregate the data on a weekly basis. Use the data for a period of a quarter (12 weeks) and using this time series,
identify the component of time series evident in the data.
a. For the components identified, extract the component values and use a multiplicative model to design a forecasting
system.
b. Use your forecasting system and obtain the forecasts for the 12-week horizon that you have.
c. Choose an appropriate measure of performance for the system. Are you satisfied with the performance of the
system?
d. If not, identify the problems in your forecasting system and improve the forecasting logic.
e. Once you have obtained a satisfactory result from the forecasting system, prepare the forecasts for the next four
weeks.
f. Write a report and discuss the results with your instructor.
CASE STUDY
ABS Caterers
ABS Caterers operate from the Green Park area of Hauz Khas in New Delhi. Green Park is a
posh locality in South Delhi and has a large number of restaurants serving the needs of the
people in the locality. Though ABS has its main operations at Green Park, its also deals with
the catering needs of educational institutions. ABS bagged the contract for catering in a well-
known educational institution in South Delhi a few years back. They are responsible for the
catering of the students and staff at the campus.
Campus Operations
At present, ABS has two units operating in the campus: “Green View”, which is on the
seventh floor of the main academic building, and the second one, “ABS Café”, which is in
the heart of the hostel zone. Green View operates between 10 a.m. and 5 p.m. during
weekdays (Mon–Fri). It caters to the needs of the entire office crowd, students, and teaching
and non-teaching staff. ABS Café operates between 4 p.m. and midnight on a daily basis and
is mostly visited by students.
Though ABS has its main operation at Green Park, Mr Jain, the proprietor of ABS
Caterers, considers the campus catering contract to be very important. There are quite a few
educational institutions around. Jawaharlal Nehru University is near the campus. If ABS can
prove its worth in their current operations, it is only a matter of time, according to Mr Jain,
before they get contracts at other nearby institutions. This would mean a lot in terms of his
business volumes.
At present, the volume of business at the campus is not substantial compared to the overall
business of ABS, but the situation has certainly improved ever since ABS took over a few
years back. ABS has registered a steady 20 per cent increase in sales and expects it to go up
even further. This is primarily because ABS has been consistently maintaining a low price
for its items.
Though the campus operations started with a small variety of food, ABS introduces new
items whenever there is a demand for them. For example, two new South Indian dishes—
Mysore Masala Dosa and Onion Uthappam—were introduced recently. They are now
available only between 11 a.m. and 3 p.m. The management expects to introduce new items
in the coming months. Serving a typical meal request takes nearly 20 minutes of one
worker’s time.
ABS has a group of service personnel and two managers who take care of the needs of the
two centres. The service personnel can move from one restaurant to another (i.e. from Green
View to ABS Café and vice versa), if the need arises. In any case, the restaurants are not too
far and they will not object to this. At present, the timings of the two centres are overlapping
only between 4 p.m. and 5 p.m. Hence, resources management has not been a serious
problem.
Future Plans
The immediate plan of Mr Jain is to keep ABS Café open between 1 pm and midnight. This
would mean an additional three hours of service timing and overlap with Green View. He
believes that this new arrangement will not only facilitate the students, but also increase his
volume of business. However, Mr Jain has to have a relook into the resource management
problem. Mr Jain has vast experience in the catering industry and particularly in an academic
set-up. He has observed some patterns in the arrival of customers at the restaurants. Table
14.24 is a sample of his observation for the month of September 2007.
a. What are the potential advantages of forecasting requirements for the campus operations of ABS Caterers?
b. If Mr Jain wishes to develop a forecasting model, develop an overall framework for the same.
c. How should Mr Jain plan his weekly plan for deploying resource requirements? Can you estimate the number of
service personnel that he needs?
SUGGESTED READINGS
B. H. Andrews and S. M. Cunningham, “L. L. Bean Improves Call-Center Forecasting,” Interfaces 25, no. 6
(1995): 1–13.
Baldigara, T. (2013). Forecasting tourism demand in Croatia: A comparison of different extrapolative methods.
Journal of Business Administration Research, 2(1), pp 84.
P. S. Fader and B. G. S. Hardie, “Forecasting Repeat Sales at CDNOW: A Case Study,” Interfaces 31, no. 3
(2001): S94–S107.
C. S. Galbraith and G. B. Merrill, “The Politics of Forecasting: Managing the Truth,” California Management
Review 38 no. 2 (1996): 29–43.
Goodwin, P., and Fildes, R. (2011). Forecasting in supply chain companies: Should you trust your judgment? OR
Insight, 24(3), 159-167.
Lamb, A., Daim, T. U., and Anderson, T. R. (2010). Forecasting airplane technologies. Foresight: The Journal of
Futures Studies, Strategic Thinking and Policy, 12(6), 38-54.
E. Mahmoud and C. C. Pegels, “An Approach for Selecting Times Series Forecasting Models,” International
Journal of Operations and Production Management 10, no.3 (1990): 50–60.
Miller, P. E. and Swinehart, K. D. (2010). Technological forecasting: A strategic imperative. Journal of Global
Business Management, 6(2), 1-5.
A. J. Scott, “Forecasting by Extrapolation: Conclusions from 25 Years of Research,” Interfaces 14, no. 6 (1984):
52–66.
R. D. Snyder, “A Computerised System for Forecasting Spare Parts Sales: A Case Study,” International Journal of
Operations and Production Management 13, no. 7 (1993): 83–92.
Wagner, M. (2010). Forecasting daily demand in cash supply chains. American Journal of Economics and Business
Administration, 2(4), 377-383.
CHAPTER 15
Operations Planning
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Organizations often manage demand during peak times by shifting the demand to a non-peak period as capacity is far in excess
of demand during non-peak periods. This is why hotel rates are high during the peak season and special off-season discounts are
offered for summer resorts and hill stations.
ideas at Work 15.1
Operating systems in a business play the vital role of providing a continuous flow of goods and
services to customers. For instance, a bicycle manufacturer may be producing at a daily rate of
10,000 bicycles in its factory. Similarly, an airline operating between the metropolitan cities of
Mumbai and Delhi may be offering 2,000 passenger seats daily. However, these operating
systems are not like a water tap, where, with a turn of the knob, the flow alters. Changing the
daily production rate at the bicycle manufacturer to 15,000 or the seating capacity in the case of
the airline to 2,500 are not simple and instantaneous decisions.
Operating systems are characterized by complex relationships between people, resources,
market, and materials, and often involve lead time, affecting several of the processes. For
example, if the order placed with suppliers is increased by 20 per cent, it requires considerable
time at the suppliers’ end to adjust the capacity to match the supply. Therefore, prior planning
and strategies for dealing with real-time changes play a very important role in managing
operating systems and ensuring a continuous flow of goods and services to the market. We
address these issues in this chapter and several others in the book. In this chapter, we discuss
planning methodologies that address capacity requirements in the medium term (typically a one-
year period).
These questions enable the top management to link the strategic intent to various operational
choices to be made during the planning horizon. Moreover, they provide some clarity on the
level of resources to be committed and the investments and policy decisions to be taken to
execute the operational plans.
Once the level of resources to be committed is arrived at, rough-cut capacity planning needs to
be done. During this stage, planning is done to adjust the demand and the available capacity on a
period-by-period basis and ensure that the available capacity could match the demand. At this
level of planning, information pertaining to products and resources is aggregated. For example, if
a manufacturing organization is offering nine variants of a product, they are all aggregated into
one “equivalent” model for the purpose of capacity calculations.
FIGURE 15.1 A hierarchical approach to planning
Consider for example the case of a fan manufacturer such as Crompton Greaves Limited
(CGL). Suppose CGL manufactures three variations of fans: standard, deluxe, and premium.
Assume that manufacturing a unit of standard fan requires 1 hour of machine capacity. Let us
also assume that the deluxe model requires 1.5 hours and the premium one 2 hours. Then, this
information could be used to convert the demand for all variants into a single measure of
capacity. For example if the demand for standard is 100, deluxe is 200, and premium is 75, then
the aggregated requirement in terms of machine-hours is 100 × 1.00 + 200 × 1.5 + 75 × 2 = 450
hours.
Similarly, if a service provider offers three levels of service—silver, gold, and platinum—that
may place varying demands on the resources, they will be aggregated into one service version on
some basis for the purpose of planning. Therefore, planning done at this level is only rough-cut
and is usually done for a period of 9–12 months.
At the second level, detailed planning is done to ensure that capacity, material, and other
resources are available to meet the projected demand on a period-by-period basis. The capacity,
material, and resource requirements are linked to specific variations in the products to be
manufactured and services to be offered during the planning horizon. In the earlier example of
the service firm, assume that the projected demand for the three types of services is 300, 200, and
50, respectively. If they need six different labour skills and four different kinds of equipment in
varying proportions, the planning exercise is done to ensure adequate availability of these labour
skills and equipment. Clearly, such a detailed capacity planning is done at a time much closer to
the planning horizon. Therefore, detailed planning is done typically for a period of 1–3 months
duration.
The final level of detailed planning is done almost at the time the plan needs to be executed.
During this exercise, decisions concerning the use of specific resources to manufacture goods as
per plan are taken. For example, after deciding to manufacture 1,500 bicycles of a particular
variety during a month, the exact scheduling will be done. This means that the set of machines to
be used for manufacturing and the time when the “job” needs to be launched into the system in
order to meet the targeted completion time are decided. This level of planning is usually done
during a period of 1–4 weeks.
After the plans are worked out to this level of detail, the actual execution of the plan is done
using an operational control system. Basically, the plans are monitored and resources are
committed as per plan to complete the requirements. However, in real time, the plans get
disturbed due to resource unavailability, breakdowns, absenteeism, quality problems, and so on.
Any problems arising in real time are addressed to ensure the targets are met with. It may call for
a certain degree of re-planning.
The hierarchy in planning offers several advantages to the planner. Splitting the decision-
making framework into various stages enables the planner to reduce the complexity of the
planning process. Moreover, as more information is available, it can be incorporated into the
planning exercise. When doing the first-level planning, the planner tends to make several
assumptions about markets, products, and the competitive scenario for the next 9–12 months.
However, at the third level of planning, the planner gets recent information and uses it to validate
and modify several assumptions. On this basis, the planner may perform a re-planning exercise
for the next 1–3 months. Therefore, it is important to reaize that an earlier planning exercise
invariably results in a certain level of re-planning at a later stage.
An earlier planning exercise invariably results in a certain level of re-planning at a later stage.
As the name suggests, it is a planning exercise done for operations using data at an aggregate
level. Aggregate operations planning serves the critical role of translating the business plans and
strategic intent to operational decisions. Using an aggregate operations planning (AOP)
exercise, firms arrive at the quantity and timing of resources to be committed to ensure
continuous flow of goods and services to customers. Usually, the decisions involve the amount
of resources (productive capacity and labour hours) to be committed, the rate at which goods and
services need to be produced during a period, and the inventory to be carried forward from one
period to the next.
For example, at the end of an aggregate operations planning exercise, a garments manufacturer
may arrive at the following plan:
Produce 9,000 metres of cloth everyday during the period January–March, increase it to 11,000 metres during April–
August, and change the operations rate to 10,000 metres during September–December.
Carry 10 per cent of monthly production as inventory during the first nine months of production.
Work on a one-shift basis throughout the year with 20 per cent overtime during July–October.
An aggregate operations planning (AOP) decision deals with the amount of resources (productive capacity and
labour hours) to be committed, the rate at which goods and services need to be produced during a period, and the
inventory to be carried forward from one period to the next.
As we can see in this example, three critical decisions are made: the rate of production, the
amount of inventory to carry, and the amount of resource (in terms of working hours) to be
committed on a period-by-period basis. We shall see later that there are several options available
for the planner and each has its own implications on cost and availability.
The entire planning exercise is done on the basis of some aggregate unit. There is no single
basis on which the demand data is aggregated. The only requirement is the need to establish
equivalences between variations. For example, suppose that one standard model of a bicycle
takes 30 machine-hours to manufacture, a deluxe model takes 60 machine-hours, and a sports
model takes 90 machine-hours. It is obvious that manufacturing one deluxe model is equivalent
to manufacturing two standard models and manufacturing one sports model is equivalent to
manufacturing three standard models from a resource-consumption perspective. Then a monthly
demand of 1,000 bicycles of the basic model, 500 of the deluxe model, and 250 of the sports
model can be aggregated as 2,750 basic models on the basis of machine-hours.
TABLE 15.1 Aggregate Units for Operations Planning: an Example
There are several reasons for every organization, whether engaged in product manufacturing or
service delivery, to plan operations at an aggregate level. The following ones merit some
attention:
Demand fluctuations: Organizations hardly experience stable or even demand. As we saw in the previous chapter,
several sectors of the manufacturing and service industry experience a significant upswing in demand during certain
periods. The demand for garments in India is high between August and October due to the festive season. Prior planning
is required to meeting surges in demand.
Capacity fluctuations: While demand fluctuations occur on account of seasonality, there are fluctuations in capacity too.
The capacity available in the month of February will be 10 per cent lower than that in the month of May on account of
fewer calendar (and working) days. Moreover, scheduled and unscheduled plant shutdowns have a significant impact on
capacity availability.
Difficulty level in altering operation rates: Operational systems are complex. Varying the rate of operation from one
level to another requires some amount of prior planning and coordination with related systems on the supplier and
distributor sides. Thus, an organization manufacturing diesel engines at the rate of 4,000 engines per day cannot
unilaterally change it to 5,000 engines. It needs to ensure that the required material, capacity, and other resources are
available for the incremental plan. Similarly, it needs to ensure that the distributors have the required capacity to handle
the additional load on them. Similar is the case of a restaurant offering multi-cuisine menu.
Benefits of multi-period planning: It is known that planning just for a particular period with no consideration of potential
events in the near future amounts to a knee-jerk reaction than an attempt to reach optimal and cost-saving decisions. For
example, while planning the production for a month, an organization is better off by taking into consideration the likely
scenario in the next few months. If the estimates indicate a rising demand in the next few months, it is better to produce
at a slightly higher, albeit even pace and accumulate inventory to handle the additional demand in the future than to
react every month. Such month-by-month reactions to the market are not only expensive but also not feasible for the
reasons discussed earlier.
It is evident that for several reasons, supply and demand may not match exactly on a period-by-
period basis in an organization. For instance, during the festive season, a manufacturer of
appliances, such as refrigerators, may experience higher demand than average since many
employees might have received bonuses, special festival allowances, and advances. If the
manufacturer has sufficient capacity to meet this demand, then there is no problem. On the other
hand, if the capacity is less, then the organization has to either forego the unfulfilled demand or
try to influence the demand or the supply.
Therefore, aggregate operations planning is done in an organization to match the demand with
the supply on a period-by-period basis in a cost-effective manner. Matching the demand with the
supply can be done in any one of the following ways:
a. For the supply capacity that an organization has during a period, ways by which the demand could be suitably modified
are identified so that both are matched.
b. For the targeted demand to be fulfilled in every period, ways by which the supply is altered are identified so that both
are matched.
c. During every period, both the demand and the supply are adjusted in such a way that both are matched.
Aggregate operations planning is done in an organization to match the demand with the supply on a period-by-
period basis in a cost-effective manner.
Figure 15.2 provides a simple framework for the AOP exercise. The forecasting module in an
organization merely indicates the likely demand for its product or service during a period. It does
not throw light on the availability of sufficient capacity to meet the demand or the manner in
which an organization needs to address issues pertaining to supply–demand mismatch. All these
fall under the purview of the aggregate planning exercise.
The aggregate operations planning exercise begins by exploring alternative ways of modifying
demand and arriving at the period-by-period net demand during the planning horizon. The
second step is to further explore the possibilities of modifying the supply so that at the end of the
exercise the demand and the supply are matched on a period-by-period basis. On the basis of
this, various decisions on the level of inventory to be carried during every period, the rate at
which goods and services are produced and the number of people to be engaged are arrived at. If
required, additional arrangements for outsourcing and augmenting capacity are also made. In the
final analysis, all decisions with respect to use of capacity are taken through an aggregate
operations planning exercise.
FIGURE 15.2 An aggregate operations planning framework
When you want to visit your doctor, don’t you take an appointment? Do you make use of
differential tariffs and make long distance calls (STD/ISD) to your friends or family members
late at night? Why do you think hotel rates are high in Darjeeling during the peak season? It is
often the practice in organizations to manage demand using some strategies. These strategies
basically aim at shifting the demand from one period to another without losing them for want of
capacity. Normally, capacity is far in excess of demand during non-peak hours and vice versa
during peak hours. Therefore, any attempt to even out the demand by shifting the demand from
peak hours to non-peak hours is a desirable step. The aggregate operations planning exercise
provides a few alternatives to achieve this goal.
Reservation of Capacity
One method used to manage demand is to moderate and shift the excess demand out of a busy
period to any future period without completely losing it. Consider, for instance, a multi-speciality
healthcare unit in which patients arrive for consultation. If there is no reservation of capacity (in
this case, the doctor’s time for consultation), some patients may go unattended and the hospital
might lose the opportunity of servicing the demand. Therefore, an appointment system helps to
capture the demand by scheduling the service on another day when there is sufficient capacity.
Capacity reservation is a very popular method in service systems.
Influencing Demand
Another method to manage demand is to influence the demand during a particular period through
some mechanisms and induce customers to voluntarily shift the demand count structures, and
limited period product promotions. Consider the differential tariffs promoted by
telecommunication companies in the country. The demand for bandwidth tends to be low during
late night and early morning (10 p.m. to 9 a.m.). Therefore, to induce customers to use the
bandwidth during this time, special rates (quarter–charges, for example) are applied. There are
several examples similar to this, which induce customers to shift the demand from peak to non-
peak periods. This includes special off-season discounts for several seasonal products such as
summer resorts, raincoats, winter wear, and weekend tariff structures for airline tickets.
Similarly, several years ago, white goods manufacturers such as Videocon and BPL used to offer
a “wait–buy–save” scheme. In this scheme, customers willing to wait for a period of about five
weeks would get a refrigerator with a substantial discount on the product price and it would also
be home delivered according to the promised schedule. It is clear that strategies for influencing
demand stem largely from a variety of initiatives in the marketing function in an organization.
Therefore, close coordination and planning by operations planning and marketing functions is
important to implement this set of strategies.
Although both these alternatives appear to be very similar there are some differences. In the
case of reservation of capacity, there is an explicit approach to book capacity in the future. On
the other hand, promotional schemes merely encourage the customers to use the capacity. There
is still an element of uncertainty in the extent to which capacity is likely to be used during the
promotional (non-peak) period. Therefore, an accurate capacity planning exercise could be done
only in the case of the former. This explains why the appointment system is a useful tool for
managing capacity in service systems efficiently.
15.5 ALTERNATIVES FOR MANAGING SUPPLY
Organizations have several options to modify supply to meet the demand during a period. In all
these cases, the plan seeks to increase or decrease the available capacity to match the supply to
the demand during the planning horizon. Supply can be modified through one of the alternatives:
(i) inventory-based alternatives, (ii) capacity adjustment alternatives, or (iii) capacity
augmentation alternatives.
In the case of inventory-related strategies, changes are made in the inventory levels to match
the demand with supply whereas in the case of capacity related strategies, changes in capacity
availability cause corresponding changes in the supply.
Supply can be modified through one of the following alternatives: (i) inventory-based alternatives, (ii) capacity
adjustment alternatives, or (iii) capacity augmentation alternatives.
Inventory-based Alternatives
Matching the demand with supply could be done in four ways using inventory. One is to build
inventory2 during periods of lean demand and consume them during periods of high demand. For
instance, consider a textile manufacturer such as Arvind Mills. Suppose the forecast demand for
denim during the months of April–June is 35,000 metres per month and during July–September
is 55,000 meters per month. Let us also assume that the production capacity is 45,000 metres per
month. Then, the firm could maintain the production at a level of 45,000 metres during April–
September and carry inventory from the first three months to meet the demand during the next
three months. Similarly, in a fast-food restaurant, a pre-processed inventory of certain food items
is built in the preceding period in order to handle the high demand during peak periods.
Another option is to back order the current periods demand in a future period. Back ordering
is a method of pushing an order to a future period on account of insufficient inventory and/ or
capacity to supply during the current period. For example, in a watch-manufacturing firm, if the
demand for watches during the month of October is 40,000 watches and if the firm has capacity
to manufacture only 30,000 watches and has on-hand inventory of 5,000 watches, then the
balance 5,000 watches is scheduled in the month of November, although the demand is for the
month of October.
Back ordering is a method of pushing an order to a future period on account of insufficient inventory and/or
capacity to supply during the current period.
Hiring/lay-off of workers
In several service systems and in certain areas of manufacturing systems, it is possible to
increase/decrease capacity and operations rate by varying the number of workers. For instance, in
a construction project, employing more people on the site could hasten the completion of the
project. This is true of several project-oriented situations such as maintenance, overhaul, new-
product development, software development, and consulting. Similarly, in a service system, it is
customary to hire temporary workers to augment capacity during peak hours. Therefore, hiring
additional workers during periods of high demand and laying-off excess workers during periods
of lean demand could help an organization match supply with demand. The cost of this includes
cost of hiring, initial training, lost productivity due to learning effects, and severance costs
during lay-off. Moreover, only semi-skilled and less-skilled workers can be frequently hired and
laid off. Furthermore, frequent hiring and laying off causes low morale and motivational
problems among the existing workers. These considerations limit the nature of jobs and the
extent to which these options could be used.
Varying shifts
The actual capacity utilized could be different from the installed capacity in most organizations.
For example, a CNC machine could theoretically work for 24 hours a day. If it can produce
components at the rate of 1,000 pieces per hour, then the installed capacity for the machine is
24,000 components per day. However, in reality, the machine may be working for only one shift
of eight hours. Therefore the realised capacity is only 8,000 pieces per day. If there is an
increased demand for the component, then it is possible to add another shift of eight hours to
increase the production to 16,000 components per day. Similarly, during periods of low demand,
the number of shifts could be reduced.
Adjusting capacity by varying shifts is a feasible option only in the case of discrete
manufacturing industries, as process industries work in a continuous fashion. Shift changes
introduce additional costs in an organization. Several fixed and semi-fixed components of the
overheads remain unchanged when the number of shifts is reduced. Moreover, the productivity
levels decrease as the number of shifts increases. These costs need to be appropriately captured
while analysing the suitability of this option. Service systems can also employ a similar strategy
to match the supply to demand.
Other options for augmenting capacity include re-scheduling maintenance plans, launching de-
bottlenecking projects and adding new capacity. In these cases, the annualized costs of these
operations constitute the cost of the option. If the demand projections for the future indicate a
growing trend, then capacity augmentation could be considered on a permanent basis. Otherwise,
temporary capacity augmentation should be considered. In the case of service organizations
similar examples can be found. When there was a huge pile up of IT returns to be processed, the
Income Tax department outsourced the process to private players.
Table 15.2 lists all the available alternatives and the costs associated with each of these
alternatives.
An organization can augment its capacity by subcontracting (outsourcing), rescheduling maintenance, de-
bottlenecking projects, and adding new capacity.
So far, we have identified several alternatives for bridging the supply–demand gap during the
planning horizon. However, we are still uncertain about the specific combination of these that
should be used, and the implications of these in practice. There are two generic approaches to
AOP that can use a combination of these alternatives: level strategy and chase strategy. These
two approaches represent alternative modes of thinking towards employing the available
alternatives for AOP. Table 15.3 summarizes the key differences between these two strategies.
Level Strategy
In the level strategy, the emphasis is on not disturbing the existing operations at all. This implies
that the system would employ a constant workforce and/or maintain constant working hours. In
this strategy, inventory plays the vital role of linking one period with the other. Therefore, firms
often employ inventory-related alternatives to address the supply–demand mismatch. During
periods of lean demand, anticipation inventory is built and during periods of high demand, the
anticipation inventory is consumed and other alternatives such as backordering/shortage are
made use of to match supply with demand. Clearly, inventory-related alternatives are useful only
when the risks of carrying inventory are low. Therefore, several sectors of industry operating in a
made-to-stock environment and products with low technological obsolescence are suitable
candidates for using this strategy.
In the level strategy, the emphasis is on not disturbing the existing operations at all.
In a constant workforce strategy, an organization having a certain number of workers may not
hire new workers or lay off excess workers in response to changes in demand. They will utilize
other means of addressing the supply–demand mismatch. This is especially true of highly skilled
employees. In such cases, the high costs of hiring, training, and laying off may be responsible for
several organizations resorting to a constant workforce strategy. A case in point is the recent
experiences of firms operating in the information technology (IT) sector in India. As the global
recession continues to affect several economies there could be a drop in the demand for IT
services in the country. Firms such as Infosys and Wipro will resort to undertime strategy rather
than laying off software engineers. At most, they may temporarily halt fresh recruitment of
employees.
The other possibility is to maintain constant working hours. A constant operations rate is often
achieved by maintaining constant working hours. For instance, a firm operating on a single-shift
basis will continue to work in the same fashion. A constant operations rate is an appropriate
method for achieving a level strategy in a highly automated and mechanized set-up. In such an
organization, the operational rate is a function of the number of working hours. Process
industries and highly mechanized batch-processing units fall under this category of such
industries—they include two-wheeler and passenger-car manufacturing and petrochemical and
pharmaceutical manufacturing.
In several organizations, level strategy is obtained only by maintaining constant working hours
and a constant number of workers. These firms may employ a variety of automated and semi-
automated set-ups for offering products and services.
Chase Strategy
At the other end of the spectrum is the chase strategy. In this method of AOP, very little or no
inventory is carried from one period to another. Rather, the supply–demand mismatch is
addressed during each period by employing a variety of capacity-related alternatives. For
example, during periods of high demand, additional workers are hired, the number of working
hours is increased, workers are permitted to do overtime, and more capacity is obtained by
outsourcing the unmet demand. Similarly, during periods of low demand, some workers are laid
off, others are permitted to go on undertime, and the number of working hours is reduced by
reducing the number of shifts, and, in extreme situations, even the duration of the shift. Clearly,
these strategies are more appropriate when it is not possible to stock inventory. Several service
systems and made-to-order project type of organizations fall under this category. As we can see
from our discussion of AOP alternatives, capacity augmentation and capacity adjustment
alternatives are suitable for a chase strategy. However, for reasons already described (and due to
the costs of these alternatives), organizations may prefer to exploit capacity adjustment
alternatives before employing capacity augmentation alternatives.
In the chase strategy, the supply–demand mismatch is addressed during each period by employing a variety of
capacity-related alternatives.
Mixed Strategy
The chase and the level strategies represent two pure forms of aggregate operations planning. In
reality, one can expect organizations to utilize a combination of the available alternatives to
devise a strategy for aggregate operations planning. Mixed strategy is therefore defined as one
that employs a combination of the available alternatives for AOP. The following examples
illustrate this concept:
Mixed strategy 1: Hire a certain number of workers at the beginning of the planning horizon, maintain a constant
operations rate, and utilize inventory-based alternatives to match supply with demand.
Mixed strategy 2: Do not hire or lay off workers, adjust the operations rate by varying the number of shifts, use
inventory and subcontracting to match supply with demand.
Mixed strategy 3: Do not hire or lay off workers, use subcontracting during periods of high demand and build inventory
during periods of low demand.
In all these examples, we see that the strategy seeks to maintain level operations during periods
of low demand by building inventory and chase the demand by additional alternatives during
periods of high demand. Thus, the focus shifts between level and chase. The cost structure of the
alternatives will determine the total cost of each of these strategies.
In reality, there can be numerous combinations of the AOP alternatives. Each combination
could form one mixed strategy with a certain cost of the plan. Therefore, arriving at the best AOP
strategy involves enumerating the pure as well as the numerous mixed strategies and choosing
the one with the least cost. Aggregate planning methods seek to fulfil this requirement.
EXAMPLE 15.1
TABLE 15.4 Forecast Demand for Models of Switchgears and the Number of Working Days
4. Inventory carrying costs are ₹1,000 per switchgear per month and unit shortage/backlogging costs are 200 per cent of unit
carrying cost.
Devise a level production strategy with constant workforce and constant working hours and
compute the cost of the plan.
Solution
Computing the Period-by-period Demand–supply Mismatch
The first step in AOP is to compute the demand– supply mismatch during the planning
horizon. Using the information available this is computed and presented in Figure 15.3. Since
the firm works for eight hours a day employing 125 people, the capacity available is 1,000
hours per day. Similarly, since each switchgear requires 100 hours, we can compute the
number of hours required during a period. Figure 15.3 shows the relevant calculations.
EXAMPLE 15.2
Consider Example 15.1. Assume the switchgear manufacturer has no opening stock of
inventory and chooses to devise a chase production strategy. The following additional
information is available:
1. Overtime costs are ₹40 per hour and undertime costs are ₹20 per hour.
2. Temporary workers can be hired on a monthly basis. The relevant cost of this option including firing them at the end of
the month amounts tò7,500 per worker.
Evaluate the following options for chase strategy and offer your suggestions to the
switchgear manufacturer.
a. Utilizing overtime and undertime alternatives
b. Using hiring and laying-off alternatives for capacity adjustment
Solution
(a) Chase strategy using OT/UT
In this strategy, the demand–supply mismatch is addressed through use of OT during periods
of high demand and UT during periods of low demand. For example, consider the
computations in Example 15.1. During April, the demand for capacity is 25,000 hours and
the available capacity is only 23,000 hours. Therefore, 2,000 hours of OT will be used at the
rate of ₹40 per hour. Similarly, during the month of September, the demand is only 18,000
hours whereas the supply is 22,000 hours. Therefore, 4,000 hours of UT will be allowed
during September at the rate of ₹20 per hour.
Since chase strategy does not result in carrying over inventory, there will be no inventory
carrying costs. By computing the extent of OT/UT required and the costs associated with this
in every period one can arrive at the cost of the plan. Figure 15.5 shows all the relevant
calculations. The total cost of the plan is ₹1,440,000.
(b) Chase strategy using hire–lay-off
In the hire–lay-off alternative we compute the number of workers to be hired/laid off by
examining the demand–supply mismatch. When the demand greater than the supply, we
resort to hiring. For the given example, consider the month of April. The demand is in excess
of supply by 2,000 hours. If we hire one worker, he/she will contribute 184 hours as there are
23 working days and each day we have eight working hours. Therefore, the number of
workers to be hired during April is 2000/184 = 10.86 ≅ 11 workers. In the case of a lean
period, we merely incur UT costs as we do not fire the enisting permanent workers. For
example, in the month of january we incur UT cost of ₹60,000 (3,000 × 20).
Since the number of workers to be hired/laid off can only be integers, we round off the
number to be hired to the next higher integer. In this process we also incur some idle time by
retaining a little more capacity than needed. The cost of the plan takes all these into
consideration. Figure 15.6 shows all the relevant details for the computation. The total cost of
this plan is ₹1,572,420.
FIGURE 15.5 Total cost of the chase strategy using the OT/UT option
Clearly, the cost of the hire–lay-off plan is higher than the cost of the OT/UT plan.
Moreover, there are other costs that are hard to quantify in the former case. For instance,
frequent hiring/lay-off of workers may not be feasible in several sectors of industry involving
skilled sets of workers. The training costs and lost productivity during the learning phase
would have been grossly under-estimated. Moreover, such practices could result in low
morale/motivation among the workers, leading to productivity and quality problems. These
costs are difficult to quantify. Therefore, it is a good practice to deploy the hire–lay-off plan
only as a last resort.
FIGURE 15.6 Total cost of the chase strategy using the hire/lay-off option
EXAMPLE 15.3
Consider Example 15.2. Assume that on the basis of these computations, the switchgear
manufacturer has come up with a plan that employs the following alternatives:
a. Hire 25 more workers at the beginning of April.
b. Lay off the 25 workers at the end of August.
c. Maintain constant working hours (1 shift of 8 hours) throughout the year.
d. Absorb the demand–supply mismatch by building inventory during periods of lean demand and by resorting to OT
during periods of excess demand.
Assume the cost of hiring to be ₹7,500 per worker and the cost of firing to be ₹5,000 per
worker. Evaluate the cost of this plan and compare it with the earlier alternatives. What are
the key inferences?
Solution
The Impact of the Hiring–lay-off Decision
25 workers are hired at the beginning of April and laid off at the end of August. This would
mean that during the months of April to August the number of workers available at the
switchgear manufacturing facility is 150. The capacity available (in hours) during this time is
computed on this basis. For example, during May the capacity available is (22 days × 8 hours
per day × 150) = 26,400 hours. The capacity available during the months of September to
March is the same as computed in Example 15.1, as the number of workers remains at 125.
Figure 15.7 shows these capacity calculations (see Column 5).
The decision affects the cost of the plan due to hiring charges at the rate of ₹7,500 per
worker for 25 workers in the beginning of April and laying-off charges at the rate of ₹5,000
per worker for 25 workers by the end of August.
We find from Figure 15.7 that except during the month of March, there is an inventory
build-up in all other periods. Only during the month of March are 1,600 hours of OT
employed at the rate of ₹40 per hour. Figure 15.7 summarizes the total cost of this strategy.
The total cost of the plan is ₹812,500.
Key Inferences
We first summarize the cost of each of the strategies that we analysed.
It has been found in several situations that the cost of chase strategy is higher than other
strategies. This is primarily due to the costs of disturbing the operating system, both at the
manufacturer and at the supply-chain-members’ level. Crash purchasing of material, uneven
utilisation of resources, and payment of premium for making various resources available at
the last minute are some of the reasons for high cost of chase strategy.
The examples we have discussed demonstrate the fact that levelling and chasing are two
pure strategies; in reality, therefore, organizations should use a mixed strategy by utilizing
several combinations of the available alternatives for AOP. It is interesting to note that for
the given cost structure, a mixed strategy has resulted in the lowest cost. Therefore, it is
worthwhile to explore all the available combinations of AOP alternatives and select the one
with the least cost. Clearly, it requires robust models to accomplish this task.
EXAMPLE 15.4
A manufacturer of industrial motors produces four varieties of motors. The Type 4 motor
requires a larger amount of customization than the other three. The production planning
department has estimated the number of hours required for manufacturing one unit of motor
of each variety. Type 1 requires 125 hours; Type 2, 175 hours; Type 3, 200 hours; and Type
4, 350 hours. The forecasts for the four varieties are given in Table 15.5.
TABLE 15.5 Forecasts for the Four Varieties
The overtime premium is estimated to be ₹20 per hour while undertime costs are 50 per cent
of the OT premium. The cost of carrying inventory is ₹15 per hour per month and the
backordering/shortage costs are 200 per cent of inventory carrying costs. The company can
subcontract capacity from a reputed vendor and the additional costs related to subcontracting
is ₹25 per hour of capacity.
Evaluate the following alternatives for AOP:
a. Prepare a level strategy. Use inventory/backlogging to absorb the demand–supply mismatch.
b. Use constant production rate for the first six months, change the production rate to another level and have a
constant production rate for the next six months. Use inventory/backlogging to absorb demand–supply mismatch.
c. Maintain a production level of 31,000 hours per month and use OT/UT alternatives to chase the demand.
d. Maintain a production level of 31,000 hours per month. Build inventory during periods of low demand and use
subcontracting when the demand exceeds available capacity.
Solution
Aggregation of Product Demand
Since there are four products, we use the number of hours as the basis for aggregation and
convert the demand in units of each variety to demand in hours. Figure 15.8 shows the
details. Let us consider the month of October. Since the demand for the number of Type 1
motors is 100, and since it requires 125 hours for each unit, the demand for Type 1 in the
month of October is 12,500 hours. Similarly, the demand for Type 2, Type 3 and Type 4 are
7,000 hours, 10,000 hours, and 3,500 hours, respectively. Therefore, the aggregate demand
during October is 33,000 hours. All the values in Figure 15.8 are calculated in a similar
fashion.
Key inferences
Example 15.4 points to a few important guidelines for aggregate operations planning.
First, the chase strategy could prove to be more expensive than level strategy in situations involving items that are
not made-to-order and vice versa. This is because some benefits of carrying inventory for a limited period and
maintaining constant production rate are missed in chase strategy.
It may be worthwhile to explore fewer production-level changes in a planning horizon even while implementing
level strategy.
Finally, mixed strategies provide a fertile ground for identifying the most appropriate AOP for an organization.
It is evident from these four examples that organizations need to devise a methodology to
generate several alternatives for AOP and evaluate these alternatives before selecting the most
appropriate one. In the absence of this, organizations may not be sure if the chosen strategy is the
best. Moreover, they may not also know how much they could have improved by enumerating
other strategies. Aggregate operations planning methods fulfil this objective. There are several
methods available to the planner to evaluate alternative strategies before selecting one. Broadly,
the available methods could be classified into heuristic methods and optimal methods.
Search methods
Search methods are useful in ensuring that “promising” candidates are not missed out in the
enumeration process. In all the search methods, certain parameters are chosen and alternatives
are generated by systematically varying the parameter values. As the parameter values are
modified, newer alternatives are considered and the quality of the solution is evaluated. The
search usually ends on the basis of some stopping criterion. The stopping criterion usually
depends on the rate of improvement of the solution and/or the number of iterations done. At the
end of the search, the best solution is identified.
A generalized search procedure is the random search process. In this process, sampling values
for the parameters under consideration randomly creates alternative AOP plans. For example,
suppose the parameters for consideration are desired production rate, number of employees,
number of working hours and the inventory at the beginning of the planning horizon. Using a
random number generator, a candidate solution for AOP could be sampled. As each plan is
evaluated, the existing solution is updated. The procedure stops when the quality of the solution
does not improve significantly and the best solution at the time of terminating the procedure is
identified to be the solution to the AOP problem. Figure 15.13 illustrates the generalized search
procedure for AOP.
Heuristic methods offer some advantages. First is the simplicity with which the method can be
implemented. In several cases, one can set up the heuristic in a spreadsheet and quickly analyse
the alternatives. It usually does not demand high levels of mathematical sophistication on the
part of the user. However, the methods suffer from some drawbacks. Heuristic methods generally
do not enumerate all the possible candidates in the solution space. Rather, they enable candidates
with some knowledge of the problem to eliminate some from consideration. Therefore, the
ability of the user to correctly identify the candidate solutions is critical to the usefulness of the
method.
The other limitation of the trial-and-error method is that there is no guarantee of obtaining the
best solution. Furthermore, users may not even know how far the best solution obtained from the
method is from the optimal solution to the problem. Therefore, before using heuristic methods,
users need to make a realistic assessment of these issues. With such a perspective, the use of
heuristic methods could provide some promising solutions to the AOP problem.
FIGURE 15.13 Search procedure for AOP
Optimal methods include the use of linear programming (LP) and the transportation and dynamic programming
(DP) models.
15.8 OPERATIONS RESEARCH TOOLS FOR OPERATIONS PLANNING
Operations research (OR) was developed during the Second World War to systematically model
and analyse several business problems. The optimum utilization of resources, maximization of
profit, and minimization of cost were some of the pressing requirements of various government
and private sector institutions at that time. Operations research tools provided them with a basis
for meeting these requirements while taking several decisions. The application of OR tools in
several areas of business happened in rapid succession in the post–World War scenario.
Operations management problems are extensively addressed by OR researchers. In this section
we will see how well known OR tools such as linear programming (LP) and transportation can
be applied to operations planning problems.
The central issue in aggregate production planning is to balance the supply with the demand on a period-by-
period basis. This structure of APP lends itself to a straightforward application of the transportation model for
solving APP problems.
EXAMPLE 15.5
A manufacturer needs to perform AOP and has the following information about to the
problem:
Demand information: The demand forecast for the next four quarters is 100,000, 50,000, 70,000, and 60,000 units
respectively. It is the company’s policy to plan for an additional inventory of 10 per cent of next period’s forecast.
Moreover, planning needs to take into consideration that 25,000 units of inventory will be available at the
beginning of Period 1 and 40,000 units must be made available at the end of Period 4.
Supply options and costs: The manufacturer can produce 80,000 units through regular production and add another
20 per cent through overtime. Adequate capacities are available with the supply base and it is estimated that up to
100,000 additional units can be obtained through subcontracting. Regular production costs ₹80 per unit while
overtime production costs ₹120 per unit. The cost per unit if subcontracted is ₹105 per unit. It is also possible to
carry over inventory from one period to another at a cost of ₹2 per unit per period.
Solution
Requirements for Each Period
Period 1: 100,000 + 10% × 50,000 = 105,000 units
Period 2: 50,000 + 10% × 70,000 = 57,000 units
Period 3: 70,000 + 10% × 60,000 = 76,000 units
Period 4: 60,000 + 10% × 100,000 = 70,000 + ending inventory of 40,000 = 110,000 units
(Since no information about the next period forecast is available, to simplify the process we
use the current year’s Period 1 forecast as a substitute)
Cost parameters:
Cr = Per unit cost of regular production
Co = Per unit cost of overtime production
Cs = Per unit cost of subcontracted units
Ch = Per unit cost related to hiring of workers
Cl = Per unit cost related to laying off workers
Ci = Per unit costs related to inventory
Other parameters:
Dt = Projected demand during the period
K = Minimum amount to be subcontracted
α = Maximum allowable overtime as a proportion of regular production
Objective function:
In this model, the total cost of aggregate planning is minimized subject to the meeting of the
constraints. Equation 15.1 shows all the cost components associated with APP. Equations 15.2
and 15.3 are the production rate balance and the inventory balance. Equations 15.4 and 15.5
pertain to the constraints for the subcontracting and overtime alternatives of APP. Equation 15.6
is the non-negativity constraint. For a planning horizon of one year split into 12 months, the
above formulation will result in 48 constraint equations and 72 variables. As this is a linear
programming model, the most important assumption that we are making is that all costs are
linear in the decision variables. In certain circumstances, this assumption many not be
appropriate.
In the above formulation, there are 9 different cost coefficients. Depending on the specific cost
structure and the nature of operations of a manufacturing set-up, these coefficients are to be
arrived at. As mentioned before, Dt is the expected demand for the period. The most appealing
feature of LDR is that the optimal policy for Pt and Wt has a simple form involving a linear
equation. However, the computational advantage of this simple policy structure is less significant
today than at the time this procedure was proposed. This is primarily due to improvements in
availability of computing power and software packages for solving LP and NLP problems today.
However, LDR is useful for the insights it offers with respect to solving AOP problems using
mathematical modelling methods.
Master operations scheduling (MOS) represents the critical linkage between planning and
execution of operations. It is the next stage in the operations planning process in any
organization after aggregate operations planning is done. Aggregate operations planning is a
rough-cut capacity planning exercise on the basis of forecast quantities of products and services.
MOS makes use of actual customer orders for the purpose of capacity planning and resource
allocation to specific customer orders. While aggregate operations planning ensures that
adequate capacity is available in a period-by-period basis, organizations need to relate the
capacity needs of specific varieties of products and services that they offer to the overall capacity
available. MOS is the process by which disaggregation of varieties is done. Using this
information, additional planning is carried out to assign the required capacity to each variety. At
this stage, planning for the material required for operations during each period is also done.
Master operations scheduling is required for a variety of reasons:
The most important reason is that at the stage of aggregate operations planning, the forecast demand is normally taken
for the purpose of estimating the capacity required. However, when actual orders are received, this information provides
a better input and another opportunity for the planning exercise.
Moreover, the AOP exercise is to ensure availability of capacity in broad terms. However, as we approach the planning
horizon, it is important to relate material and capacity availability to specific varieties of products and services that an
organization plans to produce during a planning horizon. Therefore, MOS uses actual and the most recent information
while revisiting the planning problem and ensures specific material and capacity availability in a particular time period.
As customer orders get amended and/or cancelled, the information has a bearing on capacity availability. Therefore, as
order inflow happens in real time, the marketing department may want to know if there is scope for accepting newer
order inquiries. The marketing department may also want to know if delivery commitments can be made to customers
while accepting orders. The MOS module serves the important purpose of computing capacity available to promise.
The critical linkages between MOS and other planning processes in an organization are shown in
Figure 15.15. As shown in this figure, MOS takes inputs from the order inflow system and AOP
to perform the disaggregation exercise. Based on the disaggregation, capacity planning and
materials planning are performed to ensure their adequate availability. If there is some
infeasibility in either capacity or material availability, the plans are altered to arrive at feasible
production plans. Therefore, MOS determines what needs to be ultimately produced, not what is
demanded.
Master production scheduling (MOS) is the process by which disaggregation of varieties is done.
There are two stages involved in MOS. The first step is to update the projected demand based
on earlier forecast and current market information. The second step involves disaggregation of
information and relating it to specific material and capacity requirements. After the
disaggregation process, it is possible to accurately estimate the amount of material and capacity
required for planned units of production of each variety. If either capacity or material is not
available, that information is made use of in MOS and plans are re-worked. This is an iterative
process and the actual plan is finally arrived at. The two steps are explained in some detail here.
MOS determines what needs to be ultimately produced and not what is demanded.
SUMMARY
Operations planning serves to translate business plans into operational decisions. The decisions include the amount
of resources (productive capacity and labour hours) to commit, the rate at which to produce, and the inventory to be
carried forward from one period to the next.
AOP is done to match the demand and the available capacity on a period-by-period basis using a set of alternatives
available to an organization to modify the demand and/or the supply.
Alternatives for modifying demand include the reservation of capacity and methods of influencing (changing) the
demand during a period.
Alternatives for modifying the supply include inventory variations, capacity adjustment, and capacity
augmentation.
The AOP exercise employs two generic strategies: chase and level production. A chase strategy is often found to be
expensive and hard to implement in organizations.
In reality, a mixed strategy using a combination of alternatives is employed in an AOP exercise. It uses a variety of
alternatives for modifying supply.
The structure of a transportation model lends itself to studying the AOP problem. Linear programming can also be
used to model the AOP problem.
MOS involves disaggregation of information and ensuring the required capacity and material are available as per
the plan.
REVIEW QUESTIONS
1. Why do you need aggregate units while planning for operations?
2. Give three examples each from the manufacturing and service sectors of industry for aggregation of products and
services.
3. Assume that you are a garment manufacturer. Your current production rate is 20,000 pieces per day. If you want to
increase it to 30,000 per day, what problems do you envisage in making this change?
4. Identify a suitable aggregate unit for each of the following situations:
a. A business school offering four different long-duration academic programmes and three different short-duration
programmes
b. A manufacturer of colour televisions of 30 different varieties
c. A service provider offering inter-city transport facility using six different classes of vehicles
d. A multi-specialty hospital
e. A financial supermarket such as ICICI Bank
f. A fast-food restaurant offering North Indian and South Indian dishes
g. A law firm offering a variety of legal services
h. A manufacturer of custom-made PCBs
i. A BPO service provider offering back-office services to a variety of domestic and MNC clients
5. What do you mean by hierarchical operations planning? Are there any benefits in this exercise? Use an example to
demonstrate hierarchical operations planning.
6. How does aggregate operations planning impact each of the following functional areas in an organization?
a. Marketing
b. Finance
c. Strategy
d. Materials and procurement
7. What considerations will you bear in mind while selecting either demand-oriented alternatives or supply-oriented
alternatives during an AOP exercise?
8. Which AOP alternatives are more suited to a service organization? Why?
9. What are the key differences between level strategy and chase strategy in AOP formulation?
10. Identify suitable AOP alternatives for an organization manufacturing made-to-order items. Will your recommendation
change if the items are made-to-stock?
11. “Mixed strategies are always superior to pure strategies in an AOP exercise.” Comment on this statement.
12. What are the pros and cons of using optimal methods for solving the AOP problem?
13. How is the master production scheduling exercise different from an AOP exercise?
14. How are materials planning and capacity planning related to MOS?
PROBLEMS
1. A manufacturer of ceiling fans offers three different versions of the product to its customers: basic (three blades),
improved (four blades) and deluxe (enhanced performance features). Table 15.9 has information on the machine-hours
required for each of these variations and the number of fans to be manufactured in the next quarter. Assume 25 working
days per month.
2. A manufacturing organization employs 220 people in the factory. The manufacturing is largely manual; therefore, man-
hours are a good measure of the capacity of the system. The factory currently works for a single shift of eight hours. The
number of working days for the next six months is as follows: 24, 22, 25, 19, 21, and 23. The company uses this
information to plan its capacities. The demand for its product during the next six months is such that with the current level
of employment, the company will have 80 per cent capacity utilization. During the next six months, the number of
working days remains the same but the demand is likely to go up by 35 per cent. Under these circumstances, analyse the
situation and answer the following questions:
Evaluate these options from the capacity perspective and suggest a suitable option for the company.
3. Suppose, in Problem 2, the following are the costs pertaining to overtime, inventory and shortage and idle time:
Use this additional information and suggest an appropriate option to the management.
4. Hot Spot is a fast-food restaurant located in a prominent locality in the city of Bangalore. Hot Spot offers three different
menus for breakfast: standard, extended, and custom-made. The standard offering takes nearly 20 minutes of one worker
and the extended requires 30 minutes. On the other hand, custom-made requires three workers to spend 25 minutes each.
Hot Spot is interested in identifying the right number of workers required in its restaurant. Using the data given in Table
15.10, estimate the aggregate capacity required at Hot Spot during the next four weeks.
a. What is the capacity available at Hot Spot in its current level of working?
b. Is the capacity adequate to meet the demand? If not, how much additional capacity do they need?
c. What alternatives would you recommend to Hot Spot to solve their capacity problem?
d. If Hot Spot is ready to recruit new workers, how many should they recruit?
6. A manufacturing firm is performing an AOP exercise for its product. One unit of product requires 100 hours of the
capacity. An extract of the aggregate planning exercise for the firm is given in the table. Table 15.11 has details on the
number of working days, number of people employed and working hours per day. Complete the table and answer the
following questions:
a. What is the capacity required for the planning horizon? Does the company have adequate capacity to meet the
demand?
b. Will the company fall short of capacity if you analyse on a period-by-period basis? By how much does it fall short?
c. If the cost of shortage is ₹5 per hour of capacity falling short, what is the cost that the company may incur on
account of shortages?
d. If the cost of having excess capacity is ₹1 per hour, what will the cost of excess capacity be?
e. If the company chose to adopt this plan, what is the total cost of the plan? Identify the AOP alternatives and the
AOP strategy that the company has deployed in this plan.
7. A manufacturer of garments is interested in making an aggregate production plan for its range of products manufactured.
From the perspective of capacity usage, the range of products could be divided into three: basic garments, office wear and
trendy garments. A basic garment required on an average 2 hours. The ratio of the time required for the office wear and
trendy garment to the basic garment is 2 and 4, respectively. The company can source 20 per cent additional capacity
through overtime. The overtime premium to be paid for the labour is 50 per cent. Material cost is 125 per cent of labour
cost. Currently the company is operating at a capacity of 20,000 hours. The labour charges are paid at the rate of ₹40 per
hour.
The company can also subcontract the work to nearby units for a premium of 20 per cent. However, every time a
subcontracting arrangement is made, it involves a fixed cost of ₹3,000 also. Short term funds for working capital at 24
per cent per annum is available for the company without any restrictions. The company cannot afford to run into
shortages as it is a competitive industry. The forecasted demand for the next six periods is as shown in Table 15.12.
8. A major manufacturer of vacuum cleaners is preparing the AOP for the four quarters of the next year. They are currently
operating at a production level of 24,000 vacuum cleaners per quarter. The number of employees working now is 480.
The current capacity represents the regular time production. The company can additionally produce a maximum of 2,000
vacuum cleaners per quarter through overtime. Subcontracting capacity to the extent of 750 vacuum cleaners per quarter
is also available. The cost of production (including materials) through regular time, overtime, and subcontracting are
₹7,300, ₹8,000 and ₹10,000 respectively. Unused regular time capacity costs ₹800 per vacuum cleaner. The company
plans to start the planning horizon with an opening inventory of 2,700 vacuum cleaners and wants to end with an
inventory of 2,500. No back orders are allowed and the cost of carrying the vacuum cleaners is ₹1,000 per quarter. It costs
₹9,000 to hire a worker and ₹6,000 to lay off. The demand for the planning horizon (next four quarters) is 6,700, 28,000,
49,000, and 30,000 respectively.
The company is considering two options with respect to the workforce, as shown in Table 15.13.
9. Channapatna near Bangalore is well known for specialized wooden toys. A toy manufacturer in Channapatna has
determined the demand (in units) to serve the forecasted demand for the next year. This data is shown in Table 15.14.
Each worker is paid a monthly salary of ₹8,000 and can produce 10 toys per month. The owner does not resort to
overtime, instead, he prefers to hire and fire workers depending on his requirement. He can hire and train a new
employee for ₹8,000 and lay off for ₹2,000. In this way, he enables the local community to acquire some skills and
gainfully employ themselves over time. The cost of holding inventory is ₹125 per month. At present there are 130
employees in the manufacturing unit.
a. Prepare a level production strategy for the toy manufacturer.
b. Can you also develop a chase strategy for the toy manufacturer?
c. Identify two more options for the toy manufacturer.
d. Prepare a comparative analysis of the plans and suggest a suitable course of action for the toy manufacturer.
CASE STUDY
Introduction
New Age Electronics Limited (NAEL) is a Hyderabad-based manufacturer that specializes in
the manufacture of electronic voltage stabilizers. Electronic voltage stabilizers are
increasingly popular on account of better performance, longer working life, light weight, and
low cost. NAEL was promoted by Bala Athreya, an electronics engineer with about 15 years’
experience prior to starting NAEL with two other friends. NAEL manufactures three
variations of voltage stabilizers, primarily differing on account of the type of application they
are put to and the rating. These three can be designated as VS-TV, VS-RF and VS-IN. One
unit of VS-TV requires two hours of labour. For the other two models VS-RF and VS-IN the
labour hours required are 2.5 hours and 3.25 hours, respectively. Due to their early entry into
the market, NAEL has been enjoying a good market share of the electronic voltage
stabilizers.
Production Planning
NAEL is currently in the process of developing an aggregate production plan for the next
financial year. Detailed estimates of the demand were made and based on this a strategic
planning exercise at the corporate level was done to arrive at the potential demand for the
next year. During the strategic planning exercise the marketing department highlighted the
increased competition as more players are bringing similar products to the market. According
to some market information, a number of new players were planning to launch major
marketing campaigns with a view to capture some share of the market. According to NAEL
top management, these new entrants would not make significant inroads into their market
share. The new competitors were most likely to take away the market from traditional voltage
stabilizer manufacturers. After detailed discussions, the final projections for the next year
were finalized. Table 15.15 has the demand projection for the next year.
TABLE 15.15 Demand Projection for the Next Year
Bala and his team analysed the demand pattern for the next year and came up with certain
observations. For instance, the demand for the stabilizers had pronounced seasonality and
peaked during Months 3, 7, and 11. Further, being an electrical product, the cost of carrying
inventory was high, therefore the amount of inventory build-up they could do to meet the
peak demand was a question that they needed to answer. After some discussions, they came
with two possible production plans:
a. To maintain a constant production level of 80,000 hours except during peak periods when it would be increased to
100,000 hours
b. To maintain the production at the level of the monthly demand, but to increase it to the peak production level of
120,000 hours during the peak months in order to meet the demand
They were not clear about which one to choose as there are different implications for each of
these choices. One of the major constraints for NAEL is the ability of their supply chain
partners to respond to the changes in production plan. Up to 5 per cent increase or decrease is
generally accepted without any discussion. Mere e-mail communication will do. Another 5
per cent change (both increase and decrease) requires that NAEL discusses this with them,
have a mutual agreement, and give them at least 45 days notice. Any larger magnitude of
change in excess of 10 per cent is generally resisted. It also requires a much longer time
frame to respond.
Later Development
While Bala and his team were working on finalizing the production plan for the next year,
the marketing director informed them that they would like to launch a promotional campaign
to counteract the new entrants. The specifics of the plans are as follows. The marketing plan
would cost ₹2 million and the budget will be spent during Months 2 to 5. Due to this the
increase in sales (across all the three variations) expected would be Month 3: 10 per cent;
Month 4: 15 per cent, Month 5: 18 per cent, Month 6: 12 per cent and Month 7: 5 per cent.
a. What is your advice to Bala Athreya with respect to production planning for the next year? Which of the two plans
would you recommend? Why?
b. What are the practical difficulties that Bala and his team will face while implementing the plan? How should they
handle them?
c. What are the implications of the marketing plans at NAEL? Bala needs to attend a meeting to discuss this issue.
Prepare a brief for the meeting.
SUGGESTED READINGS
R. L. Armacost, R. L. Penlesky, and S.C. Ross (1990), “Avoiding problems inherent in spreadsheet-based
simulation models—An aggregate planning application,” Production and Inventory Management 31: 62–68.
J. T. Black and J. Donald, “Mount Sinai Hospital Uses Integer Programming to Allocate Operating Room Time,”
Interfaces 32, no. 2 (2001): 63–73.
E. H. Bowman, “Production Planning by the Transportation Method of Linear Programming,” Journal of
Operational Research Society 4 (1956): 100–103.
P. Brandimarte and A. Villa, Modelling Manufacturing Systems: From Aggregate Planning to Real Time Control
(New York: Springer, 1999).
G. P. Cachon, Supply Chain Coordination with Contracts (Philadelphia: The Wharton School of Business,
University of Pennsylvania, 2002).
J. V. Caixeta–Filho, JMv Swaay–Neto, and A. P. Wagemaker, “Optimization of the Production Planning and Trade
of Lily Flowers at Jan de Wit Company,” Interfaces 32, no. 1 (2002): 35–46.
R. E. Crandall, “Production Planning in a Variable Demand Environment,” Production and Inventory Management
Journal 39, no. 4 (1998): 34–41.
P. Gazmuri and S. Maturana, “Developing and Implementing a DSS for CTI,” Interfaces 31, no. 4 (2001): 22–36.
C. C. Holt, H. Modigliani, J. F. Muth, and H. A. Simon, Planning, Production, Inventories and Workforce
(Englewood Cliffs, NJ: Prentice-Hall, 1960).
E. A. Silver, D. F. Pyke, and R. Peterson, Inventory Management and Production Planning and Scheduling (New
York: Wiley, 1998).
T. E. Vollman, W. L. Berry, and D. C. Whybark, Manufacturing Planning and Control Systems, 4th edition (Burr
Ridge, IL: Irwin/McGraw-Hill, 1997).
One of the issues that an operations manager often faces is with respect to the effective use of
resources such as labour, material, machine capacity, and other similar resources. Typically,
in any business situation, the managers face constraints with respect to several of these
resources. For example, in a manufacturing organization the availability of machines is
limited. Therefore, an operations manager would like to understand how best to make use of
the available machines while meeting certain objectives such as maximizing the production
or minimizing the waiting time for the machine. Similarly, in a service system such as a
hospital or restaurant, one of the issues for planning would be to make the best use of
available resources while minimizing the waiting time of the customers in the system. Linear
programming is a methodology which helps an operations manager to address such problems
that he/she faces in planning the operations in the system.
Linear programming (LP), as the name implies, is a methodology in which linear
relationships between various elements in the system can be modelled to study the behaviour
of the system and to identify the best set of decisions that one should make in order to
optimize the performance of the system while taking into consideration the constraints that
one faces in the system.
LP Modelling Fundamentals
An LP modelling exercise begins with the mathematical formulation of the problem that one
wishes to study. Before formulating the problem, it is important to understand the decisions
to be made, the relationships between various elements of the system, and the performance
measures for which the decision is being made. In an LP modelling exercise, these aspects
are represented in a structured fashion. The following terminologies describe them in some
detail:
Decision variables: The purpose of LP modelling is to arrive at a decision with respect to the problem under study.
For example, the problem could be to identify the number of machines required in order to meet a certain targeted
production in a manufacturing system. In this case, the number of machines required becomes the decision
variable. Another example is to decide on an appropriate production plan given a set of machines and other
resources so that the profit could be maximized. Similarly, in the design of a restaurant, the number of waiters to be
employed could be a decision variable.
If the decision on hand is the number of machines of Type A, Type B, and Type C to be procured, one can
denote this decision using three decision variables as illustrated below:
Let XA be the number of machine of Type A to be procured
Let XB be the number of machine of Type B to be procured
Let XC be the number of machine of Type C to be procured
Objectives: The operations manager analyses the problem with specific objectives in his/her mind. The objective
could be to maximize certain parameters of interest such as production, profit, availability of resources etc.
Alternatively, it could be to minimize certain parameters such as investment, waiting time, cost etc. In an LP
modelling exercise, the objectives need to be specified a priori. In the preceding example of the manufacturing
system, our objective could be to minimize the investment to meet a certain targeted production or to maximize the
production given a certain availability of machines. Similarly, in the case of the restaurant, the objective could be to
maximize the daily revenue or to minimize their cost of human resources while meeting a certain demand.
For example, in the case of the manufacturing system, if the objective is to minimize the investment, then one
can represent the objective function as follows:
Let CA be the cost of one unit of machine of type A to be procured
Let CB be the cost of one unit of machine of type B to be procured
Let CC the cost of one unit of machine of type C to be procured
The cost of the decision Z = XA × CA + XB × CB + XC × CC
The objective function is Minimize Z.
Constraints: Any management problem is solved in the context of certain constraints that the manager faces. If we
consider the examples discussed earlier, in the case of the manufacturing system, the constraint could be the need
to meet with a certain daily production. Another constraint could be the availability of raw materials for production.
A third constraint could be the demand for the product for which the production system is being designed. A fourth
could be the funds available for the procurement of machines. Similarly, in the case of the restaurant, the
constraints could be the demand for the service, the productive capacity of the kitchen, the skill requirements for
the service, and the availability of such skilled personnel in the market.
Using LP as a modelling tool also implies certain important assumptions behind the modelling exercise. These
include the following:
a. Linearity: All relationships between various elements that are being studied using an LP model are
assumed to be linear. Similarly, the objective function is also assumed to be linear.
b. Non-negativity: It is also assumed that all decision variables used in an LP modelling exercise are strictly
nonnegative. In other words in the above example this implies that XA, XB, and XC are greater than or
equal to zero.
Once these conditions are satisfied, there are well-known methodologies to obtain a solution
set for any LP problem. The solution methodologies also indicate whether a feasible solution
exists for the problem formulated, and if so, whether it is unique. The solution methodologies
also enable us to analyse the sensitivity of variables and constraints of the problem.
EXAMPLE 15A
Solution
Decision Variables for the Problem
Let us begin by identifying the decision variables for this problem. Since the problem is one
of determining the daily production schedule, the decision variables are the number of units
of Product A and Product B to be produced daily. Let us denote this with two decision
variables as shown below:
Let X1 denote the number of units of Product A to be produced daily
Let X2 denote the number of units of Product B to be produced daily
The manufacturer works on a single-shift basis of eight hours. Therefore, each machine is
available for 480 minutes per day. Since four units of Machine A are available with the
manufacturer, the daily capacity available with respect to Machine A is 1920 minutes (4 ×
480). On the other hand, one unit of Product A requires 20 minutes in Machine 1. Similarly,
one unit of Product B requires 17 minutes in Machine 1. Using this information, one can
develop the constraint equation with respect to machine A as follows:
The constraints with respect to the other machines are written in a similar fashion.
Applications of LP
Linear programming (as we have seen above) is a generic solution methodology. Therefore a
wide range of applications is possible using LP. This includes addressing several operations
management problems, advertising and media selection, inventory control, capital budgeting,
and a host of other situations. Specifically, in operations management, LP is used often for
the following problems:
Production planning: One of the frequent applications of LP is in the area of production planning. Typically,
aggregate production plans and monthly/weekly production planning could be done using LP. While the production
plan is done several constraints need to be taken into consideration including the minimum demand to be met, the
availability of machine capacity, and the availability of raw materials. This can be modelled using LP.
EXAMPLE 15B
A consumer-goods manufacturing company producing three final products needs to decide
on the capacity investment with respect to a new automated packaging/assembly line that it is
designing. There are three types of packaging machines available and each one of them can
pack the three products, albeit in varying capacities per hour. The shop works for 8 hours a
day. While certain minimum demand requirements are to be met, the exact amount of
packaging capacity to be invested needs to be finalized. Any additional production could be
sold in the market as there is adequate demand for the products. The relevant details for the
problem are summarized in Table 15B:
Formulate the problem and obtain an optimal investment plan.
Solution
Mathematical formulation of the problem
Let X1 be the number of machines of type 1 to be procured
Let X2 be the number of machines of type 2 to be procured
Let X3 be the number of machines of type 3 to be procured
Since this is an investment decision, our objective will be to minimize the investment
subject to meeting the minimum daily demand of the products. The packaging capacity of the
machines given in the table is per hour. We need to multiply them by 8 to get the daily
packaging capacity. Based on these assumptions, the complete mathematical formulation of
the problem is as follows:
Total investment cost Z = 1.65 × X1 + 2.12 × X2 + 1.60 × X3
Min (Z)
Subject to:
One can use Microsoft Excel to solve this problem as explained in the previous example.
Figures 15E, 15F, and 15G show the relevant Solver details and the final solution.
The optimal solution is as follows:
Number of Machine 1 to be procured: 2.62
Number of Machine 3 to be procured: 1.08
Total investment cost: ₹6.05 million
Consequently the daily packaging capacities are:
Product A: 1202.49 units
Product B: 1700 units
Product C: 900 units
TABLE 15B Data for Example 15B
Capacity planning: As we have seen in one of the numerical examples above LP could be used to address the issue
of investment decisions pertaining to capacity additions. Further, based on certain capacity constraints, several
operational planning decisions could be modelled using LP.
Supply Chain Management: Logistics and distribution problems, inventory planning, and materials management
problems are amenable for LP modelling. Typically, transportation and assignment models are used to study
logistics and distribution problems. Cutting-stock algorithms are frequently used to optimally make use of materials
in several industries such as textiles, steel, and paper. In addition, lot sizing problems, economic order quantities
and a host of other inventory planning problems could be studied using and LP.
SUMMARY
Typically, in any business situation, managers face constraints with respect to several resources.
Understanding how to make the best use of the available resources while meeting certain objectives such as
maximizing production is very important. Linear programming is a methodology which helps an operations
manager to address the problems that he/she faces in planning the operations in the system.
Linear programming (LP), as the name implies, is a methodology in which linear relationships between various
elements in the system can be modelled to study the behaviour of the system and to identify the best set of
decisions that one should make in order to optimize the performance of the system while taking into consideration
the constraints that one faces in the system.
Before formulating the problem, it is important to understand the decisions to be made (decision variables), the
constraints, and the performance measures for which the decision is being made (objective function).
Using LP as a modelling tool also implies the linearity of all relationships between various elements that are being
studied, as well as the non-negativity of the decision variables used in the modelling exercise.
A number of computer packages are available today to solve a well-formulated LP problem. Therefore, developing
adeptness in formulating the problem will be more valuable than developing the skills used in actually solving the
problem using a known solution methodology.
Linear programming could be used to model several operations: management problems, advertising and media
selection, inventory control, capital budgeting, and a host of other situations.
REVIEW QUESTIONS
1. What are the conditions under which one can model a problem using LP?
2. Suppose there are “n” decision variables to be studied and there are “m” constraint equations. Can you write a generic
mathematical formulation for such a problem using some standard notations?
3. Identify three advantages of using the Microsoft Excel solver for solving LP problems.
4. Suppose you successfully solved a problem using the Microsoft Excel server. How will you use the output obtained from
the solver?
PROBLEMS
1. A large manufacture of petrochemicals in the country wants to address the issue of emission of pollutants into the
atmosphere. The company is planning to install anti-pollution devices. The company makes two products HC1 and HC2;
in each of these products, the manufacturing process yields excessive amounts of irritant gases. Table 15C shows the daily
emission of each pollutant for every 1000 litres of product manufactured. The company is prohibited from using W1, W2
and W3 kg of sulphur dioxide (SD), carbon monoxide (CM) and nitric oxide (NO). The profit for each 1000 litres of
Products 1 and 2 manufactured is P1 and P2.
The production manager has approved the installation of two anti-pollution devices. The first removes 75% of the
gas CM, 50% of the gas SD, and 90% of the gas NO, regardless of the product manufactured. The second device
removes 33% of the gas CM, 10% of the gas SD and the 60% of Gas NO for Product 2. The variable cost of the first
device per 1000 litres manufactured is C1 regardless of the product; similarly the variable cost of the second device is
C2. Sales commitments dictate that at least Q1 thousand litres of HC1 and Q2 thousand litres of HC2 be manufactured.
Formulate an appropriate LP model.
2. A Logistics service provider needs to augment its fleet and has a budgetary allocation of ₹40 million. It is contemplating
three types of vehicles. Vehicle A has a 10-ton pay load and is expected to average 25 kmph. It costs ₹1.6 million.
Vehicle B has a 20-ton pay load and is expected to average 15 kmph. It costs ₹2.6 million. Vehicle C is a modified
version of Vehicle B; it carries sleeping quarters for one driver, and this reduces the capacity to 18 tons and raises the cost
to ₹3.0 million. Vehicle A requires a crew of one man, and if driven on three shifts per day, could be run for an average of
18 hrs. Vehicles B and C require a crew of two men each, but whereas B would be driven 18 hrs per day with three shifts,
C could average 21 hrs per day. The Company has 150 drivers available per day and would find it difficult to obtain
further crews. Maintenance facilities are such that the total number of vehicles must not exceed 30. Formulate this
problem in order to find out the number of vehicles to be purchased to maximize the capacity in ton-km per day.
3. Formulate the following problem as an LP and solve using Solver.
A firm produces an alloy that is made from steel and scrap metal. The cost per ton of steel is ₹1050 and that of scrap
is ₹820. The technological requirements for the alloy are: (i) a minimum of one ton of steel is required for every two
tons of scrap, (ii) one hour of processing time is required for each ton of steel and four hours of processing time for
each ton of scrap, and (iii) the steel and scrap combine linearly to make the alloy. The process loss from steel is 10%
and the loss from scrap is 20%. Although production may exceed demand, a minimum of 40 tons of the alloy must be
manufactured. To maintain efficient plant operation, a minimum of 80 hours of processing time must be used. The
supply of both scrap and steel is adequate for the production of the alloy. The objective is to produce the alloy at a
minimum cost.
4. A certain farming organization operates three farms of comparable productivity. The output of each farm is limited by the
usable area and by the amount of water available for irrigation. The data for the upcoming seasons are given in Table
15D.
The organization is considering three crops for planting which differ primarily in their expected profit per hectare and
in their consumption of water. Furthermore, the total area that can be devoted to each of the crops is limited by the
amount of appropriate harvesting equipment available. Their acreage, water consumption, and expected profit are given
in Table 15E.
It is the policy of the organization that the percentage of useable area planted must be the same at each farm. However,
any combination of the crops may be grown at any of the farms. The organization wishes to know how much of each
crop should be planted at the respective farms in order to maximize the expected profit.
What is the optimal cropping pattern to be adopted by the organization?
SUGGESTED READINGS
A. Ravindran, D. T. Philips, and S. Solberg, Operations Research: Principles & Practice,, 2nd edition (n.p.: Wiley
India Pvt. Ltd., 2007).
G. Srinivasan, Operations Research: Principles & Applications (New Delhi: Prentice Hall of India Pvt. Ltd, 2007).
H.A. Taha, Operations Research: An Introduction, 8th edition (New Delhi: Prentice Hall Inc, 2008).
CHAPTER 16
Resources Planning
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Why is it important to adopt a different planning methodology for inventory control of dependent demand items?
What are the basic building blocks of the materials planning methodology?
What are the steps involved in the materials requirement planning (MRP) methodology?
How is capacity planning done in organizations and what is its relationship with MRP?
What is the relationship between MRP and modern-day enterprise resources planning (ERP) systems?
Two types of inventories exist in an operating system—operating inventory and distribution inventory. Operating inventory
includes all the resources that are available for the operating system to consume in the production process. Distribution
inventories are meant for market consumption. The number of hotel rooms to be made available is an example of distribution
inventories.
Source: stock.xchng
ideas at Work 16.1
Material requirements planning (MRP) is a computerized information system that aids the planning of
materials in manufacturing organizations.
Before we discuss the issues in resources planning, it will be useful for us to understand the
key differences between the two types of inventories that exist in an operations system. The first
is called operating inventory. This includes all the resources (broadly of material and capacity)
that are available for the operating system to consume in the production process. Consider an
automobile manufacturer such as Hyundai. The inventory of tyres available in the stores is meant
for satisfying the production requirements. If Hyundai has 4,000 tyres suitable for its Santro
model at the beginning of the shift, it can assemble 1,000 Santros during the shift (we do not
count the extra tyre provided as standby in this computation). Similarly, if it takes 1.5 hours to
paint an assembled car at the paint shop, the capacity of paint shop required for 1,000 Santros is
1,500 hours. There is no statistical estimation involved in arriving at the number of Santros that
Hyundai can assemble on that day or the capacity required at the paint shop.
Operating inventory includes all the resources (broadly of material and capacity) that are available for the
operating system to consume in the production process.
On the other hand, operations systems carry another type of inventory known as distribution
inventory. Distribution inventories are meant for market consumption. In our example, finished
Santros waiting for dispatch belong to the distribution inventory. If the distribution inventory of
Santros is 1,000, we cannot come to the conclusion that 1,000 Santros will indeed be sold. It
depends on the demand in the open market and may be subject to statistical fluctuations. Unlike
the operating inventory, it requires alternative estimation techniques to approximately guess the
number of Santros likely to be demanded on that day.
One can give similar examples in the service system as well. Consider a hotel in the Andheri
area of Mumbai. The number of rooms to be made available and the number of people to be
served lunch and dinner are examples of distribution inventories. On the other hand, the amount
of raw materials made available for use in the kitchen and the number of waiters and
housekeeping staff deployed in the hotel on that day are examples of operating inventories.
Look at the converse of the problem we just discussed. How do we decide how much
inventory to carry? In the case of operating inventories, it is a matter of using arithmetic to arrive
at this decision, whereas in the case of distribution inventories, the decision is a little more
complex. In other words, while it is difficult to decide how many Santros to assemble on a
particular day, it is very easy to compute the exact number of tyres required on that day once we
have decided on the assembly schedule. In this chapter, we focus our attention on operating
inventories and developing a methodology to help managers plan for these inventories. Planning
for intermediate components and raw material/bought-out items using this methodology is
known as materials requirement planning. However, one can extend this idea of planning to
capacity and other resources in an operating system. Therefore, we broadly define this
methodology as resources planning. We will discuss the planning issues pertaining to
distribution inventories in the next chapter.
Why should a manager address planning issues associated with operating inventories differently
from those associated with distribution inventories? In order to answer this question, we need to
know more about the demand attributes of these inventories. As we have already discussed,
operating inventories have dependent demand. Distribution inventories, on the other hand, have
independent demand. All other differences stem from this attribute of demand.
The foremost difference stems from the nature of the demand. There is no uncertainty in a
dependent demand item. Moreover, the demand is lumpy, manifesting at specific points in time
in response to a requirement in the system. Consider the domestic fan manufacturer, Crompton
Greaves Limited. If Crompton Greaves decides to produce 20,000 fans in May 2014, the number
of blades that it requires is 60,000 (since each assembled fan is fitted with three blades). Going
one step further, if each blade consumes two hours of machine time, 350 g of a grade of steel,
and 50 ml of paint, then it requires 120,000 machine-hours, 21,000 kg of steel, and 3,000 litres of
paint if it were to manufacture the 60,000 blades during the month. In this example, we notice
that a parent–child relationship between inventories causes the dependency. The same cannot be
said in the case of independent demand items.
Due to certainty, the goal for a planning manager is to make the dependent demand items
available to exactly match the requirement. It is feasible to attain a 100 per cent service level in
this case. On the other hand, due to uncertainty, independent demand items cannot be made
available at a 100 per cent service level. The extra inventory required increases substantially as
we approach 100 per cent service level in this case. Therefore, managers plan for a certain
targeted service level in the case of independent demand items.
Since dependent demand items exhibit some structure and have a causal relationship with
other items in the system, it is possible to estimate the demand by appropriate planning
methodologies. Moreover, as our example demonstrated, it is possible to accurately estimate how
much to order. These issues are far more complex in the case of independent demand items.
However, in the case of dependent demand items, the timing of the order is very crucial. This is
because the dependency relationship among the items requires that items arrive exactly at the
required time. Otherwise, shortage and schedule slippages could cascade through the system and
create severe schedule disruptions. Table 16.1 lists the key differences between dependent
demand and independent demand items.
Since dependent demand items exhibit some structure and have a causal relationship with other items in the
system, it is possible to estimate the demand through simple planning methodologies.
TABLE 16.1 Key Differences Between Dependent Demand and Independent Demand Items
Before we unfold the planning methodology for dependent demand items, let us look at the
following aspects of the planning framework: (a) multiple levels of dependency, (b) the product
structure–the bill of materials (BOM), (c) time-phasing the requirement, (d) determining the lot
size, (e) incorporating lead-time information, and (f) establishing the planning premises.
Multiple Levels in Products
Clearly, a parent–child relationship in dependent demand items causes dependency. In reality,
multiple levels of relationships exist in a product. While computing the requirement for an item,
it is important that we proceed level by level. Otherwise, the amount of inventory that we make
available will be different from what is required. Let us study the example of a basic telephone
instrument to illustrate this idea. As shown in Figure 16.1, the telephone instrument consists of a
base unit and a handset. The handset has a speaker and a microphone and is covered by a pair of
cover plates firmly held together using a pair of screws. There is a wire that connects the handset
to the base unit. In this simple example, let us consider the parent–child relationship.
Let us assume that we have in stock an inventory of 30 telephones, 27 handsets and 16 cover
plates. Suppose that we want to plan for the production of 100 telephones during the next week.
How much inventory do we need in each of these components? One method is to merely subtract
the on-hand inventory from the requirement at each level to arrive at this number (see Method A
in Figure 16.2). Another method is to proceed level by level in the computation. In this method,
after computing the requirement at one level, we use the parent–child relationship to compute the
requirement at the next level (see Method B in the Figure 16.2).
FIGURE 16.1 A telephone instrument
This iterative process of computing all requirements at a particular level and then moving down one level is
known as an explosion in resources planning literature.
As we see in Figure 16.2, both the methods result in different numbers. However, the second
calculation truly depicts the requirement. Let us understand this using another kind of logic.
Every item in an operating inventory exists on its own and as part of its parent. For instance, in
addition to 16 units of on-hand inventory of cover plates, 54 more exist in the assembly of 27
handsets and another 60 in the assembly of 30 basic telephones. Therefore, the total inventory
already on hand is 130 (60 + 54 + 16 = 130). Since 100 telephones are required, only 70 more
cover plates are needed.
FIGURE 16.2 Methods for calculating inventory requirement
Therefore, it is clear that while planning for operating inventories, we need to proceed on a
level-by-level basis. This iterative process of computing all requirements at a particular level and
then moving down one level is known as an explosion in resources planning literature.
Product Structure
As it is important to compute the requirement level by level, an unambiguous definition of the
levels is crucial. For this, knowledge of the product structure is important. In any product, the
final assembly stage puts several major assemblies together. In order to understand this, let us
look at a typical fountain pen. Figure 16.3 shows an exploded view of a fountain pen. The
various components that go into one fountain pen, their exact specifications, and the quantity
required for assembling every fountain pen can be arrived at from Figure 16.3.
Let us return to our example of the basic telephone instrument. If we disassemble the
telephone instrument progressively, we will be able to understand the product structure. In our
example, two assemblies and two components make up the final product: the base unit assembly,
the handset assembly, a connecting cable, and a pair of jacks to connect the ends of the cable
with the base unit and the handset.
At the next level, we can identify the sub-assemblies that make up each assembly. The base
unit assembly consists of the operating unit (consisting of the basic circuitry and processor), the
panel board (consisting of the buttons to dial the numbers), the control panel (consisting of
controls for volume, pulse/tone settings etc.), a pair of cover plates, four lugs (to support the
instrument on any surface), and four screws to cover the entire unit using the base plates form the
next level. One can similarly identify all the components and proceed downwards until we reach
the basic raw material/bought-out components. The resulting graphical structure is the product
structure. Figure 16.4 shows the product structure for the basic telephone unit. When common
items appear in more than one level, they must be represented at the lowest level of their
appearance. This is known as low level coding. In our example, the item “screw” appears at
Level 2 and Level 4, but we have represented them at Level 4. Low level coding does not alter
the logic behind planning for the requirement but merely serves to improve the efficiency of
processing.
The product structure graphically depicts the dependency relationships among various items that make up the
final product.
The product structure graphically depicts the dependency relationships among various items
that make up the final product. Every level in the product structure has a parent relationship with
those below it. In Figure 16.4, the numbers in parentheses show the number of items at a
particular level needed to assemble one unit of its parent. Thus, 12 buttons and 12 screws are
required to make one button assembly. Similarly, one speaker, one microphone, two cover plates,
and two screws make up one handset.
FIGURE 16.3 The product structure of a typical fountain pen
Solution
a. From a description of the product and its sub-assemblies, we note that Item D occurs at two different levels.
Therefore, we employ low-level coding for Item D. The product structure is shown in Figure 16.5. The product has
three levels.
b. If we assume that no inventory of sub-assemblies or the end product exists in the system, then the number of D’s
required is computed as follows:
Number of C’s required for assembling 10 units of A: 40
Number of D’s required for assembling 40 units of C: 160
Number of D’s required for assembling 10 units of A: 20
Therefore, the total number of D’s required for manufacturing A = 180
c. Since an inventory of the intermediate sub-assemblies exists, it is important to compute the number of D’s required
using a level-by-level approach.
Number of C’s required for assembling 10 units of A: 40
Number of C’s already available: 10
Net requirement of C’s: 30
Number of D’s required for assembling 30 units of C: 120
Number of D’s required for assembling 10 units of A: 20
Total requirements of D’s: 140
Number of D’s already available: 10
Net requirement of D’s for assembling 10 units of A: 130
The bill of materials is a list of all the parts, ingredients, or materials required to assemble or put together one
unit of a product.
A variety of formats are available for BOMs. The simplest format is the single-level BOM. It
consists of a list of all components that are directly used in a parent item. An indented BOM is a
form of multi-level BOM. It exhibits the final product as Level 0 and all its components as Level
1. The level numbers increase as you proceed down the tree structure. If an item is used in more
than one parent within a given product structure, it appears more than once, under every sub-
assembly in which it is used. A third variation is the modular BOM. Modular BOMs are very
useful to represent product structures with several varieties.
In the telephone example, let us assume that four different colours (grey, blue, black, and
beige) are available. Further, let us assume that three different control panels (simple, deluxe,
and elegant) and four different memory settings (20-call memory, 50-call memory, 100-call
memory and 250-call memory) are available. This means that we can make 48 (4 × 3 × 4) unique
offerings to the customer. Therefore, we need 48 different BOMs to represent each variant. One
way to solve this problem is to have a modular BOM. By picking up one variation for each of the
three attributes, it is possible to construct one unique BOM for the purpose of planning. In
several other cases such as automobiles and IT hardware, the number of unique offerings could
be as high as 250,000 considering the numerous options available for each variation. In such
situations, it is prudent to use a modular BOM. Table 16.2(a), (b), and (c) illustrate the
alternative BOMs for our telephone example.
TABLE 16.2(a) A Single-Level BOM for a Telephone Instrument
Therefore, we may conclude that we need to place an order for 120 base cover plates to meet
the planned requirements. While this computation is correct, it may not reveal “when” to place
the order and if there are any problems that the organization may face by planning in this
manner. Let us look at the same data in a different fashion in Table 16.3:
TABLE 16.3 Time Phased Estimation of Net Requirement
Notes: *Increase in on-hand inventory by 300 at the end of Week 4 (beginning of Week 5) is due to an order placed earlier;
**On-hand data pertains to the inventory at the end of the period.
Table 16.3 reveals important information to the planner arising out of the lumpy nature of
demand that is characteristic of dependent demand items. First, we find that the on-hand
inventory, due to arrive at the beginning of Week 4 (on account of an earlier order) is “out of
phase” with the requirement. Although the order can cover the requirement for the first five
weeks, it is wrongly scheduled. Second, it is also clear from the computation that our order of
120 needs to arrive at the manufacturing system only at the beginning of Week 8. There is no
need to order much in advance and keep the base cover plates idle in the stores.
This illustration highlights another important principle associated with resources planning.
“Time phasing” of the information is a critical step in the process. In this case, we used a time
bucket of “weeks”. The planner needs to choose an appropriate time bucket for the system for
which planning is done and analyse the inventory-related information. Time bucketing has
significant implications for the computational efficiency and the accuracy of information. In the
example being discussed, by aggregating all the data to a two-month bucket, we first obtained
gross information by using just four data points (on hand, on order, gross requirement, and net
requirement). Later, we obtained better information by using a weekly bucket involving 25 data
points (one data point for on-hand inventory at the end of time 0 and 24 data points for the eight
weeks, as can be seen in Table 16.3). This is a managerial trade-off and needs to be resolved
before using the system for planning purposes.
It is also important to realize that the accuracy and integrity of the inventory data is crucial for
obtaining good schedules for the components. Organizations incorporate periodic checking
routines (called cycle counting) to ensure that the integrity of the data is established. Physical
inventory is verified with the inventory data and corrective measures are taken if there are any
discrepancies.
It is evident that computing the net requirements and appropriately time phasing the
requirement is the first step in the planning process for dependent demand items. A sample
illustration for the operating unit of the telephone example is given in Table 16.4.
TABLE 16.4 A Sample Illustration for the Operating Unit
Note: *On-hand data pertains to the inventory at the end of the period
Lot sizing is the process of determining the size of the order quantities for each component in a product.
Lot-for-lot (LFL)
Lot-for-lot is perhaps the simplest of the lot-sizing rules. Under this rule, the lot size is equal to
the net requirement during every period in the planning horizon. If, for a component, the net
requirements for the next 4 weeks are computed to be 120, 0, 200, and 50, then the order sizes
scheduled to arrive during these four weeks are 120, 0, 200, and 50, respectively. Table 16.5
shows a sample illustration of the rule for the operating unit of the phone.
The LFL rule is appropriate under the following conditions:
a. The cost of carrying inventory from one period to the next is very high compared to the cost of set-up.
b. The demand for the item is sparse and highly discontinuous (several data points in the planning horizon have zero
demand).
One may find that in situations with complex product configurations and several varieties in
the end product, these assumptions hold as we approach the final assembly stage.
Note: *On-hand data pertains to the inventory at the end of the period.
TABLE 16.6 The FOQ Rule for the Operating Unit of a Telephone
Note: *On-hand data pertains to the inventory at the end of the period
Alternative methods could be used for computing the fixed order quantity. One method is to
estimate the economic order quantity for the item on the basis of set-up/ordering costs and
carrying cost.1 In the case of components manufactured “in-house”, there are preferred run
lengths or economic run lengths. Similarly, for items bought from outside, minimum order
quantity restrictions imposed by the suppliers, other considerations such as truck load capacity,
and the prevailing quantity discounts structure will influence the choice of order quantity.
FOQ provides an alternative perspective to the “carrying cost–ordering cost” trade-off. When
the unit value of the item is low and there is a continuous and more or less stable demand for the
item, it is appropriate to use this rule. Items that are at the lower ends of a product structure share
a greater degree of commonality among numerous end-product variations. Therefore, the
demand is likely to be continuous, justifying the choice of FOQ rule. Moreover, they tend to
have a low unit cost. Hence, the cost of carrying any leftover inventory from one period to the
next is likely to be very low. In reality, depending on the net requirements, the order placed will
be in integral numbers of FOQ (i.e., k × FOQ, where k is an integer).
Let us return to the example of operating units. The total net requirement for the 8-week
period is 560 (150 + 120 + 130 + 160 = 560). Let us suppose that the economic order quantity is
210. Therefore, using Eq. 16.1, we obtain P to be 3 periods. Table 16.7 shows the
implementation of the POQ rule for operating units with P = 3. An order is scheduled to arrive at
the beginning of Week 3 and the quantity ordered is sufficient to cover the net requirements
during Weeks 3 to 5. Similarly, another order is scheduled to arrive at the beginning of Week 7.
TABLE 16.7 Implementation of the POQ Rule for Operating Units with P = 3
Note: *On-hand data pertains to the inventory at the end of the period
FIGURE 16.7 Computation for planned receipt and planned order release
Note: *On-hand data pertains to the inventory at the end of the period
Let us assume that based on the on-hand inventory of basic telephones, base unit assemblies
and operating units as well as the scheduled receipts of these during the planning period, the
planned order releases for base unit assemblies are computed to be 140 for Week 3, 80 for Week
5, 130 for Week 7, and 120 for Week 8. Let us further assume that the MPS has projected the
demand for operating units as spares to be 60 units in Week 3, 40 units in Week 5, and 40 units
in Week 8. In general, we can compute the gross requirements of components for a period t at
any level using the following notations:
The planned order release of the parent = PORt
In our telephone example, each base unit assembly requires one operating unit. Therefore one
can use the given information and Equation 16.3 to compute the gross requirements of operating
unit for the planning period. Figure 16.8 explains the computation of gross requirements in the
case of the operating unit of the phone.
FIGURE 16.8 Computation of gross requirements
Material requirements planning is a structured approach that develops schedules for launching
orders for materials in any manufacturing system and ensures the availability of these at the right
time and at the right place. It uses the basic building blocks of resources planning that we
described in the previous section to develop these schedules. Figure 16.9 shows the core logic of
the MRP process, its input, and output.
As shown in the figure, four key processes drive the MRP procedure. These processes occur in
a cyclic fashion. The first process is the “net” process. The MPS for the end product provides
information on the gross requirements for the end product. By utilizing the information available
in the inventory records, the “net” process computes the net requirements for the end product.
The second process is the “lot” process. Once the net requirements are computed, the lot-sizing
rule is used to schedule planned receipts of the product. The third process is the “offset” process.
Once the planned receipts are identified, lead-time information is used to offset and obtain the
planned order releases for the product. The planned order releases are either work orders for a
manufacturing shop to assemble as many components as per the schedule or a purchase order to
obtain sub-assemblies from outside.
Once these three processes are completed, the requirements for the end products are estimated
and orders are scheduled. Then the next step is to cascade the process down the product structure
and repeat the procedure with all the components at the next level in the product structure. This
process is the last in the cycle denoted as “explode”. In order to perform the explosion process,
BOM data is required. The planned order releases of a parent creates dependent demand for the
offspring as specified in the BOM. This becomes the gross requirements for the offspring. The
procedure continues iteratively, level by level, until the lowest level is reached and all
component schedules are determined.
Therefore, the key inputs for the MRP processes are MPS, BOM, inventory status, lead-time
data, and lot-sizing rule. We have already discussed in detail in the earlier sections the role each
one of them plays in accurately determining the quantity and timing of the material requirements.
As we proceed through the lower level components, two types of outputs are generated from the
MRP system. The first output is a work order. Work orders are generated for items that are
manufactured in-house. In our telephone example, we would have generated work orders for
base set and handset assemblies as they are likely to be done in-house. The second output is a
procurement notice. Procurement notices are generated for items that are bought from outside
and directly used in the assembly. It triggers the purchase ordering process in an organization. In
our telephone example, it is possible that the manufacturer might directly source connecting
cables and jacks instead of producing them in-house. In such a situation, the outcome of the
MRP process will be a procurement notice.
FIGURE 16.9 Material requirements planning core logic
EXAMPLE 16.2
A manufacturing organization needs to plan the materials required for the next six weeks for
the manufacture of its end-product (Product A), as per a master production schedule. In
addition to the end product, there is an independent requirement of Component C, as it is sold
as a spare in the market. The master production schedule for both the end product and the
spares are given in Table 16.9. In order to assemble one unit of Product A, Components B to
G are required. Figure 16.10 shows the product structure. In the figure, the number alongside
each component denotes the number of each component required to assemble its immediate
parent. An extract of the inventory status (Table 16.10) reveals the inventory on hand. There
are no pending orders for delivery. Different lot-sizing rules are used for the components of
Product A. Moreover, the components have different lead times. All this information is
available in Table 16.10. Perform an MRP exercise to estimate the quantity and timing of the
components required for the manufacture of Product A as per the MPS.
Solution
We perform the MRP exercise by using the four-step process given in Figure 16.8. We begin
with Product A. Figure 16.11 shows the calculation of the net requirements. Since the lot-
sizing rule is lot-for-lot, the net requirements and the planned receipts are the same. We
offset the planned receipt by lead time to obtain the planned order releases. Figure 16.11
shows the calculations.
Now, we continue the process by exploding the product structure and moving to the next
level. There are two components at this level: Components B and C. In our example, if an
order needs to be released for assembling 100 units of Product A in Week 1, then we need
300 units of Component B and 100 units of Component C at the beginning of the week so
that we can launch the work order. Therefore, the planned order releases of a parent
determine the gross requirements for the offspring. We repeat the process once for B and
then for C before we explode to the next level. The two tables in Figure 16.12 show the
calculations for Components B and C.
While computing the gross requirements for C, we take into consideration both the
dependent demand of C (as indicated in the product structure) as well as the independent
demand (as indicated in the MPS in Table 16.9). Consider Period 2. The planned order
releases for Product A during this period is 200. Therefore, there is a gross requirement of
200 units of C. There is also an independent demand of 50 units of C during Period 2.
Therefore, the gross requirement for Component C is 250 (200 + 50 = 250). Similar
computations have been made for all other periods also.
FIGURE 16.10 The product structure
TABLE 16.9 MPS for the Next Six Periods
Component D is used by both B and C. Therefore, while arriving at the gross requirements
of D, we take into consideration the planned order releases of both the parents. For example,
during Period 1, the planned order releases of Components B and C are 200 and 50,
respectively. Since they both require one component of D, the gross requirement for D during
Period 1 is 250. Since the first instance of planned receipt is Week 3, we add up the
requirements of three weeks (Weeks 3 to 5) in order to implement the POQ policy and
schedule a planned order release during the beginning of Week 1.
We perform similar computations for the remaining items in the product structure. Figure
16.13 presents the detailed calculations for Components D and E, while Figure 16.14 has the
calculations for Components F and G.
FIGURE 16.11 Calculations for Product A
FIGURE 16.12 Calculations for components B and C
Perhaps the most significant impact that a well-designed MRP system could provide to an
organization is the reduction in inventory. MRP systems were first developed in the early 1960s
and organizations that started using them reported dramatic reductions in their inventory. The
reasons are obviously related to the logic of exploiting the peculiar characteristics of dependent
demand items. Using traditional EOQ-based inventory control systems will often result in having
inventory when not required. The other advantage of the MRP system is the increased visibility
of items and their dependencies through a BOM representation of products being manufactured.
Further, it could potentially inculcate a certain discipline in the planning process.
Despite their simplicity and their initial success, MRP installations faced several problems after implementation.
Despite the simplicity and initial success, MRP installations faced several problems after
implementation. In several cases, MRP systems suffer from three major problems:
The data integrity is low. The quality of the solution is only as good as the data used for the computation. If the lead-
time data is wrong, organizations may either have too much inventory or frequent shortages. Similarly, if the inventory
status is wrong it could jeopardize the entire computation.
Users did not have the discipline of updating the required databases as and when changes were taking place elsewhere in
the organization. If the R&D department creates new designs and revisions in existing product design, this data needs to
be incorporated in the BOM file. Failure to do so will mean introducing errors in the process, resulting in inappropriate
planning.
There are uncertainties associated with several issues that lie outside the control of the people and the system (for
instance, bad supply management resulting in many uncertainties in lead time and quantity delivered, and so on).
The net result of these problems is that the predictions made by MRP systems may often turn
out to be less accurate and the system may have to be rerun often. This could also result in
several production schedule changes and consequent delays in the downstream supply chain.
Moreover, there are other limitations in using the MRP system. The amount of computation
involved in generating component-wise schedules for the planning horizon is large. Real-life
examples require thousands of iterations that consume time. In some cases, it is not uncommon
to have a single run of an MRP extending for about 12–16 hours. Although speed and
availability of computing power keep increasing continuously, this issue still merits some
attention and puts realistic limits to the frequency of generation of MRP schedules.
Therefore, an organization needs to incorporate certain aspects into the MRP planning
framework to minimize problems arising out of these issues. Alternative methods are available to
re-run an MRP system and they have implications on the accuracy, cost, and time pertaining to
the exercise. However, there are methods available to handle some of the uncertainties in the
system and thereby reduce the risk of shortage, but such alternatives have cost implications as
well. We shall look at some of these alternatives in the following sections.
Regeneration is a method to update the MRP schedules where the MRP system is run from scratch.
The decision to use net change or regeneration depends on the magnitude of changes that
occur in an organization. If the number of changes to be incorporated in the system tends to be
large, then organizations are better off using the regenerative method for updating MRP
schedules. One sees a similar kind of logic being applied to bringing out the next edition of the
telephone directory in a large metropolitan city such as Mumbai. Every year, they may come up
with a new telephone directory (equivalent to regenerative method). However, between two
editions of the telephone directory, they bring out a small supplement announcing the changes
(equivalent to net change). The cost of running an MRP system and the number of changes
happening in the planning horizon influence the type of updating procedure employed and the
frequency of updation.
Consider a hypothetical plan of Maruti Udyog Limited (MUL) to produce 20,000 Omni vans,
15,000 Altos, and 9,000 Zens during a month. Using MRP logic, MUL can schedule the arrival
of materials, sub-assemblies, and components at various shops in the factory. MRP thus
addresses the material aspect of the planning problem. However, in addition to material, we need
capacity in the form of resources such as machines and skilled labour. What is the guarantee that
there is sufficient capacity available in the factory to complete the tasks as per the MRP
schedule? Clearly, MRP merely addresses “what needs to be produced” during each time period
in a planning horizon.
However, MUL should also take into consideration “what can be produced” during each time
period in a planning horizon and do the required planning to match these two. Therefore,
capacity requirement planning (CRP) is necessary to ensure that what needs to be produced
during a period can in fact be produced. CRP is a technique that applies logic similar to MRP to
address the capacity issues in an organization. Similar to MRP, CRP develops schedules for
planned releases of capacities to specific work orders as identified in an MRP schedule. The
output of an MRP process becomes the basis for the CRP exercise.
Capacity requirement planning (CRP) is a technique that applies the MRP logic to address the capacity issues
in an organization.
The notion of dependency applies very well to capacities also. Every manufacturing process
generates a dependent demand for the resources involved in the conversion process. In the MUL
example, if it takes three hours to complete the painting of one Alto and 20 minutes to complete
the final inspection process using a specialised photo-sensitive gadget, then 45,000 hours of
paint-shop capacity and 5,000 hours of the testing facility are required during the month. Just as
the BOM provides information on materials required to complete one unit of the parent, a routing
file or a process plan will provide information on the resources that are required to complete the
processing of one unit of the parent. The basis for the MRP is an MPS. Similarly, CRP systems
employ the detailed schedule generated by an MRP system as the basis for capacity planning.
Just as MRP systems perform netting of the requirements using on-hand inventory data at the
beginning of a planning horizon, a CRP system uses the capacity status as the starting point for
the “netting” process. Figure 16.17 illustrates the underlying logic of CRP.
FIGURE 16.17 The underlying logic of CRP
There are striking similarities between MRP and CRP and therefore organizations use logic
similar to that of MRP for performing CRP (see Table 16.11 for a comparative picture of both).
However, there are certain important issues that one needs to understand about MRP–CRP–MPS
interfaces. In a simple hierarchical mode, MPS will drive MRP and MRP in turn will drive CRP.
If both the schedules are feasible, then the plan is finalized. On the other hand, if there are
mismatches between capacity and material schedules due to non-availability of capacity to meet
the MRP schedule, then it calls for a few iterations of the process. The MRP schedules are first
modified to obtain feasible MRP and CRP schedules. However, if this is not possible, then the
MPS is modified until feasible schedules are obtained for both material and capacity. The
process is presented in Figure 16.18.
The logic of MRP is sufficiently general to apply to several other areas of business. One
extension of MRP is to plan distribution inventories in the downstream supply chain, consisting
of depots and stocking points at the dealer level. Consider a downstream supply chain consisting
of a factory warehouse, four distribution centres, and 20 dealers geographically spread out in the
country. The demand at the dealers’ level aggregates up in the chain and eventually reaches the
factory warehouse. Transporting the products through the supply chain involves a certain lead
time. Further, the products are transported in some preferred lot sizes. In general, ordering
decisions happen based on a periodic review of stock and the emerging demand. This causes the
demand at the next level in the supply chain to be lumpy. Therefore, it calls for time phasing of
the planning process.
TABLE 16.11 MRP and CRP: A Comparative Picture
Criteria for Comparison Material Requirements Planning (MRP) Capacity Requirements Planning (CRP)
Input Bill of material (product structure) Routing information (process plan)
MPS MRP schedules
Lead-time data Lead-time data
Inventory status Capacity status
Lot-sizing rules
Output Purchase orders Capacity loading schedules
Output Purchase orders Capacity loading schedules
Work orders Capacity usage profiles
It is easy to notice that the logic of MRP could be readily applied to planning distribution
requirements. Using a time phased data, organizations could schedule the planned receipt of
products at various points in the supply chain and offset these by the required lead time to arrive
at planned order releases for dispatch of material to the demand points. A dealer may review the
inventory on hand and the likely demand during the planning period and arrive at the net quantity
to order. Based on his lot-sizing considerations, he may lot the orders and determine the planned
receipts. Further, by offsetting the order to the extent of the lead time required for receiving the
order (due to transportation time and administrative processes) he may arrive at planned order
releases for each order. This becomes the gross requirement for the distribution centres. The
entire process could be repeated at the distribution centres and the planned order releases of the
four distribution centres eventually influence the MPS at the factory.
A distribution requirement planning (DRP) exercise will help organizations and their supply
chain partners to jointly plan and reduce investment in inventory in the supply chain. They will
be in a position to respond to changes in the demand, such as sudden surges and drops, and have
a cost-effective operation. They will also be able to offer a high level of service to their
customers. However, unlike the MRP exercise, a DRP exercise relies on key information
pertaining to planned order releases outside the domain of an organization. Retailers should be
willing to share with dealer the planned order releases and their estimates of upcoming demand,
for better planning. Similarly, dealers need to share similar information with the manufacturer. If
the information exchange is not proper and if there is no data integrity, then the value of the
entire exercise will be severely undermined. It will eventually result in inventory build-up or
shortages, poor service, and increased costs of the operation in the supply chain.
As we see from this list, MRP II covers all activities, from business planning to servicing the
customer. In reality, the business planning exercise triggers dependency relationships for all
resources in an organization. The forecasting/demand management module and the order entry
system essentially interface with the outside world and bring recent information into the planning
system. Based on these, production planning, MPS, and other requirements planning can be
done. Since the outcome of these exercises is to procure items and services from outside and
perform in-house activities as per plan, the relevant modules have also been included to close the
gap. Essentially, the focus is on planning for all the resources that an operations system requires.
The advantage of MRP II lies in its ability to provide numerous feedback loops between
different modules and minimize re-planning on a piece-meal basis. We saw in an earlier section
that with the inclusion of CRP it was possible to close the gap between material and capacity.
Similarly, adding DRP could address distribution-related issues. As more and more gaps are
closed, it promotes a centralised approach to planning and promises to bring additional benefits
arising out of integration.
Enterprise resource planning is an organization-wide planning system that utilizes some common software and
an integrated database for planning and control purposes.
In addition, other tools are also available for generating Web interfaces, coding and
programme generation for specialised requirements, report generation routines, data
import/export, and a library of best practices from which an organization could choose processes
for implementation. A wide number of software options are available for ERP today. The most
popular among them include SAP, Oracle, Ramco Systems, PeopleSoft, and JD Edwards. Ramco
Systems provide ERP software using a cloud based architecture focusing on small and medium
enterprises. Initially, ERP software were proprietary standards that were expensive and
inflexible. However, all of them are now available as cloud based and customizable. They are
also relatively less expensive and amenable to a greater degree of customization.
The heart of ERP is the organization-wide integration of several activities. This ranges from
the integration of functions to markets, divisions, plants, products, and customers across the
globe. The greatest benefit organizations get from implementing ERP is its ability to link various
functional areas of business tightly through the software. Figure 16.19 is an illustration of the
idea. Let us consider the issue of cash management activity pertaining to treasury. Purchase and
sales order information directly affects the cash forecasting exercise. Similarly, the accounts
payable and accounts receivable functions affect the actual cash management position. The ERP
software will ensure that with the input new information in each of these modules, the treasury
module will get automatically updated with this information.
Prior to the implementation of ERP, an organization re-engineers its processes. It also
identifies the level of customization it needs on various functionalities in the software. Usually, a
consulting firm assists the whole process and the software is implemented in a project mode.
Through the efficient use of a host of IT-enabled mechanisms, ERP promises to cut down
cycle time, transaction costs, layers of decision making, and thereby improve responsiveness and
flexibility. The other benefits include a greater sense of empowerment, better visibility of
information, and improved customer service. Moreover, an ERP system could be the backbone
of IT infrastructure for an organization with which several outward-facing IT applications can be
built in the future. All these will eventually lead to improving the competitiveness of the
organization at the marketplace.
Consider a typical eye hospital, such as the Aravind Eye Hospital in Madurai, which conducts a
large number of eye surgeries. A typical eye surgery requires certain resources. Let us look at
four key resources: eye surgeon, operation theatre (OT) and medical equipment, pre-operative
care, and post-operative care. Let us assume that a typical eye surgery requires 1 hour each of the
eye surgeon and the OT, 8 hours of pre-operative care, and 20 hours of post-operative care. If we
need to plan to perform 1,000 operations in a 6-day week, one can compute the amount required
of each of these resources. For example, the hospital requires 1,000 hours each of the eye
surgeon and OT, the 8,000 hours of pre-operative care, and 20,000 hours of post-operative care.
Furthermore, we also know that an item in this list could in turn depend on a host of other
resources. Pre-operative care may mean the requirement of certain skills of medical
professionals, diagnostic and curative equipment, medical consumables, medicines, support staff,
and so on.
There are striking parallels between the eye hospital example and that of the telephone set,
which was used to develop the logic of MRP. The end product in this case is a treated and
discharged patient (similar to the telephone instrument in the manufacturing example). In the
telephone examples, four sub-assemblies were required to complete the product. In this case, the
product structure in the previous level will have post-operative care as a single item.
Therefore, as in the manufacturing situation, one can develop a bill of resources (BOR) in the
case of service organizations and employ a similar planning exercise to schedule service
deliveries and the associated resources and materials.
Ramco SRP provides an end-to-end solution for customers of service industry by tightly
integrating key functions such as HR and Finance with Project Management. Its core
functionality is project based billing where projects can be Fixed Bid/Milestone or Time and
Material (T and M) in nature. SRP is also equipped with built-in adaptors to integrate with
project management tools such as Microsoft Projects and Primavera. It further provides
comprehensive project tracking with revenue recognition. Figure 16.20 provides the product
guide for Ramco’s SRP.
Ramco’s SRP provides industry specific applications for the services sector. ITes/BPO
industry is project-centric, making project management very essential for its success. SRP
supports milestone-based, T and M based as well as capped T and M based projects. In
addition, it also helps in comprehensive project tracking by integrating with MS Project and
Primavera. SRP provides an integrated platform for process-centric consulting companies
where timesheet entry, invoice generation, and billing are all inter-linked to one another. This
reduces data redundancy and manual data entry into excel spreadsheets. Being a resource-
centric industry, workforce management plays a key role in staffing companies. SRP
application for staffing company includes processes and the flexibility and variations
required for recruitment and selection, onboarding and resource deployment, and a
transparent client billing process from Time/Attendance approval through payroll till
invoicing.
SRP provides seamless real-time flow of information between multiple departments within
the company such as HR, Finance, Project Control Office, and Resource Management Group.
Employees, who form the core of services industry, can be managed very efficiently by
optimizing their skills and effort across multiple projects or assignments simultaneously. On
the whole, SRP reduces the time and effort spent in manual activities through process
automation and also makes day-to-day transactions of the organization paperless by
providing a single electronic source of information as a global data repository.
Source: Adopted from http://www.ramco.com/hcm/why-ramcohcm/services-resources-planning/; Last accessed on 15 June
2014.
SUMMARY
Dependent demand items exhibit certain characteristics such as lumpy demand and parent-offspring relationships
that can be exploited while planning for materials in organizations.
Bill of materials (BOM) and product structures depict the dependency relationships among various components in a
manufacturing set-up.
Various lot-sizing rules can be utilized to batch requirements and schedule orders. The lot-sizing rules strike an
appropriate trade-off between carrying cost and ordering cost.
MRP logic involves a four-step logical process: net–lot–offset–explode. Through an iterative process, the material
requirements of all components in an organization during a planning horizon are arrived.
Regeneration and net change are the two methods used for updating the MRP schedules.
Capacity requirements planning (CRP) uses logic similar to that of MRP.
Furthermore, the logic of MRP can be extended to several other spheres of an operations system. MRP II systems
provide this functionality. Modern-day ERP systems are further extensions of MRP logic to other areas of business
such as marketing, finance, and personnel.
REVIEW QUESTIONS
1. What do you mean by dependent demand and independent demand? Is knowledge of this useful to an organization? If so,
in what way is this information useful?
2. Consider each of the following operating systems and identify three examples of dependent and independent demand
items.
3. What are the key attributes of a dependent demand item? Explain with one example each from manufacturing and service
organizations.
4. Why should an organization plan the requirements for a dependent demand item differently from that of an item with
independent demand?
5. Suppose that an organization uses normal inventory control methods to plan for its dependent demand items. What will
the impact of this be on material planning? Under what conditions will the planning result in satisfactory results?
6. What is meant by a bill of materials (BOM)? Why do organizations need BOMs?
7. Are BOMs and product structures the same? Explain with an example.
8. An organization engaged in the manufacture of electrical appliances is not clear if they need to use a modular or a single-
level BOM. Prepare an advisory note clearly specifying the conditions under which each of these alternatives is
appropriate.
9. Suppose that an organization prepared an MRP on the basis of incorrect data on lead time of the components. What will
the impact of this be on the actual operations?
10. What is meant by the “explosion” of a BOM? Explain with a simple example
11. Identify the appropriate lot-sizing rule for each of the following components. Justify your choice of the rule.
PROBLEMS
1. Omega, a manufacturer of office supplies, produces a desktop calendar unit using a simple process and the following
components. The desktop calendar unit is basically assembled using a base plate, a sliding unit and a calendar assembly.
The sliding unit is made up of a plastic tray and a set of 30 cards to store telephone and address information. The sliding
unit is fixed to the base plate using a pair of 3 mm screws. The calendar assembly is made up of two U-shaped hooks and
a calendar (consisting of about 400 pages containing the daily calendar details). The calendar assembly is fixed to the base
plate by clamping the U-shaped hooks using a pair of 3 mm screws each.
Each base plate is made of plastic sheets and requires 150 square inches of plastic sheet. Similarly, the sliding unit is
also made of the same plastic sheet and requires 30 square inches of the plastic sheet. Each U-shaped hook is made from 7
inches of 4mm steel rod. The calendar is bought directly from a supplier. Finally, the product is packed in a specially
made carton box and sealed using 12 inches of tape. There are also four bar-coded labels stuck to each carton box after it
is sealed. This makes it a packaged calendar ready for shipping.
Two different varieties [one is the basic version and the other has provision for keeping pens/pencils (these affect
only the base plate)]
Five different colour choices in the desktop calendar unit (these affect the base plate and the sliding unit)
Three different configurations of the calendar itself (one page per day, half a page per day, and month planner sheet
at the beginning of every month in addition to one page per day)
a. How many different varieties of desktop calendar units can the organization offer to the customer?
b. Develop a BOM for the product range offered by Omega.
3. A gearbox manufacturer has 20 gearboxes in stock. Each gearbox has four gears. There are 200 gears already in stock.
The gears are made from gear blanks. The stock of gear blanks in the stores is 100. Each gear blank requires 30 kilograms
of alloy steel. The stores have 7,000 kilograms. of alloy steel. Compute the requirement of components for manufacturing
570 gearboxes in the next month.
4. Consider the product structure given in Figure 16.21 pertaining to a product manufactured by Oriental Housings & Seals.
The numbers in parentheses in the figure indicates the number of units of the item required to assemble one unit of its
parent. Use the information available in the product structure to answer the following questions:
FIGURE 16.21 Product structure for Oriental Housings & Seals product
a. How many units of C are required to manufacture one unit of Product A? Did you make any assumption in
computing this value?
b. How much inventory of each component is required for satisfying a demand of 100 units of the final product if: (i)
there is no stock of finished goods (ii) there is a stock of 30 units of the final product?
c. Consider Problem (a). Will the computations change if there is some stock of Item B? Why and by how much?
(Hint: Assume that you have x units of Item B to proceed with the analysis)
5. In Problem 4, the lead times for manufacture/assembly of the components are as follows:
Product A 2 weeks
Component B 1 week
Component C 2 weeks
Component D 2 weeks
Component E 1 week
a. How early can Oriental deliver an order of 100 units of the product to the customer if (i) there is no stock of Item B
(ii) there is a stock of 220 units of Item B?
b. Will the results change if there is a stock of Product A?
6. Given in Table 16.12 is a partially completed MRP working for Component X. Using the information provided, complete
the table.
Use this information to analyse the following five lot-size rules and suggest an appropriate lot-sizing rule for the
component.
8. Consider the product structure of a manufacturing firm, as shown in Figure 16.22. The company is interested in
performing an MRP exercise for the next eight weeks to cover the requirements indicated in the MPS. The MPS
requirements for the next eight weeks are 20, 80, 100, 200, 100, 0, 120, and 150.
TABLE 16.14 Lead Time, Inventory Status, and Lot-sizing Rules for Problem 8
a. Do we need to schedule order releases for Component B? If so, when and how much?
b. How long will the on-hand inventory of Item D satisfy the requirement?
c. How many orders need to be placed for Component F?
d. If the cost of scheduling an order is ₹2,000, what is the total cost of ordering for the MRP plan?
9. Figure 16.23 shows the product structure for a manufacturing firm. In addition to the final product (Product A),
Component B is sold as a spare in the market. The MPS for Product A and Component B and other relevant information
for the problem are given in Tables 16.15 and 16.16.
Perform an MRP exercise to cover the requirements of the MPS and answer the following questions:
a. What is the gross requirement for Component B during the six periods?
b. How many orders do we have to place for Component C and in what quantities?
c. Suppose the lead time for Component G is two weeks instead of one week, what will its impact be on the planning
process?
d. Suppose that the economic order quantity for Component C is 320 units. What will the appropriate POQ lot-sizing
rule be for Component C?
10. Consider the following MRP (Table 16.17) for Component C, which is used in a product manufactured by an
organization.
Of late, the organization has been experiencing frequent shortages of this component. It has been estimated that on an
average, the company loses ₹5,000 by way of lost business and goodwill due to these shortages in a six-week period.
Therefore, it has been decided to include a safety stock of 100 units. Rework the planning details for this component by
incorporating the safety stock requirement.
11. In Problem 10, assume that the organization decided to implement a safety lead time of one week instead of safety stock.
a. Rework the planning details for this component by incorporating the safety lead time requirement.
b. Compare the two results and list two significant observations that you have made.
NET-WISE EXERCISES
1. Visit the Web sites of the major ERP vendors (SAP, PeopleSoft, Oracle, Ramco, J.D. Edwards). Study the product
offerings of all these vendors and prepare a detailed report to answer the following questions:
a. What are the key components of the ERP system offered by the major vendors?
b. How do these vendors differentiate differ their offerings?
c. If an organization wishes to choose one of the four vendors, what factors will you suggest to the organization to
enable appropriate selection?
d. Are there industry-specific offerings of ERP by these vendors? If so, which sectors of the industry are covered and
how well?
MINI PROJECTS
1. Take a simple household gadget or appliance such as a dining table set or a sofa set with a centre table. Carefully analyse
the assembly of the appliance and identify the parts that make up the appliance. Identify the major assemblies and sub-
assemblies of the appliance.
a. Using this information prepare a complete BOM and a product structure for the appliance
b. What are the various ways by which you think product varieties can be created for this appliance? Identify them
and prepare an indented BOM for the entire set of product offerings.
c. Assume an MPS and other relevant data for the components that you have identified for the appliance. Can you set
up a standard table for performing the MRP exercise for the appliance?
2. Set up the table created in Problem 1 in a Microsoft Excel worksheet (assume a planning horizon of 6 periods). Use the
worksheet to accomplish the following tasks:
a. Perform the MRP exercise for the appliance that you have chosen.
b. Assume that the MPS is wrong. Quantities have been underestimated to the extent of 10 per cent during Periods 4
and 5 and overestimated to the extent of 5 per cent during Period 6. What is the likely impact of this on customer
commitment?
SUGGESTED READINGS
G. M. Bayhan, “Implementation and Evaluation of an MRP II Software Package in a Sanitary Ware Company,”
Production & Inventory Management Journal 40, no. 4 (1999): 41–48.
T. H. Davenport, “Putting the Enterprise into the Enterprise System,” Harvard Business Review 76 (July–August
1998):121–131.
P. Gazmuri and S. Maturana, “Developing and Implementing a Production Planning DSS for CTI Using Structured
Modelling,” Interfaces 31, no. 4 (2001): 22–36.
J. Haddock and D. E. Hubicki, “Which Lot Sizing Techniques are Used in Material Requirements Planning?”
Production and Inventory Management Journal 30, no. 3 (1989): 53–56.
S. Kaushika Madhavan, “Erhardt + Leimer India: Strategic Transformation Through ERP,” Vikalpa 25, no. 3
(2000): 57–69.
L. X. Li, S. S. Chaudhry, P. E. Chaudhry, and Y. Wang, “Evaluation of Acquiring and Implementing a
Manufacturing Resource Planning System,” Production and Inventory Management Journal 42, no. 3 (2001): 1–8.
S. Nahmias, Production and Operations Analysis, 4th edition (New York: McGraw-Hill Publishing Co., 2000).
J. Orlicky, Material Requirements Planning: The New Way of Life in Production and Inventory Management (New
York: McGraw-Hill Publishing Co., 1975).
P. J. Rondeau and L. A. Litteral, “Evolution of Manufacturing Planning and Control Systems: From Re-order Point
to Enterprise Resources Planning,” Production and Inventory Management Journal 42, no. 2 (2001): 1–7.
T. E. Vollmann, W. L. Berry, and D. C. Whybark, Manufacturing Planning and Control Systems, 4th edition (New
York: Mc-Graw-Hill Publishing Co., 1997).
T. F. Wallace, MRP II: Making It Happen: The Implementers’ Guide to Success with Manufacturing Resource
Planning, 2nd edition (Essex Junction, VT: Oliver Wight Ltd. Publications, 1995).
O. W. Wight, Manufacturing Resource Planning: MRP II (Essex Junction, VT: Oliver Wight Publishers, 1984).
CHAPTER 17
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
Organizations need to spend a considerable amount of money to carry inventory. The cost of stores and warehousing and the
administrative costs related to maintaining inventory and accounting for it form a significant part of this cost.
The term inventory refers to any idle resource that can be put to some future use. Manufacturing
and several service organizations have significantly invested in inventory. Often, investment in
inventory has a direct bearing on the profitability of a firm. Experiences in the last twenty years
suggest that the world-class performance of a firm hinges on the firm’s ability to cut investment
in inventory to very low levels. Automobile assembly plants and component manufacturers are
known to carry very little inventory and yet provide good quality finished goods at the right time.
In addition to cutting down cost, reduced inventory levels help an organization improve quality,
planning systems, and supply chain coordination. They also reduce wastage and obsolescence.
Hence, inventory planning and control continues to derive considerable attention of the
management in organizations.
The term inventory denotes any idle resource that could be put to some future use.
In manufacturing organizations, finished goods and spare parts typically belong to the category
of independent demand items. While planning for a dependent demand item is done to meet
manufacturing requirements, in the case of independent demand items, it is done to meet
customer requirements. Two attributes characterize and distinguish independent demand items:
They are in continuous demand, and there is an element of considerable uncertainty in the
demand for such items.
In manufacturing organizations, finished goods and spare parts typically belong to the category of independent
demand items.
Continuous Demand
Independent demand items are in continuous demand. Consider the sale of consumer appliances
of manufacturers such as Videocon or LG. The demand for 32” LCD colour television panels in
a particular city will be continuous. When there is a continuous demand for an item, constant
availability of items and periodic replenishment of stock are important elements of the planning
process. Non-availability of items translates into lost sales, poor customer goodwill, and
additional costs in servicing the promised deliveries.
Uncertainty of Demand
There is an element of considerable uncertainty of demand in the case of independent demand
items. On the other hand, in the case of dependent demand items, the demand is always derived
and hence known with certainty. For example, the number of tyres required per month for new
vehicles is known accurately once the monthly production is finalized. However, the number of
spark plugs required for sale as spare parts needs to be estimated on some basis. Therefore,
inventory planning for independent demand items should include some cushion for handling
uncertainties whenever they are significant.
Given these requirements, inventory planning of independent demand items must address the
following two key questions:
a. How much? Good inventory planning must devise means by which the planner can continuously replenish the stock as it
depletes over time. The replenishment could be of a fixed quantity or variable. It could be based on the costs associated
with inventory. It could also satisfy certain practical considerations such as truck capacity, quantity discounts, and
minimum order quantity restrictions posed by the supplier.
b. When? The other issue is the timing of replenishment. Since demand is continuous and uncertain, the timing of
replenishment is equally crucial. There are numerous ways of addressing the timing. It could be based on consumption
patterns, the value of the item, managerial preferences for review, and other practical considerations.
Inventory planning of independent demand items must address the two key questions: how much, and when?
Before planning for inventory, it is important to know why organizations carry inventory and
what factors influence the level of investment. Five types of inventory are normally found in
most organizations.
Seasonal Inventory
Organizations carry inventory to meet fluctuations in demand arising out of seasonality. During
festival periods, the demand for consumer durables may be high due to an increase in disposable
income in the hands of consumers. In order to meet this surge in demand, there is inventory
build-up during non-peak periods. Similarly, in a fast-food restaurant, a certain amount of
inventory build-up happens during non-peak hours to handle the increase in demand during peak
hours. In both these cases, inventory plays the crucial role of addressing short-term capacity
issues in an organization.1
Five types of inventory are normally found in organizations: seasonal inventory, decoupling inventory, cyclic
inventory, pipeline inventory, and safety stock.
Decoupling Inventory
Manufacturing systems typically involve a series of production and assembly workstations. Raw
material passes through these stages before it is converted into finished goods. Each stage
behaves idiosyncratically on account of varying process times, downtimes, and resource
availability. Therefore, the planning and control of such multi-stage production processes
becomes very complex. One way to simplify the production planning and control problem is to
decouple successive stages using inventory at some intermediate points. Each stage will have an
input buffer and an output buffer. The output buffer of the preceding stage becomes the input
buffer of the succeeding stage. Inventory decisions in this case require analysis of workstation
capacities, resource availability, and bottlenecks.
Figure 17.1 illustrates the use of decoupling inventory. In the first case, there is no decoupling
inventory. In this case, all the ten stations of production are to be closely monitored and flow
balance is to be ensured on a continuous basis. If there is a problem in one of the stations, it will
affect all upstream and downstream stations. Further, the vagaries of demand uncertainty will
directly affect the system’s performance. Planning and control is challenging, especially when
the number of stations is very large. On the other hand, the second case in Figure 17.1 has
decomposed the production into three major stages. Each stage is linked to the other using
decoupling inventory. This enables each stage to work with reasonable levels of independence.
Further, the adverse impact of one stage will not immediately affect the others. Therefore, a
certain amount of decentralized planning and control is possible.
FIGURE 17.1 An illustration of the use of decoupling inventory
Pipeline Inventory
Pipeline inventory pertains to the level of inventory that organizations carry in the long run due
to non-zero lead time for order, transport, and receipt of material from the suppliers. Because of
the geographical distances between the buyers and the suppliers and the host of business
processes involved in ordering and receiving material, there is a time delay between order
placement and order receipt. The inventory carried to take care of these delays is known as
pipeline inventory.
Consider the example of the hospital. Suppose it takes three days to supply disposable
syringes. Then the hospital should place an order at the end of Day 17 in order to replenish the
stock to 10,000. This means that an order will be placed when the inventory level reaches 1,500
(3 days at the rate of 500 syringes per day).2 In general, if the lead time for supply is L, and the
mean demand per unit time is μ, then the pipeline inventory is L × μ.
In the long run, the system will always have this inventory in order to take care of lead time
for supply. The only way organizations can reduce this inventory is to cut down lead time for
procurement, production, and distribution. Locating suppliers nearby, re-engineering the
business processes related to procurement and distribution of materials, and using efficient
alternatives for logistics are some of the ways in which an organization can reduce lead time and
pipeline inventory.
FIGURE 17.3 Cyclic inventory, pipeline inventory, and safety stock
Safety Stock
Organizations also have additional investment in inventory to buffer against uncertainties in
demand and supply of raw material and components. We know from the elementary theory of
probability distributions that when demand is stochastic, carrying average inventory will ensure
that the demand is met only 50 per cent of the time (in the long run). However, in order to
improve the availability to meet uncertain demand, an additional quantity known as safety stock
is kept. Greater investment in safety stock leads to a lower probability of inventory going out of
stock. Similarly, the higher the uncertainty, the greater is the need for safety stock. Safety stock
only serves to prevent shortages in the short run. However, in the long run, the demand will tend
toward an average value and the safety stock will not be consumed. Figure 17.3 represents an
inventory system with safety stock and other types of inventory.
EXAMPLE 17.1
A manufacturer of transformers requires copper (both in plate and wire form) as a key
ingredient. The average weekly requirement of copper is 200 tonnes. The lead time for the
supply of copper is two weeks. If the manufacturer places monthly orders of copper, analyse
the various types of inventory in the system.
Solution
Order quantity Q = 1 month requirement = 800 tonnes.
Reducing safety stock calls for reducing uncertainty in the system. Investment in superior
planning and forecasting systems and good supplier development practices will help
organizations reduce uncertainty in the system. In the case of components manufactured in-
house, better production planning and improved maintenance of resources will help reduce
uncertainty in the system.
Cyclic inventory, pipeline inventory, and safety stock are critically linked to both the “how
much” and “when” questions in inventory planning. Lead time influences the “when” decision
directly and determines the level of pipeline inventory in the system. Similarly, cyclic inventory
is the outcome of the “how much” decision that an inventory planner makes. Safety stock
influences both “how much” and “when” in an indirect sense. However, in order to understand
how these decisions are made, one needs to understand various costs associated with inventory
planning and control.
Cyclic inventory, pipeline inventory, and safety stock are critically linked to the “how much” and “when”
questions in inventory planning.
There are several costs associated with inventory planning and control. These costs could be
classified under three broad categories:
The cost of carrying inventory. All costs related to maintaining inventory in organizations will be classified under this.
The cost associated with ordering material and replenishing it in cyclic intervals.
The cost arising out of shortages.
Inventory control models should take these into consideration and aim at minimizing the sum of
all these costs. Sometimes, the unit cost of the item for which inventory planning is done is also a
relevant cost for decision making. This is because when large quantities are ordered, there could
be some discount in the unit cost of the item.
Inventory-carrying Cost
Organizations need to spend a considerable amount of money to carry inventory. The most
significant component is the interest for the short-term borrowals for the working capital required
for inventory investment. The second significant cost relates to the cost of stores and
warehousing and the administrative costs related to maintaining inventory and accounting for it.
The elements of storing and warehousing costs include the following:
Investment in store space and storage and retrieval systems
Software for maintaining the inventory status
Managerial and other administrative manpower to discharge various activities related to stores
Insurance costs
Cost of obsolescence, pilferage, damages, and wastage
In the case of high-tech electronics and short-shelf-life items, such as active chemical
compounds, food items, and pharmaceutical formulations, obsolescence costs could be
significant. Table 17.1 shows a sample computation of carrying cost for a company. The
components of inventory-carrying costs exert considerable pressure on an organization to keep
inventory levels low. However, the ability of an organization to reduce inventory depends on the
nature of business processes in place, the lead time for procurement, and the quality of planning,
as pointed out in the earlier section.
Note that all the costs related to carrying inventory are directly related to the level of
inventory. Graphically, it can be shown as a simple linear relationship (Figure 17.4). Let Cc
denote the inventory-carrying cost per unit per unit time. Since the interest component is the
predominant part of Cc, the usual practice is to represent Cc as a percentage of the unit cost of the
item. Thus, if Cu is the unit cost of the item, then Cc = iCu, where i is in percentage. For example,
if the unit cost of an item is ₹5,000, the annual interest charges are 12 per cent, and the other
annual costs related to carrying inventory are 3 per cent, then the inventory-carrying cost is 15
per cent of the unit cost, that is, ₹750 per unit per year.
For an order quantity of Q, the average inventory carried by an organization is Q/2. Therefore,
Cost associated with carrying inventory
Cost of Ordering
Replenishment of cyclic inventory is achieved by ordering material with the suppliers.
Organizations perform a series of tasks related to ordering material. These include search and
identification of appropriate sources of supply, price negotiation, contracting and purchase-order
generation, follow-up and receipt of material, and eventual stocking in the stores after necessary
accounting and verification. In Chapter 7, we have discussed several aspects pertaining to this in
some detail. All these tasks involve manpower, resources, and time that can be classified under
cost of ordering. Since several of these costs are fixed in nature, ordering a larger quantity could
reduce the total costs of ordering. A larger order quantity Q will require fewer orders to meet a
known demand D, and vice versa. The relationship between Q and the total cost of ordering is
graphically shown in Figure 17.4. Table 17.2 shows a sample computation of ordering cost for a
company.
The number of orders to be placed to satisfy a demand of D = D/Q. If C0 denotes the cost of
ordering per order, then:
Cost of Shortages
Despite careful planning, organizations are likely to run out of stock for several reasons. There
could be a sudden surge in demand. Alternatively, the suppliers might not have delivered the
material as per schedule, or a lot could have been rejected because of defective components.
Such events disrupt production and have a cascading effect down the supply chain. Delivery
schedules are missed, leading to customer dissatisfaction and loss of goodwill. It also introduces
additional costs arising out of pushing the order back and rescheduling the production system to
accommodate these changes. Rush purchases, uneven utilization of available resources, and
lower capacity utilization further escalate the costs in the system. All these form part of the cost
of shortage.
The cost of carrying and the cost of ordering are fundamentally two opposing cost structures in inventory
planning.
The cost of shortage per unit is denoted by Cs. Since the effects of shortage are vastly
intangible, it is difficult to accurately estimate Cs. In practice, managers tend to use other
measures to incorporate the cost of shortage if estimation of Cs proves to be futile.
Let us consider a situation in which the demand for an item is continuous and is known with
certainty. Since the demand is known, we exclude the possibility of having shortages. Better
inventory control requires that we answer the “how much” and “when” questions by balancing
the total costs of carrying inventory and ordering. For an order quantity of Q, we can compute
the total cost of carrying and ordering from Eqs. 17.2 and 17.3.
Total cost of the plan = Total cost of carrying inventory + Total cost of ordering
When the total cost is minimum, we obtain the most economic order quantity (EOQ).
By taking the first derivative of Eq. 17.4 with respect to Q and equating it to zero, we can obtain
the EOQ. Differentiating Eq. 17.4 with respect to Q we obtain:
Note that the second derivative is positive and hence we obtain the minimum cost by equating
the first derivative to zero. Equating the first derivative to zero and rearranging the terms, we
obtain Eq. 17.5.
Denoting EOQ by Q*, we obtain the expression of Q* as:
Q* answers the “how much” question directly. Every time the inventory depletes to zero, it is
economical to place an order equal to Q*. Similarly, time between orders answers the “when”
question. Since the demand is continuous, the demand rate will determine “when” to place an
order. For instance, if the annual demand for an item is 2,500 units and if there are 250 working
days, the daily demand is 10. If Q* is 300, then it is implied that an order needs to be placed
every 30 days.
2500
EXAMPLE 17.2
Solution
Annual demand for the item (D) = 2,500 boxes
Number of working days = 250
Hence, the manufacturer will place an order for 200 boxes of the component once in every
20 days and will incur a total cost of ₹27,000 for the plan. Any quantity above or below will
increase the cost of the plan.
In practice, several of these assumptions do not hold. Despite this, the EOQ model can be
applied with suitable modifications because it is robust. Because of the square root in the
formula, changes in model parameters, such as demand and cost, have less impact on the
solution. If demand is not known accurately, one can use representative figures for D and assess
the impact of changes on the total cost of the plan (see Problem 4 at the end of the chapter). The
assumption of instantaneous replacement can be easily taken care of by placing an order ahead.
If the lead time is L, then by placing an order L periods ahead of depletion of inventory, one can
ensure availability of material. Similarly, if the items are manufactured in-house, it is possible to
estimate the economic run length by a simple modification to Eq. 17.5.4
The EOQ model is robust and can be applied in several real-life settings with some modifications.
Assumptions (d), (e), and (f) indicate the need for examining the suitability of utilizing order
quantities different from Q*. Large order sizes result in price discounts. There are other reasons
for placing a larger order than what the EOQ suggests. Sometimes, it will be economical to
transport a truckload of items and save on transportation cost. In other cases, the supplier may
impose a minimum order quantity. In all these cases, the relevant cost for analysis includes cost
elements other than carrying and ordering. An analysis of these relevant costs will help the
inventory planner make an appropriate choice in order quantity. See Example 17.3 for an
illustration of this.
EXAMPLE 17.3
Consider Example 17.2. Assume that the carrying cost of the component remains ₹135 per
unit per year and that the supplier is willing to offer a discount on the unit price as per the
following structure:
Up to 399 boxes = No discount
400 − 799 boxes = 2 per cent discount
800 − 1,000 boxes = 3 per cent discount
What should the company do in this case?
Solution
The economic order quantity is 200 boxes and the unit price of the item is ₹750.
There are two other order quantities at which the unit price changes on account of
discount.
At Q1 = 400, unit price of the item = 0.98 × 750 = ₹735.00
At Q2 = 800, unit price of the item = 0.97 × 750 = ₹727.50
Since there is a discount on the price as the order quantity is varied, the total cost
comparison between alternatives can be made only after incorporating the purchase price.
Total cost for Q*
Since TC(Q1) is the lowest among the alternatives, the firm can make use of the discount
offered and reset the order quantity to 400 boxes.
With some modifications, the EOQ model can accommodate several practical considerations.
However, when the demand is uncertain, instances of shortages occur while operating the system
with EOQ. There are two ways of handling the issue of shortages. One is to incorporate shortage
cost in the model. The other is to use the concept of service level. Service level is the desired
probability of not running out of stock between the time an order is placed with a supplier and
the order is received. For example, if an organization wants to operate with a 90 per cent service
level, it means that in the long run, the organization is able to meet the demand on 90 per cent of
all occasions.
Service level is the desired probability of not running out of stock between the time an order is placed with a
supplier and the order is received.
Consider an item with uncertain demand. When the item is available in stock, there is no risk
of shortage. However, when an order is placed, the demand can be met only with the available
stock until the stock is replenished by an order. Hence, if an organization is protected from the
risks of shortage during the supply lead time, then inventory planning is better. The concept of
service levels seeks to achieve this objective. Let us understand the concept of service level
through a simple example.
Consider an item with uncertain demand that has a lead time of one week. Based on observed
consumption patterns one can estimate the weekly demand. Table 17.3 has illustrative data for
the item. The last two columns show the cumulative frequency of demand exceeding the lower
class. For example, the weekly demand exceeded 30 units during 98.25 per cent of the observed
occasions and exceeded 180 units during 18.42 per cent of the occasions. The information in the
table can also be plotted as a frequency ogave (see Figure 17.5).
From the data, one can compute the average demand during lead time to be 143 units. If the
organization stocks 143 units at the time of placing the order with the supplier, then in the long
run, it will be able to meet the demand during lead time only on 50 per cent of the occasions. In
other words, the organization has a 50 per cent service level if it operates with this inventory
level. Clearly, this will be an undesirable situation and managers would want to offer better
service. From the data, we can compute the number of units to be stocked to offer 90 per cent
and 95 per cent service levels to be 203 and 224 units, respectively. By stocking 60 more units
(203–143), the organization can improve the service level from 50 per cent to 90 per cent.
Similarly, by investing an additional 21 units, it can further increase the service level to 95 per
cent. This additional stock is known as “safety stock” because it provides the needed safety to an
organization to handle uncertain demand during lead time.
TABLE 17.3 The Service Level Concept: An Illustration
The concept of service levels permits managers to sidestep problems related to estimating
shortage cost per unit. Using this concept, it is possible for managers to relate the amount of
additional investment required and the improvement in service and strike appropriate trade-offs
between additional investment and service. In the example we just discussed, a 5 per cent
increase in service level from 90 per cent to 95 per cent entails an additional investment of 21
units. Depending on the unit cost of the item in question, the cost of carrying and the competitive
scenario, managers can strike a better trade-off in their choice towards the appropriate level of
service that needs to be offered.
The concept of service level permits managers to sidestep problems related to estimating shortage cost per unit.
FIGURE 17.6 Illustration of service level and the safety stock concept
If the organization places an order when the stock on hand is equal to the sum of mean demand
during lead time and safety stock, then it will be able to provide a service level of (1 − α) in the
long run. Therefore, an expression for reorder point (ROP) for systems with uncertain demand is
given by:
The previous sections provide organizations with the necessary building blocks to put inventory
control systems in place. In practice, organizations employ some methods to manage and control
inventory. Prominent among them include the use of continuous review of inventory and
periodic review of inventory. Let us understand the working of these two systems, and also how
various decision parameters of these systems can be estimated using the building blocks
provided in previous sections.
The daily demand for an item is stochastic and follows the normal distribution with a mean
of 100 and a standard deviation of 20. The supplier of the item takes two weeks to deliver the
item from the date the order is placed. What will be the appropriate reorder point for 90 per
cent and 95 per cent service level? The cost of ordering is ₹1,000 per order and the carrying
cost is ₹250 per unit per year. There are 250 working days in a year. If the organization
places orders in fixed quantities of 500, what will be the total cost of the plan?
Solution
Mean daily demand = 100
Number of working days in a year = 250
Total annual demand = 25,000
Lead time for supply (L) = 2 weeks = 14 days
Therefore, mean demand during LT, μ(L) = 14 × 100 = 1,400
Standard deviation of daily demand = 20
Variance of daily demand = 400
Variance of the demand during LT = 14 × 400 = 5,600
Standard deviation of demand during LT5, σ(L) = 74.83
For 90 per cent service level, α = 10 per cent and (1 − α) = 90 per cent.
From standard normal tables (Zα) = 1.28
Safety stock (SS) = Za × σ(L) = 1.28 × 74.83 = 95.78 ≈ 96
Therefore, ROP = μ(L) + Zα × σ(L) = 1,400 + 96 = 1,496
Total cost of the plan,
The “when” decision is answered by ROP. We can employ the computations in the previous
section to arrive at ROP. The “how much” decision is answered by our choice of Q. Several
alternatives are available to fix Q.
One is to compute EOQ and fix Q at this value.
The other is to fix Q to be the minimum order quantity that the supplier insists, if EOQ value is lesser than this.
The third option is to have any other preferred quantity arising out of practical considerations such as full truckload
requirement, quantity discounts, savings in transportation costs, economies of scale, or any other consideration.
The total cost of using a Q system can be computed using Eq. 17.8 for a chosen value of Q.
In a periodic review system, the inventory level in the system is reviewed at fixed intervals of time.
The order up to level S determines how much to order. Hence, it is important to derive an
appropriate expression for S. Let us consider an instant of time t, when a review is taken and an
order is placed with the supplier. The organization gets the next opportunity to review the
inventory R periods later. An order placed at that point in time will reach the organization only
after L time periods, which is the lead time for supply. Hence, at every review, the organization
needs to protect itself from the risks of shortages for a period of (R + L) periods. Therefore, an
expression for S can be derived by modifying Eq. 17.9 as follows:
Where, μ(L + R) = Mean demand during (L + R) and σ(L + R) = Standard deviation of demand
during (L + R)
The Q system is less responsive to changes in demand and poses greater difficulty in ordering multiple items
from the same supplier.
QR = S − IR
The P system is responsive to demand and enables ordering multiple items from the same supplier at the same
time.
However, the P system overcomes several of these limitations of the Q system and offers
many advantages over the Q system. This includes more responsiveness to demand, the ability to
order multiple items from the same supplier at the same time, ease of planning and control,
greater chances for linkages with MRP, and other planning systems in the organization. Table
17.4 compares the salient features of P and Q systems of inventory control. The safety stock
investment in the P system, however, is more than that of the Q system because it requires
protection from shortage for a longer period of time.
The safety stock investment in the P system is more than that of the Q system because it requires protection from
shortage for a longer period of time.
EXAMPLE 17.5
The purchase department has indicated that the lead time for procurement of this raw
material is two weeks. Past experience with the supplier suggests that there is no uncertainty
with respect to the lead time. The organization has been using EOQ for the purpose of
scheduling orders. However, there is a general feeling that it is not working satisfactorily. It
is not uncommon for the organization to experience stock-outs. Work out the parameters of
the P and Q systems of inventory control.
Solution
EOQ Model
Weekly demand = 200
Number of weeks per year = 52
Annual demand, D = 200 × 52 = 10,400
Ordering cost, C0 = ₹460 per order
Unit cost of the item = ₹300
Carrying cost, Cc = ₹60 per unit per year
Since there is a two-week lead time, an order needs to be placed two weeks ahead of
complete depletion of the inventory, that is, by the time the inventory position in the system
reaches 400 units. The order quantity is 400. There is no safety stock to protect the system
from shortages arising out of uncertain demand. The system can be improved either by using
a Q system or a P system, as shown here.
Q System
Standard deviation of weekly demand = 40
Lead time, L = 2 weeks
Mean demand during L, μ(L) = 2 × 200 = 400
Using EOQ as the fixed order quantity, the Q system can be designed such that, as the
inventory position in the system reaches 493, an order is placed for 400 units. In the long run
this will ensure a service level of 95 per cent.
P System
Using the time between orders derived from the EOQ model as the basis for review period
Mean demand during (L + R), μ(L + R) = 200 × (2 + 2) = 800 Standard deviation of demand
Managing inventory invariably amounts to handling a large number and variety of items. For
instance, an automobile manufacturer such as Ashok Leyland may have an inventory of about
50,000 items in their stores. A vast majority of them will be consumed (albeit in varying rates)
and will require a mechanism for monitoring inventory on-hand, establishing reorder points,
order quantity, and the type of inventory control system that needs to be adopted. Clearly,
establishing the same level of monitoring and control for all the items may not be practically
feasible. Therefore, organizations devise suitable ways of categorizing the items and adopt
mechanisms that have variable levels of control on the different categories of items.
Consider two items in inventory, a simple fastener that costs less than a rupee and a
specialized electronic sub-assembly that costs several thousands of rupees. Managing the latter
item with close monitoring and tighter control will benefit the organization more than exercising
greater managerial attention and control on fasteners. If there were errors in judging the demand
for the item, or if extra items were ordered, the impact of these factors will be minimal in the
case of fasteners. Also, if there were shortages, the response time to procure fasteners will be
very low. On the other hand, in the case of the electronics sub-assembly, the impact will be
significant. It is therefore obvious that managers need alternative methods and levels of control
while dealing with a multitude of items in the inventory. Selective control of inventory achieves
this objective.
ABC Classification
It is clear from the arguments outlined earlier that the unit cost or total cost of items consumed
could be one basis on which items could be classified. The ABC classification of inventories is
based on the cost (or value) of items consumed. Very high-value items are “A-class items” and
may require tighter control. Medium-value items are categorized as “B class” and low-value
items as “C class”.
TABLE 17.5 Sample Data for ABC Classification
Let us consider an example of twenty items in inventory, as shown in Table 17.5. For the twenty
items, the unit value and annual consumption are available. Using this, it is possible to compute
the cumulative percentage of items and cumulative value of consumption (see the last two
columns of Table 17.5) and rank them in the descending order of cumulative value. Figure 17.9
shows a graphical representation of this data.
As shown in Table 17.5, 5–10 per cent of the items typically contributes to 70–80 per cent of
the value and constitutes A-class items.8 These items require greater control and close
monitoring. By virtue of constituting a small percentage of the total in terms of volume, it will be
practically feasible to exercise close monitoring and greater control. At the other extreme, about
60–70 per cent of items constitute about 5–10 per cent of the total value. These are the C-class
items. These items require simple control mechanisms such as level-based reordering. The
middle portion of about 10–25 per cent by value and about 20–30 per cent by number of items
constitute B-class items. These items require a moderate level of control.
The periodic review and the continuous review systems of inventory control can be linked to
the category of items. Since A-class items require closer control and better response to changes
in the demand pattern, periodic review systems are more appropriate. In the case of “B” class
items, continuous review systems are appropriate. “C” class items can have simple level-based
rules for inventory control. C-class items are often available readily off the shelf and it is
possible to obtain them by ordering over the phone. Hence, issuing blanket purchase orders for a
year and following up with specific requests for supply against the purchase order is often
practiced in the case of C-class items.
FIGURE 17.9 ABC classification of inventories
XYZ classification is based only on the unit cost, whereas ABC classification takes into account the
consumption pattern as well.
2. On the basis of movement of inventory (FSN classification): (a) fast-moving, (b) slow-moving, (c) non-moving.
This method of classification is ex-post, as opposed to ABC classification, which is future oriented. Items that have
not been moving for sometime incur carrying costs and may call for managerial attention for disposal decisions. On the
other hand, fast-moving items can be controlled using available inventory control systems.
3. On the basis of criticality of items (VED classification): (a) vital, (b) essential, (c) desirable.
This classification is relevant in the case of maintenance items. In several cases a vital item for maintenance may not
be very expensive. For instance, a wide variety of oil seals are used in hydraulic systems. If the oil seals are not
available, the entire equipment becomes inoperative and has a ripple effect on the entire system. However, by virtue of
having smaller unit value or lower consumption value, these items may escape the attention of the management without
this classification scheme in place.
It is often possible to combine more than one classification scheme and make use of them to further sub-
categorize the inventory and devise appropriate inventory control system for each of them.
4. On the basis of sources of supply: (a) imported, (b) indigenous (national suppliers), (c) indigenous (local suppliers).
Imported items are usually of high value. Moreover, they have long lead times owing to several statutory and
procedural complexities involved in the buying process. Therefore, they may require tighter control. On the other hand,
items procured from the local suppliers are available “off the shelf”.
So far we have considered the problem of inventory planning in the case of recurring demand
that requires continuous replenishment. However, there are a number of situations that require
inventory planning for a single-period demand. In a single-period demand, the unfulfilled
demand cannot be back-ordered to the next period because the demand ceases to exist after the
period for which planning is done. In other cases, even though demand exists in the future, what
is ordered for a period cannot be used for future periods due to the perishable nature of the item.
Examples include the demand for morning newspapers, tickets for journeys, advertising space
for a mega entertainment event, and expensive maintenance spares. This type of inventory
models are referred to us news vendor model. Often, there is a high degree of uncertainty
involved in estimating the demand for a single period.
In a single-period demand, the unfulfilled demand cannot be back-ordered to the next period.
Planning for appropriate levels of inventory in such situations requires careful balancing of
two opposing costs. The main driver of costs in this inventory planning is the fact that uncertain
demand manifests only for a period. Therefore, carrying lesser inventory than demand directly
results in lost opportunity to make profit. This represents the cost of understocking. In the same
way, any excessive inventory cannot be consumed afterwards. At the most, the unused inventory
may fetch some salvage value and the resulting loss on account of this could be termed as the
cost of overstocking.
If d > Q, then we incur costs on account of cost of understocking. On the other hand, if d < Q,
then we incur the cost of overstocking. At a very low value of Q, we tend to experience costs
arising out of understocking and as we increase Q incrementally we will approach optimal Q. At
very high values of Q, we will incur overstocking costs. By incremental analysis, we find that
while taking a decision to stock Q units, we would like to ensure that:
The expected cost of overstocking ≤ The expected cost of understocking
Therefore, we choose that largest value of Q that satisfies Eq. 17.11 as the optimal value.9
EXAMPLE 17.6
Navratri is a popular festival in India. In South India, beautifully painted dolls made of clay
are bought by customers during this festival. After the festival, there is no demand for these
dolls. A manufacturer of dolls needs to decide on the optimal stock of dolls that he needs to
carry in his inventory to satisfy the demand during the festival time. The item fetches a sales
value of ₹1,300 per box. The cost of production is ₹1,000 per box. After the festival is over,
the items at best can be salvaged at a value of ₹800 per box. Table 17.6 presents the
distribution of demand for the item during the festival time. What is the optimal quantity to
stock?
TABLE 17.6 Distribution of demand for Example 17.6
Solution
Since each unfulfilled demand results in a foregone profit, the cost of understocking is the
profit per box. Similarly, by salvaging each unsold box after the festival, the manufacturer
loses an amount equal to the difference between the cost of manufacture and salvage value,
which represents the cost of overstocking.
On examination of the cumulative probability values in the last column of the demand table,
we notice that Q lies between 200 and 300. We round up the value of Q to 300. Therefore,
the manufacturer should plan for an inventory of 300 boxes for sale during the festival.
In practice, inventory control systems vary from theoretical formulations in many ways.
Moreover, there is a lack of understanding of certain critical issues that influence inventory
management. One serious drawback of inventory control systems in practice is that they tend to
stand aloof from reality. Any inventory control system should have provisions for linking key
parameters such as reorder level and safety stock to actual changes in consumption pattern,
measured in terms of mean and standard deviation of demand. In the absence of this linkage, the
model parameters become outdated and result in either piling up of inventory or in frequent
shortages.
With the installed base of ERP systems and other MIS systems for transactional data capture,
it is possible to upgrade computerized inventory control systems from an ordinary MIS system
generating a host of shortage reports and accurate inventory status records to one of a decision-
support system. On a periodic basis, these computerized systems could analyse the consumption
patterns of items, detect impending changes, and adjust the model parameters of the inventory
control system.
Better management of inventory does not happen only because of good inventory planning
tools. Organizations need to take several other measures to bring down inventory. A frequently
encountered issue is part number proliferation. Over the years, an organization creates a vast
number of part numbers and managing such a large variety of parts in inventory could be a
herculean task. Efforts to reduce part numbers by variety reduction measures are an important
aspect of better inventory management. Similarly, an organization having several
manufacturing/service divisions must engage in inter-divisional joint planning efforts towards
standardization of classification and coding systems and stock holding of expensive items such
as capital spares. These efforts go a long way in better inventory control.
Opportunities for better inventory control and management often exists outside of the
materials management function in organizations. For example, alternative methods of launching
material onto the shop floor, such as kit launching, helps reduce inventory mismatch and non-
moving inventory. In a kit launching system, items required for manufacture and assembly are
always launched into the system in exact quantities of ship sets or assembly sets. This reduces
the chances of ending up with excess and unwanted work in progress at the end of the planning
period. Another example relates to performance measurement systems in organizations.
Management control systems that emphasize on utilization-based measures for rewards and
incentives tend to promote inventory build-up in the system. As shown in Chapter 13, better
structures and systems also contribute positively to reducing inventory build up in the system.
Opportunities for better inventory control and management often exist outside of the materials management
function in organizations.
An inventory planning manager should take all these into consideration while actively
pursuing the models suggested in this chapter. Such an approach will ensure greater chances of
reducing inventory in the system.
SUMMARY
Every organization carries five different types of inventory. This includes cyclic stock, pipeline inventory, safety
stock, decoupling inventory, and seasonal inventory. The purpose of carrying these types and the manner to assess
the level of investment required in each of these vary.
Inventory planning is done in order to minimize the total cost of the plan. The costs include the unit cost of the item
for which planning is done, the cost of carrying inventory, the cost of ordering, and the cost of shortages.
The key decisions in any inventory planning scenario is to answer the “how much” and the “when” questions.
These decisions are made by balancing various costs associated with inventory.
The EOQ model is useful for inventory planning in the case of multi-period deterministic demand situations. The
EOQ model is robust to model parameters and could be suitably modified to incorporate some real-life situations
such as quantity discounts and non-zero lead time for supply.
Service level is a useful concept for modelling inventory planning in the case of stochastic demand. Safety stocks
can be built commensurate to the desired service level.
A fixed order quantity (Q system) or continuous review system of inventory planning and control is useful for B-
class and C-class items of inventory. A popular application of the continuous review system in organizations is the
two-bin system.
A fixed order interval or a periodic review system (P system) is useful for planning and control of high-value or A-
class items. The P system is more responsive to changes in demand patterns than the Q system.
Selective control of inventories is achieved through alternative classification methodologies. The ABC, VED, and
XYZ classifications are often used by organizations.
FORMULA REVIEW
Total cost,
REVIEW QUESTIONS
2. How is inventory planning for independent demand items different from that for dependant demand items?
3. Do organizations need to carry inventory? Why?
4. What is the relationship between inventory investment and profitability in an organization?
5. Consider the inventory planning problem. Write down the relevant total cost equation in each of the following situations:
a. The organization can fetch a discount on the purchase price if large quantities are ordered
b. There is a benefit in transportation costs if ordered in full truckloads
c. The supplier is willing to set up elaborate systems for easy ordering, monitoring, and follow-up of orders if the
order quantity is large
d. There is uncertainty in demand and the organization wishes to carry safety stock
6. What are the difficulties in using shortage cost measures in inventory planning models?
7. Can you identify four alternatives that organizations use as a basis for fixing the order quantity?
8. On what basis would you recommend the periodic review system of inventory control?
9. Why should organizations adopt a selective system of inventory control? If you were asked to recommend a suitable
classification scheme, how would you go about the task?
10. What is the relationship between the service level and safety stock?
11. There are N locations at which an organization faces demand for an item. The demand for the item is uncertain and
hence the organization needs to carry safety stock. If it were to make the choice between centralized storage of the item
and localized storage of the item in each of the location, which alternative will result in lower investment in inventory?
(Hint: Assume that all locations have the same mean and standard deviation for demand).
12. What is the effect of the cost parameters on the economic order quantity? Perform a “what if” analysis using a
spreadsheet by varying these cost parameters. What did you deduce from the results?
13. When is it appropriate to use the ABC classification scheme and the FSN classification scheme?
14. Suppose you develop a classification scheme for selective control of inventories using ABC and VED classification as
two dimensions. Relate the discussions pertaining to service level to this two-dimensional classification scheme.
PROBLEMS
1. Consider an item that is ordered once per month. The daily requirement is 200 and the lead time for supply is two days.
There are 25 working days in a month. The cost of ordering is ₹300 per order and the cost of carrying is ₹150 per unit per
year.
a. Draw a sketch showing the cyclic and pipeline inventory in the system
b. What is the reorder point for this item?
c. What will the cost of this plan be?
2. An auto-component manufacturer requires a certain steel forging in large quantities. The annual requirement is 40,000
pieces, each costing ₹450. The ordering cost is ₹600 per order and the carrying cost is ₹100 per unit per year.
3. Consider Problem 2. Perform a “what if” analysis for the following scenarios and compute the total cost of the plan for
ordering an optimal order quantity:
a. The actual demand is −12 per cent, −8 per cent, −4 per cent, 0 per cent, 4 per cent, 8 per cent, and 12 per cent of the
demand mentioned in the previous problem.
b. The actual ordering cost has a similar variation as that of the demand
c. The actual carrying cost has a similar variation as that of the demand
Plot the total cost in each of the three scenarios and make your observations.
4. Excel Toys, a manufacturer of plastic toys for children, requires constant supply of high-density polyethylene (HDPE) in
the resin form. HDPE is available in the resin form and one tonne costs approximately ₹2,000. The manufacturer runs his
factory for 250 days a year and the daily consumption rate for HDPE is four tonnes. The suppliers normally take a week
to replenish an order. The ordering cost is ₹2,000 per order and the carrying cost is 20 per cent.
a. Estimate the cyclic and pipeline inventory in the system at Excel Toys.
b. What is the total cost of the plan?
c. Suppose if the supplier insists on a minimum order quantity of 75 tonnes, what is your recommendation to Excel?
Will the recommendation change if the minimum order quantity is 150 tonnes?
d. Suppose Excel launches some improvement initiatives in the company and brings down the ordering cost by 20 per
cent. Recompute the economic order quantity.
e. If Excel spent a sum of ₹10,000 towards improvements in the system, when will the improvement efforts bring
payback? (Hint: Estimate the savings in costs due to reduced ordering cost)
5. Consider Example 17.2 in this chapter. If the company continues with the same plan even when there is an increase in the
demand by 10 per cent, what will the impact be? What do you infer from the result?
6. A manufacturing organization has been consuming a certain item in large quantities and is currently procuring the item
from Supplier A. The price offered by the supplier is ₹400 per piece. The ordering cost is ₹2,800 per order and the
carrying cost is ₹350. The annual demand for the item is 10,000. The supplier is currently not offering any discount.
However, another supplier, Supplier B, is willing to offer the following discount structure:
Up to an order size of 999 = No discount
For an order size of 1,000–1,999 = 2 per cent discount in price
For an order size of 2,000 and above = 5 per cent discount in price
Switching over to this supplier means incurring an initial cost of ₹15,000. This cost is primarily to set-up new
communication systems with the new supplier. What should the company do in the light of the new offer?
7. Siemens India Limited manufactures switchgears at their Kalwa plant. One of the key components, Kombishraube
(contacts), is manufactured in-house. If the machine producing these contacts manufactures at the rate of 800 pieces per
hour and the demand rate is 450 pieces per hour, what is the economic run length for this production process? How often
should they set up the machine if the factory works for 250 days in a year on a single shift (of 8 hours) basis? The cost of
setting up the machine is ₹3,000 per set-up and the cost of carrying the components is ₹400 per unit per year.
8. Consider an item with the following demand attributes:
Mean daily demand = 20; Standard deviation of daily demand = 8.
Compute the safety stock required for a continuous review system in each of the situations given in Table 17.7:
What do you infer from these calculations?
9. For Problem 8, compute the order up to levels and safety stock for a periodic review system with a two-week review
period.
10. The weekly requirement of furnace oil in a foundry has a mean of 3,000 litres and a standard deviation of 900 litres. A
litre of furnace oil costs ₹400 and the carrying costs are estimated at 25 per cent. If the foundry uses a continuous review
system of inventory control with a desired service level of 90 per cent, then
a. What will the reorder point and safety stock for furnace oil be if the lead time for supply is 1 week?
b. What will the service level be if the reorder point is changed to 3,100 litres?
c. If the reorder point is left unchanged when there is a 5 per cent increase in the mean demand, how will it affect the
service level?
d. If the foundry estimates that the monetary value of the benefit that it may obtain is ₹3,000 when increasing the
service level from 90 per cent to 95 per cent, what is the appropriate service level for the item?
11. The demand for springs in the coil form for an auto component manufacturer is continuous and uncertain. The
manufacturer places orders for the springs with a local supplier who takes two weeks to deliver. The cost of the spring is
₹400. The weekly demand for the springs has a mean of 500 and a standard deviation of 100. The cost of ordering is
₹2,000 per order and the cost of carrying is 18 per cent. The supplier specifies a minimum order quantity of 2,000.
Assume a service level of 90 per cent.
12. Oriental Healthcare is a multi-specialty hospital catering to a variety of illnesses connected to the heart and respiratory
system. The demand for a class of medical consumable is generally random. Initially, Oriental followed a practice of
ordering 200 boxes every two weeks. This practice was not found to be satisfactory. After some review, they resorted to
ordering 300 boxes every two weeks. Even this practice was not found satisfactory. Recently, an examination of the stores
records over a period of 10 weeks revealed the following weekly consumption pattern (Table 17.8):
The supplier of the item takes (on an average) two weeks to deliver once the order is placed. The item costs ₹450 per box.
The ordering cost is ₹2,000 per order and the carrying cost is 20 per cent. What should Oriental do in the light of this
information? Will they be better off with the new recommendation? Assume a service level of 90 per cent.
13. A pharmacy needs to decide on monthly procurement quantity of a critical yet a highly perishable drug that has a shelf-
life of just one month. The unit price of drug is ₹ 7,000 and its cost is ₹6,000. As per the historical data, the monthly
demand for the drug is random with a certain distribution. The details are given in Table 17.9. Since the drug is an
emergency drug, the associated hospital offered to reimburse ₹3,000 for each unit of drug left unsold. At the same time,
the store would be penalized ₹2,000 per unit by the hospital for any unmet demand.
What should be the monthly order size the pharmacy must plan for?
14. In problem number 13 consider the following. The hospital is contemplating a change in the contractual terms with the
pharmacy. Suppose the hospital is willing to make an offer of either withdrawing the reimbursement or penalty. What
should the pharmacy owner do now?
15. A manufacturer of electric ovens has made an estimate (shown in Table 17.10) for annual consumption and unit cost of
the components that are used for manufacture. Perform an ABC analysis and advise how the manufacturer should plan
and control for inventory.
1. Identify a sector of industry of your choice and select three companies operating in that sector. Analyse their annual report
and identify portions of the annual report that have a direct impact on inventory-related issues.
a. Prepare a report highlighting your understanding of the performance of the companies with respect to inventory
planning and control. Can you identify some possible explanations for your observations pertaining to the
inventory data?
b. Among the three, which one is better in managing their inventory? How do these firms compare with the industry
average?
c. Can you make some comparisons with similar firms elsewhere (outside the country)?
d. What are your key inferences from this exercise?
2. Study the reports produced by ICICI (Financial Performance of Companies) and the Centre for Monitoring Indian
Economy on performance of corporate sector.
a. Identify the inventory investment across different sectors of our industry. Do you see any significant trends in these
movements?
b. How has the performance been in the last five years? What factors do you think might have contributed to this?
c. Which sectors have performed better and which have performed the worst? Can you identify some reasons for
these?
3. Analyse the multiplier effect of inventory using the Du Pont equation for return on investment for a pair of close
competitors.10 What are your key inferences from this exercise?
CASE STUDY
MML Limited
The Background
MML Limited was incorporated as a private limited company and later converted into a
public limited company in 1988. Though the company started with the manufacture of soaps
and perfumes, it has added a variety of products to diversify its product line. Today, it is a
conglomerate with businesses in soaps, perfumes, plastics, petrochemicals, paints, industrial
electronics, and agro-business. Their unit at Hosur manufactures electronic components. The
inventory position has been deteriorating over the last few years. The year-end inventory
position, as revealed by the annual reports of the conglomerate, showed an increase from
₹1,230 million as on 30 June 2004 to ₹2,490 million as on 31 March 2007.
The conglomerate has been operating in a highly centralized fashion. All major planning
activities were done at the corporate office. However, many inputs for this decision making
were made available by all the operating units. The headquarters sent the tentative monthly
production plan two months prior to the commencement of production. The final production
plan arrived 5 days prior to start of the production. It was not unusual for the corporate office
to change the production plans while production was progressing. This was communicated
through telephone, telex, fax, or mail. Once the production plan was made, the requirement
of each item was also computed using the bill of materials. Table 17.11 presents the tentative
and final plans, and the actual production of the Hosur unit for 2005–06 and Table 17.12
presents the consumption pattern of the “A-class” item.
TABLE 17.11 Tentative Plan, Final Plan, and Actual Production for Some Products
Notes:
*The lead time is an average figure based on six months’ data.
The consumption details are expressed in units of items. The units of measurement are different for the items. For example, the unit of
measurement for Item 3 is boxes (each box contains 250 items) whereas the unit of measurement for Item 8 is kg.
Item 6 is sold by all the suppliers in integral units of 250 only. This has been attributed to the fact that once the seal is broken it should be used
within three days.
Items 7 and 10 are supplied by the same supplier, who is the only available domestic source.
3. In case of sudden shortages of an item at a location, the corporate office could advise another location to transfer stocks
because it knew the safety stocks at each location.
The corporate office advises the purchase departments at the units on the following:
1. What to order?
2. How much to order?
3. When to order?
4. On whom to place the order?
5. What should be the delivery date?
6. What will be the price?
On receipt of this, the purchase department at the operating units took the responsibility for
the procurement of items on the basis of the advice and subsequently made the payment to
suppliers.
For the corporate office to take decisions on these items, it needed the following:
1. Stock levels of each item under inventory management.
2. The production plan for the following month.
3. The maximum and the minimum stock levels to be maintained for each item.
4. The previous month’s consumption.
The operating unit provided the information pertaining to Item 1. Items 3 and 4 were policy
decisions made at the corporate office based on historical data and future plans and Item 2
was market-related information that was forecast.
Notes: * All values are yearly expenditure in rupees; ** All numbers are expressed as percentages. For limited purposes
many departments have been combined with general administration.
Recent Developments
Recently, the company took a decision to decentralize most of the decision-making
processes. Under this scheme, each unit was focused on certain areas of the business. The
individual unit heads entered into an agreement with the corporate office regarding the target
for sales turnover, profitability, and a few other broad parameters. The unit took all
operational decisions with the required autonomy.
A study group was constituted at the Hosur unit to analyse the various expenditures
incurred at the plant level. The unit head initiated this as soon as decentralization was put
into effect. Until recently, such statistics was not compiled and analysed at the unit level. The
units would merely send a host of weekly and monthly reports to the corporate office. The
corporate staff performed the task of analysing the data and informing the various unit heads
regarding the variances from budgetary provisions. A portion of the cost data gathered so far
by the study group is given in Table 17.13. Certain service items such as forklift truck
maintenance, insurance premiums freight, and demurrage were allocated to specific
departments using the basis developed for costing purposes. Table 17.13 also shows the
apportionment details for the items.
SUGGESTED READINGS
R. B. Chase and N. J. Aquilano, Production and Operations Management: Manufacturing and Services, 9th edition
(Boston: Irwin, 1999).
M. A. Cohen, Y A. Zhang, and Y. Wang, “Identifying Opportunities for Improving Teradyne’s Service Parts
Logistics System,” Interfaces 29, no. 4 (1999): 1–18.
H. D. Gupta, Inventory Control in Public Sector (New Delhi: Anmol Publications, 1991).
R. W. Hall, Zero Inventories, (Homewood: Dow Jones, Irwin, 1983).
A. C. Hax and D. Candea, Production and Inventory Management (Englewood Cliffs, NJ: Prentice Hall, 1984).
P. C. Jones, G. Kegler, T. J. Lowe, and R. D. Traub, “Managing the Seed Corn Supply Chain at Syngenta,”
Interfaces 33, no. 1 (2003): 80–90.
S. Nahmias, Production and Operations Analysis, 3rd edition (Boston: McGraw-Hill International Editions, 1997).
N. V. Reinfled, Handbook of Production and Inventory Control (Englewood Cliffs, NJ: Prentice-Hall, 1987).
E. A. Silver, D. F. Pyke, and R. Peterson, Inventory Management and Production Planning and Scheduling, 3rd
edition (New York: John Wiley, 1998).
M. K. Starr and D. W. Miller, Inventory Control: Theory and Practice (Englewood Cliffs, NJ: Prentice Hall, 1990).
CHAPTER 18
Operations Scheduling
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
“Shop configuration” refers to the manner in which machines are organized on the shop floor and the flow pattern of the jobs
utilizing these machines. Two alternative configurations—flow shop and job shop—are available for configuring machines in a
manufacturing system.
ideas at Work 18.1
Operations management requires planning and control at various levels and time horizons. So
far, we have discussed operations management issues in the long term and the medium term.
Configuration of an operations system at the right location and choice of appropriate process and
product designs are issues associated with the long term. On the other hand, aggregate planning
and material and capacity requirements planning are issues associated with the medium term. In
the same manner, there is a host of issues that one needs to address in short-term planning. In this
chapter, we focus on planning and control issues associated with the short term. In the short
term, there is more emphasis on operational control than on planning. This, however, does not
imply that there is no planning involved. In fact, planning does half the job of control.
“Short term” denotes a time horizon close to real time. Typically, in a majority of
manufacturing and service organizations, this would imply a horizon ranging from a day to at
most a week or two. In some special cases, “short term” may mean a few hours or a shift, as in
the case of a power transmission firm or an event management firm. One question that comes to
our mind is: When we have exercised so much care in the previous stages of planning, why do
we still need to continue with the planning exercise? Before we proceed with the concepts and
techniques of operations scheduling, it is best to clarify this issue. Planning for the short term
becomes inevitable for three important reasons:
1. As we approach real time, several kinds of additional information become available to an organization. Use of this new
information makes the plans far more robust and representative of reality. For instance, some customer orders could be
cancelled, some new orders booked, and the terms of the existing orders could be revised. Obviously, using this
information makes operations more accurate than choosing to simply ignore them.
2. The occurrence of random events is inevitable in business. These include the sudden breakdown of a machine,
absenteeism of skilled labour, delays in the supply of key raw material, and a sudden revision of job priorities. It is not
possible to include these details accurately while planning for the medium term. Broad assumptions and approximations
are made about these at that time of planning. Therefore, we need to revisit these assumptions and fine-tune our
planning and decision-making methodology.
3. In the short term, we need to focus on micro-resources, a single machine, a set of workers, and so on. Such a focus is
neither possible nor warranted in medium or long-term planning.
In the short term, several questions need to be answered. How can jobs be assigned to various
work centres? Within each work centre, how can the jobs be rank-ordered? How can other
resources such as skilled workers and special gadgets be assigned to the operating system? How
do we react to a breakdown in the system? How can the performance of the operating system be
measured? All these questions are equally applicable to both manufacturing and service systems.
In this chapter, we address many of these questions and develop methodologies to address these
questions using a defined framework. We refer to the collective framework for addressing these
issues as scheduling of operations.
Operations scheduling uses a defined framework to address issues associated with the use of available
resources and the delivery of products and services as promised to the customers.
Operations scheduling has a direct linkage with MRP systems in an organization. Recall that
the function of an MRP exercise is to generate feasible production orders and procurement
notices. Production orders are nothing but specific sets of instructions to utilize a set of
resources for the manufacture and/or assembly of the components under question during a
particular period of time. Consider a sub-assembly, X. If the MRP schedule indicates that during
Week 16 a designated shop, S1, should manufacture 200 units of X, then it is clear that during
Week 16, the planner needs to ensure that the job is assigned to Shop S1 and provide capacity for
manufacturing 200 units of X. During the same week, there could be several jobs assigned to
Shop S1. Therefore, the shop planners also need to prioritize all such jobs in the shop with
respect to each machine they visit, and ensure completion of the job without delay. This example
indicates that the output of an MRP exercise becomes the input for job scheduling.
18.2 SCHEDULING: ALTERNATIVE TERMS
In practice and in literature, several terms are used in scheduling. These terms convey important
concepts related to scheduling. Yet, those unfamiliar with the subject are often confused by these
terms and experience considerable difficulty in understanding their significance. We will,
therefore, define these terms here and use them consistently in the rest of the chapter.
Planning-related Terms
Planners often use a pair of terms—loading and scheduling. Loading is defined as a planning
methodology that is used to assign an adequate number of jobs to the resources in an operating
system during a planning horizon (of, say, a week). Loading is done to ensure that the resources
in an operating system are utilized in the best possible way. Moreover, loading also ensures that
the workload among competing resources is not lopsided, and hence does not create an uneven
strain on the operating system. In a dental hospital, let us assume that there are three dental
specialists visiting the outpatient department for four days. On each day, let us assume that they
spend three hours advising or treating patients. The loading function will ensure that the
specialists are adequately engaged and that no specialist is grossly over or underutilized. (We
make the important assumption here that the patients do not make the choice of the consultant. If
the patient has a choice, then the loading methodology will suggest shifting appointments from a
particular day to another day or that the patient consult another specialist.)
Loading is a planning methodology that is used to assign an adequate number of jobs to the resources in an
operating system during a planning horizon.
Scheduling is defined as the process of rank-ordering the jobs in front of each resource with a
view to optimize some chosen performance measure. If there are n jobs waiting in front of a
machine, using scheduling, the planner could rank-order them for the purpose of processing.
Rank-ordering could be done to minimize the waiting time of all the jobs in the queue. In the
dental hospital example, if it is decided to allow 10 patients for consulting during a three-hour
schedule, some rules are utilized to schedule the ten patients one after the other. In service
systems, the rule is often first-cum-first served. But in manufacturing systems, several other
possibilities exist.
Scheduling is the process of rank-ordering the jobs in front of each resource to maximize some chosen
performance measure.
Routing is the order in which the resources are used by jobs for processing.
Sequencing is the ordering of the operations of the jobs in the operating system. It combines
the routing and scheduling information to provide a basis for the planners to ensure that jobs are
processed in the shop without violating technological constraints.
Sequencing is the ordering of the operations of the jobs in the operating system.
Administration-related Terms
In organizations, dispatching and expediting are terms frequently associated with the scheduling
of operations. Dispatching is defined as the administrative process of authorizing the processing
of jobs by resources in the operating system, as identified by the scheduling system. The
authorization could be in the form of a shop order or information communicated electronically to
the respective work centre. Dispatching physically signals the assignment of jobs to work centres
and places, and the onus for reviewing the progress is on the controllers. Expediting, on the other
hand, is the administrative process of reviewing the progress of the job in the operating system
and ensuring that it travels through various stages of the process as indicated in the routing,
without harming the performance measure. If there are deviations and delays, it calls for pushing
the job forward through the system.
EXAMPLE 18.1
A manufacturer of dies for the automotive component industry utilizes a job shop structure.
During a planning horizon, the manufacturer needs to first assign five jobs to the five
machines that are in one section of the factory. Table 18.1 provides details on the processing
time of the jobs in each of these machines if assigned. The differences in the processing time
are attributed to differences in the ratings of these machines. Use the assignment method to
solve the loading problem.2
Solution
We use the standard assignment method for solving the problem. We first subtract each row
by the minimum number in the row. The resultant table is as shown in Table 18.2.
After that, we subtract each column by the minimum number in the column. There is a
change only in Column 2 as other columns already have a zero. The table after column
subtraction is as shown in Table 18.3.
Since the minimum number of lines required to cover all the zeros is equal to the number
of rows, we can use Table 18.3 to perform the final assignment. The assignments are as
follows:
Job 1 assigned to Machine 2;
Job 2 assigned to Machine 1;
Job 3 assigned to Machine 3;
Job 4 assigned to Machine 5;
Job 5 assigned to Machine 4.
This assignment results in a minimum total processing time of 80 time units.
TABLE 18.1 Processing Times for the Jobs
The nature and the complexity of the scheduling problem are dependent on the context in which
we solve the problem. Moreover, past research suggests that even the type of performance
measure that we use for scheduling jobs can influence the ease with which we could obtain the
solution. Therefore, a good introduction to the parameters of the problem is essential before we
analyse the solution techniques available for scheduling. The following parameters adequately
describe the scheduling problem that one tries to solve:
The nature and the complexity of the scheduling problem are dependent on the context in which we solve the
problem.
Number of machines: The number of machines, or more generally, the number of resources for which the planner
proposes to perform the loading and scheduling function is the first parameter for the problem. Let us denote the number
of machines by m. The simplest case is m = 1. In this case, the jobs that require processing in that machine need to be
simply rank-ordered on some basis. However, as m becomes larger, the problem size increases as numerous
combinations are introduced. Moreover, if multiple machines are available for the jobs as parallel resources for
processing, we need to develop methods by which one out of those multiple machines is chosen for assigning the job.
As the number of machines and jobs increases, the problem size also increases as numerous combinations
are introduced.
Number of jobs: The number of jobs is the second parameter for the scheduling problem.
We denote the number of jobs by n. If the number of jobs increases, it introduces more
complexity into the scheduling problem. As multiple jobs compete for the same set of
resources, additional methods are required to assign the jobs to the resources.
Shop configuration: “Shop configuration” refers to the manner in which m machines are organized on the shop floor and
the flow pattern of the jobs utilizing these machines. Two alternative configurations—flow shop and job shop—are
available for configuring machines in a manufacturing system.3 Since these configurations differ in terms of the
deployment of resources, the scheduling methodology also differs. Moreover, we shall also see that the complexity of
the problem varies between these two configurations.
The simplest configuration is flow shop. In a flow shop, the resources are organized one after
the other in the order the jobs are processed. In a flow shop, jobs start their processing at
Machine 1. After completing processing, they visit Machines 2, 3, and so on before they exit the
system from Machine m. Figure 18.2 graphically portrays a pure flow shop. A pure flow shop is
one in which all the jobs visit all the machines in the same order (beginning at Machine 1 and
ending at Machine m). However, in a mixed flow shop, some jobs are allowed to skip machines
in between.
The simplest shop configuration is flow shop. In a flow shop, the resources are organized one after the other in
the order the jobs are processed.
Since in a pure flow shop all jobs follow the same order of visiting the machines, the
scheduling function is essentially reduced to one of ordering the jobs in front of the first
machine. Once the jobs are ordered in front of the first machine, it follows the same order in the
rest of the machines. Since there are n jobs, there are n! ways in which one can draw up
alternative schedules in the shop. It is easy to note that scheduling complexity increases in the
case of a mixed flow shop.
The second configuration is the job shop. In a job shop, machines are not organized in any
processing order. Rather, similar types of resources are grouped together. Every job follows a
unique order in which it visits the machines for processing. Moreover, there is no requirement
for all jobs to visit a particular machine for their first operation or a particular machine for their
last operation (as in the case of a flow shop). Figure 18.3 presents a simple illustration of a job
shop consisting of seven machines. In this example, the route that Job 1 follows for processing is
1−4−2−5−6. On the other hand, Job 2 follows the sequence 3−2−1−4−6−7 and Job 3 follows
2−3−4−7−5−6.
In a job shop, machines are not organized in any processing order. Rather, similar types of resources are
grouped together.
Since each job has its own unique route, the set of jobs visiting each machine is not the same.
Therefore, the ordering of jobs in front of each machine is to be separately determined. There are
n! ways of rank-ordering jobs in front of a machine. Since there are m machines in the shop, the
number of alternative schedules that one can draw for a job shop is given by (n!)m. Clearly, the
number of alternatives is far too high for one to complete by enumerating these alternatives and
arriving at the best schedule.
Scheduling Rules
A critical parameter for the scheduling problem is the basis on which competing jobs queueing in
front of a machine are rank-ordered. Alternative rules can be used for establishing priorities.
These rules are known as scheduling rules. The following are the most frequently used rules:
Shortest processing time (SPT): This rule simply chooses the job with the least processing time from the competing list
and schedules it ahead of the others. For example, let us consider four jobs waiting in front of a machine for processing.
If the processing times (in minutes) are 12, 9, 22, and 11, then the order in which the jobs will be scheduled using the
SPT rule is Job 2–Job 4–Job 1–Job 3.
Longest processing time (LPT): This rule is the reverse of the SPT rule. The job with the longest processing time is
scheduled ahead of other competing jobs. For our example, the order in which the four jobs are scheduled is exactly in
reverse (Job 3–Job 1–Job 4–Job 2).
Earliest due date (EDD): It is possible to establish priorities on the basis of the due date for the jobs. The logic here is
that a job that is due tomorrow needs to be scheduled ahead of another that is due the day after even though the second
job may require lesser processing time compared to the first. Therefore, jobs can be rank-ordered on the basis of due
dates. If the chronological order of the due dates of the four jobs is Job 4–Job 3–Job 2–Job 1, then EDD rule will rank
the jobs in that order.
Critical ratio (CR): The SPT and LPT rules take only the processing time information into consideration. On the other
hand, EDD considers only the due dates of the jobs. In reality, it will be desirable to consider both while scheduling. A
job with a long processing time due in the near future requires greater attention than a job with very little processing
time but due at the same time. The critical ratio estimates the criticality of the job by computing this information using a
simple ratio:
A smaller CR value indicates that the job is more critical. Particularly, if CR <1, it denotes that the available time is not
sufficient and the job is already running late. On the contrary, a larger CR value (CR >1) indicates that some slack is
available for the job.
First-come-first-served (FCFS): This rule schedules the jobs simply in the order of job arrival. There will be no
consideration of the processing time or any other information such as the due date of the jobs. This is akin to the
working of Indian Railways’ computerized passenger reservation system in various cities.
Random order (RAN): It is also possible to assign priorities on a random basis. Suppose we get four random numbers
from a random number generator, as follows: 563, 125, 490, and 923. Then one can use this set of numbers for rank-
ordering the four jobs (on the basis of ascending order of the three digit numbers). The resulting order will be Job 2–Job
3–Job 1–Job 4.
Table 18.4 illustrates these five rules for the four-job example. In reality, it is possible to
construct alternative scheduling rules on the basis of the specifics of the problem that one is
trying to solve.4
TABLE 18.4 Illustration of the Scheduling Rules (Current Time = 0)
Performance Criteria
The scheduling problem needs to be solved for the purpose of meeting certain characteristics in
the operating system. Therefore, the performance criterion is an important parameter for the
scheduling problem. Three types of performance criteria are normally considered in operations
scheduling. These include completion-based, due date-based and inventory-/cost-based
measures.
Three types of performance criteria are normally considered in operations scheduling: completion-based, due
date–based, and inventory-/cost-based measures.
Completion-based measures
In completion-based measures, an assessment of the time taken to complete the operations is
made and used for the purpose of evaluating the goodness of the chosen scheduling rule. Two
measures are frequently used: flow time and make span. Flow time is the elapsed time between
releasing a job into the shop and the time of completion of the job. Flow time equals the sum of
processing time of all operations of the job and other non-productive time, including waiting and
moving in the shop at various work centres. Consider the following notations for Job i:
Flow time is the elapsed time between releasing a job into the shop and the time of completion of the job.
Alternative variations of these measures can be computed. For example, maximum flow time,
mean flow time, maximum completion time, and mean completion time can also be computed
for the set of jobs under consideration. However, the maximum completion time is denoted by
another measure known as make span.
Make span is the time taken to complete all the jobs released into the shop for processing.
Even when all the jobs are released at the same time, jobs come out of the shop at different times
due to time differences in processing. Therefore, make span is the completion time of the last job
in the shop.
Make span is the time taken to complete all the jobs released into the shop for processing.
Flow time, also known as throughput time, is an appropriate measure for jobs that are
continuously launched on the shop floor. It provides a good measure of the length of the time the
job resides on the shop floor. This is directly related to the work-in-progress inventory likely to
be accumulating in the shop. On the other hand, if the shop launches batches of components and
requires setting up the entire system for another batch of components, then make span will be a
more appropriate measure.
If a job is completed ahead of its due date, then there will be negative lateness (earliness) as
per Equation 18.4. For the set of jobs being scheduled, one can compute additional measures
such as mean lateness and maximum lateness.
Lateness is the difference between completion time and the due date.
As in the case of other measures, one can compute the mean, maximum, and minimum values
of tardiness for the set of jobs scheduled using a particular scheduling rule. The third measure is
the number of tardy jobs, denoted by nT.
Inventory-/Utilization-based measures
Inventory- or utilization-based measures are based on the actual inventory build-up in the shop or
the utilization of the shop resources. The measures include the mean number of jobs waiting, the
mean number of finished jobs, the mean number of unfinished jobs, and the mean utilization of
machines (or other resources) in the shop. It is also possible to convert these numbers into
monetary terms by appropriate cost-based measures. Although these measures are very
meaningful, computing these numbers is neither straightforward nor easy. These measures are
typically used while the scheduling problem is studied using a simulation model.
It is possible to obtain optimum schedules for certain combinations of shop configurations, scheduling rules, and
performance criteria.
EXAMPLE 18.2
A manufacturing organization is in the process of assessing the usefulness of SPT and EDD
rules for the purpose of scheduling four jobs on a machine. Table 18.5 presents information
on the processing time and due dates for the jobs. Assume that all the jobs are available for
scheduling at the beginning of the planning horizon so that the release time is zero. Further,
due dates in the table denote the time at which the component is due at its next stage of
manufacturing.
Compute the relevant performance measures and comment.
TABLE 18.5 Processing Time and Due Dates for the Jobs
Solution
Job sequencing using SPT rule: 3–1–2–4.
Job sequencing using EDD rule: 1–2–4–3.
In order to understand when the jobs get completed, we need to estimate the beginning and
end times for all jobs. In the case of SPT, Job 3 is first scheduled in the machine. At the end
of 2 time units, Job 1 is scheduled and it takes 4 time units. Therefore, at the end of 6 time
units, Job 2 is scheduled, and finally, at the end of 13 time units, Job 4 is scheduled. Job 4
completes its processing by 21 time units, which is the make span. Using this information we
can compute all the measures. The workings for both SPT and EDD are given in Tables 18.6
and 18.7. It may be noted that all the jobs are indeed available for processing at time zero.
Therefore, the release time for all the jobs is zero. Due to the scheduling rule, there are
differences with respect to when they undergo the processing.
Some Observations:
Both the scheduling rules result in the same make span. This is because there is only one
machine involved and all the jobs undergo processing in this machine. However, the rules
differ in their performance. EDD rule performs better in due date–based measures even at the
expense of increasing the number of tardy jobs by one.
TABLE 18.6 Performance of the SPT Rule
TABLE 18.7 Performance of the EDD Rule
Past research in the area of scheduling has established that it is possible to draw optimum
schedules for certain combinations of shop configurations, scheduling rules, and performance
criteria. Therefore, choice of these parameters could significantly influence the complexity of the
problem and the feasibility of obtaining optimum schedules.
It is not an easy task to arrive at the optimum schedule for a manufacturing shop. Numerous
combinations are possible for scheduling jobs in a shop. In a flow shop, the number of
combinations grows non-linearly (on account of the n! factor). Even for a small problem
involving 10 jobs, the number of alternatives to be enumerated is 3,628,800. Therefore, it will
not only be difficult but also futile to resort to complete enumeration of these alternatives.
However, we have noted earlier that a combination of the shop configuration, the scheduling
rule, and the performance measure could lend itself to identifying the optimum combination
without much computational effort. We shall concentrate on one such algorithm pertaining to the
flow shop here.
Development of context-specific scheduling rules is a better investment of managerial time to solve the
scheduling problem in organizations.
Johnson’s Rule
Let us consider a flow shop with two machines and n number of jobs to be scheduled. Let us also
assume that we are interested in minimizing the make span. Johnson proposed a simple
algorithm that will provide the optimum schedule for the shop.5 The algorithm is as follows:
Step 1:Let t1i denote the processing time of Job i in Machine 1 and t2i denote the processing time in Machine 2.
Step 2: Identify the job with the least processing time in the list. If there are ties, break the tie arbitrarily.
a. If the least processing time is for Machine 1, place the job at the front of the sequence immediately after any jobs
already scheduled.
b. If the least processing time is for Machine 2, place the job at the back of the sequence immediately before any jobs
already scheduled.
c. Remove job i from the list.
Step 3: If there are no more jobs to be scheduled, go to Step 4. Otherwise, go back to Step 1.
Step 4: The resulting sequence of jobs is the best schedule to minimize the make span of the jobs.
Johnson’s rule is quite simple and it applies to a restrictive set of assumptions about the
number of machines and the shop configuration. However, there have been some extensions of
Johnson’s rule that apply to a wider set of situations involving flow shops. The first variation is
Jackson’s rule that generalized Johnson’s rule to a situation involving two machines with jobs
skipping one of the two machines.6 Thus, the flow shop becomes a job shop with four sets of
jobs possible. One set containing jobs visiting only Machine 1, the second set visiting only
Machine 2, the third set visiting Machine 1 and then Machine 2, and the fourth set visiting
Machine 2 first and then Machine 1. The second extension of the Johnson’s rule is the CDS
method.7 This method extends Johnson’s rule to a multi-machine situation and derives schedules
that are near-optimal.
EXAMPLE 18.3
Five jobs are to be scheduled in two machines in a manufacturing shop. All the five jobs
undergo processing in both the machines (flow shop). Table 18.8 provides information on the
processing time in both the machines.
TABLE 18.8 Processing Time
Processing Time
Job no. Machine 1 Machine 2
1 4 7
2 6 3
3 2 3
4 7 7
5 8 6
Solution
Johnson’s Rule:
The least processing time is for Job 3. Since it is for Machine 1, we schedule the job at the
front end of the sequence:
Job 3 – – – –
Job 3 is removed from the list. The next least processing time is for Job 2. Since the
processing time is for Machine 2, we schedule the job at the end of the sequence:
Job 3 – – – Job 2
The next job with least processing time in Machine 1 is Job 1. Therefore, we schedule it
immediately after Job 3 at the front:
The fourth job with least processing time in Machine 2 is Job 5. Therefore, we schedule the
job immediately before Job 2 at the back of the sequence:
Finally, we schedule Job 4 in the remaining slot. Since no more jobs are available, we end the
procedure with the following optimum schedule for the problem:
There are several realistic assumptions that one needs to take into consideration while
scheduling the jobs that makes the problem mathematically intractable. For example, there are
situations of no job waiting during certain intermediate stages of processing as in the case of a
heat treatment operation with a pre-heating schedule. There are other situations that warrant
minimum waiting before being scheduled for the next operation, as in the case of a painting
application. Further complications can arise on account of sequence-dependent set-up times
between two jobs.
As more and more such constraints are introduced into the problem, obtaining optimum
schedules evades the planner. The realistic alternative is one of obtaining good enough and
workable schedules. Production planners benefit in this process by gainful exploitation of the
available computing. Development of context-specific scheduling rules and heuristic procedures
will serve the management better and save time in solving the scheduling problem in
organizations.
As we have already noticed, scheduling is much more complicated in the case of a job shop.
Unlike the flow shop, developing optimal methods is difficult and computationally complex.
Complete enumeration is also not possible as the number of schedules [(n!)m] grows
exponentially with n and m. Therefore, the use of appropriate scheduling rules and performance
criteria are the only viable options for the planner. Use of graphical methods such as Gantt charts
is very useful for job-shop scheduling. However, unlike the case of flow shops, the performance
of various scheduling rules cannot be easily predicted in a job shop. Therefore, one can resort to
simulation modelling of the system and evaluate alternative scheduling rules.
The use of graphical methods such as Gantt charts is very useful for job-shop scheduling.
EXAMPLE 18.4
Consider a job shop with five machines and four jobs. Each job has a processing sequence in
which it visits three out of the five machines. Each job also has a due date for completion.
The relevant information is given in Table 18.9.
a. Use SPT and EED rules to obtain the schedules for the jobs.
b. Compute the make span, mean lateness and number of tardy jobs.
Solution
We use a Gantt chart to develop the schedule for the shop.
SPT Rule:
Jobs 2 and 4 require processing in Machine 1. However, since the processing time of Job 4 is
less than for Job 1 it is scheduled first in the machine using SPT. Similarly, Job 1 is first
scheduled in Machine 2. The partial Gantt chart is shown in Figure 18.5.
TABLE 18.9 Details for each Job
Note: *The numbers in the parenthesis indicates the machine in which the job gets processed
At the end of time unit 3, Job 1 exits Machine 2 and it is time to take the next scheduling
decision. Job 1 will be scheduled in Machine 4 as there is no other job competing for that
machine. Similarly Job 3 will be scheduled in Machine 2. The Gantt chart in Figure 18.6 has
these additional schedules included.
FIGURE 18.5 The partial Gantt chart for the SPT rule
FIGURE 18.7 The final Gantt chart for the SPT rule
FIGURE 18.8 The Gantt chart for the EDD schedule
One can proceed in this fashion to arrive at the final chart for the SPT rule, as shown in
Figure 18.7. It may be noted that the SPT rule is invoked only when more than one job
competes for a resource. The make span for the problem is 21 units as evident from the Gantt
chart.
The EDD rule schedule is drawn using a similar logic. When there are competing jobs for
a resource, the EDD criterion is used. The Gantt chart for the EDD schedule is given in
Figure 18.8.
In this case, the make span has increased by two time units to 23. Table 18.10 presents a
summary of the performance measures for SPT and EDD rules. Interestingly, the EDD rule
has fared badly on completion time, make span and lateness. This example clearly
demonstrates the complexity of job shops. Predicting the performance of the rules is not as
easy as in the case of a flow shop. In both the cases, all the jobs were tardy.
TABLE 18.10 Performance Measures for SPT and EDD Rules: A Summary
So far, we have concentrated on flow shops and job shops and issues pertaining to scheduling
and operational control in these systems. However, there are several examples of mass
production systems in operation today. For example, a two-wheeler manufacturer such as Bajaj
Industries produces over 9,000 two wheelers everyday. Various sub-assemblies in the Bajaj plant
need to be configured to match the production rate. Similarly, the final assembly stations should
also have the required number of resources at each station to meet the targeted demand. In such a
scenario, much of control and scheduling boils down to appropriately arriving at a balanced flow
of components on the shop floor. There are two ways of achieving this objective:
a. Estimate the number of resources required at each station of the manufacturing system for a targeted production level.
Conversely, if there are a certain number of resources available, estimate the production level and adjust the plans
accordingly.
b. Given a certain availability of resources at various stages of the production system and a targeted daily production,
estimate the rate at which the entire system should work and accordingly modify the work practices and resource
deployment to meet the target.
The first method lays emphasis on designing operational control systems for mass production.
On the other hand, the second method focuses on the ability to craft alternative operational
control strategies to meet the changing targets on a period-to-period basis. The former is a
traditional application of the concept and is often referred to as the line-balancing technique,
while the latter is the modern Japanese approach to manage mass production systems. A detailed
description of line balancing is available in Chapter 4 of the book.
The line-balancing technique does not solve all the operational control problems in a mass
production system. It provides a basis for the initial design and deployment of resources.
However, mass production systems face constant revisions in the production rate. These changes
occur due to changes in the market demand and revisions in component off-take by customers.
Consider an auto-component manufacturing firm, such as the Chennai-based Lucas TVS that
supplies alternators to Maruti Udyog Limited (MUL). Suppose the daily requirement of MUL is
480 alternators during the first week of a month. If Lucas works for two shifts of eight hours
each, then the hourly production of alternators is 480/16 = 30. Furthermore, the cycle time for
alternator assembly is 60/30 = 2 minutes. This means that the production system responsible for
manufacturing alternators for MUL should work at a uniform pace with a cycle time of two
minutes everywhere. The final assembly will have to be tuned to two minutes, and so will all the
feeder shops to the final assembly.
Suppose that there has been a revision in the daily requirement by the third week of the month.
Due to increased production at their plant, let us assume that MUL wants a daily supply of 600
alternators. To cope up with the increase in the demand, Lucas will have to work with a new
cycle time 20 per cent less than the earlier measure (that is, 1.6 minutes). The final assembly as
well as all the feeder units will have to be tuned to this new requirement. This measure therefore
provides a rhythm for the overall functioning of the shop. In mass production systems, this
measure is known as TAKT time.8 Operational control issues in mass production systems begin
with the computation of TAKT time. Based on this, several choices are made to ensure that the
actual cycle time is as per the computation of TAKT. These choices can be grouped under the
following categories: machine redeployment, altering operator allocations, and adjusting material
feed rates.
Operational control issues in mass production systems begin with the computation of TAKT time.
Machine Redeployment
Mass production systems lay emphasis on the flexibility of the varying deployment of resources
to alternative product groups. Therefore, one method of responding to changing TAKT time is to
alter the assignment of machines (or manufacturing cells) to specific product groups. By
incremental addition/deletion of machines, it is possible to adjust the production rate and cycle
time.
In this chapter, we have discussed various techniques for scheduling operations in a job shop and
a flow shop. However, despite the use of these, there is a common problem of high WIP
inventory in several manufacturing organizations. Moreover, manufacturing organizations suffer
from the “end-of-the-period” syndrome. Typically, as organizations approach the end of a
reporting period (the year-end is a classic example, but there are monthly and quarterly reporting
cycles also), workflow gets considerably complex and inventory piles up beyond normal limits.
Several reasons are attributed to this and significant among them are the following:
a. Organizations do not use appropriate measures for performance.
b. Operations scheduling does not take into consideration some ground realities in the shop floor.
Measures based on standard costing, such as variances at the department or even work-centre
level, promote local optimum at the expense of global optima. Therefore, it is possible to create
mountains of inventory, yet products may not flow evenly out of the manufacturing system.
Similarly, scheduling on the basis of balancing capacity at individual work-centre level may not
ensure a smooth flow of material through the manufacturing system.
Measures of Performance
The theory of constraints developed by Eliyahu Goldratt addresses these concerns and proposes
alternative methodologies for operations scheduling.9 Goldratt also proposed three new measures
of performance for manufacturing firms.
According to the theory of constraints, it is possible to increase the throughput even while the operating
expenses and inventory are being reduced.
Operational measures: The operational measures for a company are the throughput, inventory, and operational
expenses. These measures could be used to make all decisions pertaining to the use of manufacturing resources.
Throughput: Throughput is the quantity of money generated by a firm through sales over a period of time. The standard
definition of throughput is total output, but this definition considers only sales and not finished goods (FG). According
to Goldratt, this is because finished googs do not generate revenue for the firm until they are sold.
Inventory: The quantity of money invested in materials that a firm intends to sell is defined as inventory. Contrary to the
traditional logic (where inventory valuation includes value-added components to estimate WIP and FG inventory), here,
it is a simple measure of the value of the material only.
Operating expenses: Operating expenses include all expenses incurred by a firm to convert inventory into throughput
over a period of time.
According to the theory of constraints, it is possible to increase the throughput even while the
operating expenses and inventory are being reduced. This is identified as the single most
important goal of managing manufacturing organizations. The principles governing the theory of
constraints are best explained using the analogy of marching soldiers.
As in the case of the soldiers, resources in manufacturing organizations differ from one
another in their ability to process components. Some machines may have more than the required
capacity to process in a planning horizon. Some other resources may have less than the total
capacity required for the planning horizon. Statistical fluctuations and dependent events are
characteristic of resources in a manufacturing organization. All these generate similar effects in a
manufacturing system as that of increasing the gap between the soldiers. However, we know that
increasing WIP inventory and manufacturing lead time is directly related to the increase in the
gap between the soldiers as they march. Therefore, operational control is primarily a function of
making sure that the gap between the soldiers does not increase with time.
The theory of constraints is a systematic body of knowledge that recognizes these effects in a
manufacturing system and uses specific methods to improve the performance of the system
under these conditions. Synchronous manufacturing is a specific application of the theory of
constraints to scheduling and operational control of the manufacturing systems.
Synchronous manufacturing is a specific application of the theory of constraints to scheduling and operational
control of manufacturing systems.
Synchronous Manufacturing
The marching-soldiers example shows that the rate at which they can cover ground is a function
of the slowest marching soldier in the column. If soldiers ahead of the slowest continue to march
at their higher speeds, they will only contribute to increasing the gap between them and the
slowest. On the other hand, those behind the slowest will be forced to march only at the speed of
the lowest. One way to solve this problem is to first tie a rope connecting all the soldiers (from
the first to the last), and provide a drumbeat at a rate equivalent to the slowest and enable the
marching soldiers to respond to the rhythm of the drumbeat. By doing this we not only ensure
that the gap does not increase with time, but also create a synchronized movement of the soldiers
on the ground. Synchronous manufacturing deploys these features in the scheduling and
operational control logic. In synchronous manufacturing, the focus is on synchronizing flow
rather than on balancing capacities.
This discussion clearly brings out the need to address the scheduling problem in a different
context. It is important to plan the scheduling and control of the entire manufacturing system on
the basis of availability of resources. The following definitions are common in synchronous
manufacturing literature.
Bottleneck resource: Any resource whose capacity is equal to or less than the demand placed on it is a bottleneck
resource. On the contrary, any resource whose capacity is more than the demand placed on it is a non-bottleneck
resource.
Any resource whose capacity is equal to or less than the demand placed on it is a bottleneck resource.
Capacity constrained resource (CCR): In a manufacturing system it is observed that some resources have the ability to
significantly deviate the product flow, from the planned flow, if not properly scheduled. Any resource that can deviate
the actual product flow from the planned flow is known as a capacity constrained resource. A bottleneck or a non-
bottleneck resource may become a CCR.
Any resource that can deviate the actual product flow from the planned flow is known as a capacity
constrained resource (CCR).
Activation of a resource: Activation merely refers to the deployment of a manufacturing resource to process materials.
Merely processing material may not in itself guarantee increase in throughput.
Utilization of a resource: Utilization refers to the activation of a manufacturing resource that will lead to improving the
throughput.
Process batch: A process batch refers to the quantity of a product (or part) processed at a resource before that resource
changes over to another product (or part). For example, if two parts—X and Y—are processed in Resource R1, then, if
300 units of Part X are processed before changing over to processing Part Y, then the process batch for Part X is 300.
Transfer batch: A transfer batch indicates the quantity moved at the same time from one resource to another resource in
a manufacturing system. For example, if 100 units of Part X are moved at a time from Resource R1 to Resource R2 for
further processing, then the transfer batch of Part X is 100.
From the definitions and the guiding principles we have discussed, a few important aspects can
be inferred:
The separation of resources as bottleneck and non-bottleneck is very important because while bottleneck resources
determine the (planned) output of the system, CCR will ensure that the actual throughput does not deviate from the plan
in a manufacturing system. Focusing on maximizing the utilization of a bottleneck resource is instrumental to
maximizing the throughput in a manufacturing system. On the other hand, scheduling is done in synchronous
manufacturing with reference to CCRs.
Placing buffers in front of bottleneck resources will protect the throughput of the system
Careful choice of the transfer and process batch sizes will ensure the protection of the throughput at the bottleneck as
well as proper scheduling of the CCRs
Once these aspects are determined, the rest of the system must work in tandem with these decisions A drum–buffer–rope
(DBR) methodology helps one to achieve these while scheduling a manufacturing system.
Several organizations have benefited from the use of the DBR methodology for scheduling
operations in a manufacturing system. In response to the demand for easy application of these
principles, several computer-based software packages have also been developed to implement
the DBR methodology. Synchronous manufacturing principles represent a recent approach to
addressing scheduling complexities in organizations. Moreover, the notion of the theory of
constraints is also applicable to other operations systems such as management of large projects.
SUMMARY
The focus shifts from operations planning to operational control in the short term. Scheduling aids operational
control in manufacturing and service systems.
The scheduling context relates to the number of jobs and machines in the system and the physical configuration of
the machines. These factors greatly influence the complexity of scheduling.
Flow shops and job shops are two alternatives for the configuration of a manufacturing system. The scheduling
methodology and complexity differ vastly between these two. Job shops are far more complex to schedule than
flow shops.
Johnson’s algorithm provides an optimal schedule for a two machine–n job problem using the shortest processing
time rule for scheduling. Several variations of Johnson’s algorithms are available for flow-shop scheduling.
Operational control in mass production systems are primarily achieved through use of TAKT time–based
scheduling.
The theory of constraints indicates that scheduling of operations must take into account the existence of bottlenecks
and statistical fluctuations in operations.
Synchronous manufacturing principles apply the theory of constraints and develop alternative schedules using a
drum–buffer–rope methodology.
FORMULA REVIEW
1. Critical Ratio
REVIEW QUESTIONS
1. How is scheduling linked to other aspects of production planning? Explain with the help of an example.
2. What do you mean by the loading of machines? What do you achieve by performing the loading function?
3. Classify the following as a job shop or as a flow shop for the purpose of scheduling. Also, identify the relevant parameters
of the system that have a bearing on the scheduling complexity.
PROBLEMS
1. Consider a one-machine–six-jobs scheduling problem. All the jobs are available for scheduling at the beginning of the
planning horizon itself. The processing times and due dates of the jobs are given in Table 18.13.
2. A service provider has two stages in service delivery. Initially, the customers arrive at Stage 1, get the initial part of the
service, and then reach Stage 2, where the remaining part of the service is completed. All arriving customers wait in a
lounge and are called on some basis into the system once a batch of eight accumulates. All of them go through these two
stages, although the time they spend in these two stages may vary depending on their specific requirements. The time
spent by a set of eight customers on these two stages is shown in Table 18.14.
3. A pure flow shop has three machines and all the jobs visit these machines. The processing time of each job in each
machine is given in Table 18.15.
a. Use the SPT rule to schedule the jobs in the flow shop and draw a Gantt chart for the schedule.
b. Compare the performance of the SPT rule with the EDD rule using the following performance criteria:
i. Mean tardiness
ii. Mean flow time
iii. Number of tardy jobs
iv. Make span
c. Compute the utilization of the three machines for both the schedules.
(Hint: Convert it into a two machine problem. First alternative is to combine processing times of m/c1 and m/c2
into a single equivalent machine. Alternative 2 is to combine the processing times m/c2 and m/c3. Solve using
Johnsons rule and choose the best alternative)
5. A job shop has five jobs for processing. There are four machines in the shop. The jobs visit the machines in a particular
order (indicated by each job’s routing file). Table 18.17 has details of the machines visited by each job and the processing
time of each machine. The table also has information on the due dates.
a. Develop a schedule for the jobs using (i) SPT, (ii) EDD, and (iii) FCFS rules.
b. Use three appropriate performance criteria and evaluate these rules.
c. Compute the machine waiting time and job waiting time for all the schedules obtained using the SPT, EDD and
FCFS rules. If you were to select a scheduling rule on the basis of the minimum job waiting time, what will your
recommendation be?
d. If there is a penalty for delay at the rate of ₹250 per minute and a reward of ₹100 per minute for completing early,
compute the cost of the schedules attributable to these.
a. The TAKT time for all the shops in the manufacturing system.
b. If the successive workstations in the manufacturing plant are uniformly placed at a distance of three metres, at what
speed should the material handling system move between the work-stations?
MINI PROJECTS
1. A vast number of scheduling packages are available in the market today. Scan through the following sources:
Journals: OR/MS Today (a bimonthly magazine from the Operations Research Society) and IIE Solutions (a journal
of the Institute of Industrial Engineering). These journals periodically bring out a survey of available software for
specific areas of production management. Look for surveys pertaining to scheduling software.
American Production & Inventory Control Society (APICS).
Search engines: Search with keywords such as “APS”, “advanced production scheduling”, and “scheduling
software”.
a. Based on this exercise, prepare a technical note on a survey of the available software options for
scheduling.
b. Develop a classification scheme to list all the software that you have come across.
2. Visit a health management system (such as a multi-specialty hospital with facilities for in-patients, out-patients,
laboratories for diagnostics, operation theatres, and special care systems) and study the scheduling practices with respect
to the following:
3. Read the book The Goal by Goldratt and Cox.11 After reading the book, write a book review. While writing the review,
answer the following questions amongst other things that you may like to cover in the book review:
a. What are the key elements of the alternative method by which Alex Rego tries to solve the problems at his
manufacturing set-up? Clearly outline the key conceptual details of his alternative strategy that helped him find a
solution.
b. How do you link his strategies for controlling a manufacturing system to that of the notion of scheduling that we
have discussed in the chapter?
c. What are the key performance measures that Alex Rego uses for controlling his manufacturing system?
d. Do you think that his concept has a wide applicability? If so, give some examples of situations where this concept
can be applied.
SUGGESTED READINGS
J. E. Ashton and F. X. Cook Jr, “Time to Reform Job Shop Manufacturing,” Harvard Business Review 67, no. 2
(1989):106–111.
K. R. Baker, Introduction to Sequencing and Scheduling (New York: John Wiley and Sons, 1974).
K. R. Baker, “The Effects of Input Control in a Simple Scheduling Model,” Journal of Operations Management 4,
no. 2 (1989): 99–112.
M. Berrada and K. E. Stecke, “A Branch and Bound Approach for Machine Load Balancing in Flexible
Manufacturing Systems,” Management Science 32, no. 10 (1986): 1316–1335.
G. G. Brown, R. F. Dell, R. L. Davis, and R. H. Daff, “Optimizing Plant-Line Schedules and an Application at
Hidden Valley Manufacturing Company,” Interfaces 32, no. 3 (2002): 1–14.
H. C. Campbell, R. A. Dudek, and M. L. Smith, “A Heuristic Algorithm for the n Job m Machine Sequencing
Problem,” Management Science 16, no. 10 (1970): B630–B637.
R. W. Conway, W. L. Maxwell and L. W. Miller, Theory of Scheduling (New York: Dover Publications, 2003).
E. M. Goldratt, Critical Chain (Great Barrington, MA: The North River Press Publishing Corporation, 1997).
E. M. Goldratt and J. Cox, The Goal [Chennai: Productivity Press (India) Pvt. Ltd, 1993).
J. R. Jackson, “An Extension of Johnson’s Results on Job Lot Scheduling,” Naval Research Logistics Quarterly 3
(1956): 201–203.
R. S. Kaplan, “Yesterday’s Accounting Undermines Production,” Harvard Business Review 62, no. 4 (1984): 95–
101.
S. Kontogiorgis and S. Acharya, “US Airways Automates its Weekend Fleet Assignment,” Interfaces 29, no. 3
(1999): 52–62.
Y. M. Lee and E. J. Chen, “BASF Uses a Framework for Developing Web-Based Production Planning
Optimization Tools,” Interfaces 32, no. 6 (2002): 15–24.
L. S. Murty “An Indian Case of Throughput Improvement,” Productivity 42, no. 2 (2001): 337–343.
S. Nahmias, Production and Operations Analysis, 4th edition (New York: McGraw-Hill Publishing Co., (2000).
O. Oliver and C. Rajendran, “Efficient Job Shop Dispatching Rules: Further Developments,” Production Planning
and Control 11 (2000): 171–178.
S. S. Panwalker and W. A. Iskander, “A Survey of Scheduling Rules,” Operations Research 25, no. 1 (1977): 45–
61.
V. Portugal and D. J. Robb, “Production Scheduling: Just Where Is it Applicable?” Interfaces 30, no. 6 (2000): 64–
76.
V. Suresh and D. Chaudhuri, “Dynamic Scheduling: A Survey of Research,” International Journal of Production
Economics 32 (1993): 52–63.
M. M. Umble and M. L. Srikanth, Synchronous Manufacturing (Cincinnati: South Western Publishing Co., 1990).
T. E. Vollmann, W. L. Berry, and D. C. Whybark, Manufacturing Planning and Control Systems, 4th edition (New
York: McGraw-Hill Publishing Co., 1997).
CHAPTER 19
CRITICAL QUESTIONS
After going through this chapter, you will be able to answer the following questions:
In an organization like a five-star hotel, it is not enough to main tain a 99 per cent quality level. Six sigma helps such
organizations to improve their quality to near-zero defect levels.
Source: stock.xchng
ideas at Work 19.1
several days, the sample observations were outside the control limits. In the case of acidity
levels, nearly two-thirds of the data was outside control limits.
These pointed to an immediate need for better control at IFS. If the situation was out of
control due to the process, further investigation might be required. On the other hand, if it was
due to workers not adhering to specified procedures, better training and discipline might be
required.
Every organization uses several of the ideas that IFS has tried to implement to improve
quality control in its production processes. There are several tools and techniques available to
monitor a process and detect situations when the process deviates from specifications or goes
out of control. Furthermore, these tools also enable organizations to predict the quality level of
the processes and target improvement towards zero defects in the long run.
Source: Based on S. Y. Deodhar and D. Tirupati, “Indian Food Specialties Limited,” Vikalpa 27, no. 1 (2002): 55–68.
In Chapter 12, we discussed several issues that an organization needs to address to ensure that
quality becomes an ingrained value among all its employees. TQM principles set the stage for an
organization to build specific subsystems for quality management. Recall that one of the three
pillars of TQM is the establishment of tools and techniques for quality assurance. In this chapter,
we focus our attention on the application of statistical tools. Statistical tools and techniques play
a central role in setting up a robust quality assurance system in any organization. Therefore, an
operations manager should have a sound conceptual understanding of these techniques.
Multinational firms such as Motorola and GE pioneered the use of the six-sigma approach to quality.
Six sigma is a new approach to process control based on a set of principles that enables
organizations to improve their quality to near-zero defect levels. These include understanding
customer needs well, appropriate and disciplined use of data and statistical tools, statistical
analysis, and a closer attention to managing and improving business processes using a set of
tools. The basic unit of analysis with respect to the quality of any process is the defect. A defect
is an unacceptable state of a product or service for a customer, arising out of a business process
that produced it. Since the focus of six-sigma quality is primarily business processes, it is equally
applicable to both manufacturing and service organizations. Since quality improvements have
been happening in manufacturing shop floors ina big way due to the TQM movement, one can
expect a greater overall impact of the six-sigma approach in the non-manufacturing portions of a
manufacturing organization, as well as in service organizations. This in turn will greatly improve
the competitive position of an organization.
Six sigma is a new approach to process control based on a set of principles that enables organizations to improve
their quality to near-zero defect levels.
A defect is an unacceptable state of a product or service for a customer, arising out of a business process that
produced it.
Six-sigma methodology seeks to provide a set of tools and techniques and an organizational
framework to eliminate defects from any business process by ensuring that it is an
“extraordinarily rare event”. We define a new metric to capture the essence of this notion of
“extraordinarily rare event”. Six-sigma quality control differentiates itself from the traditional
quality control methodology on the basis of three features:
a. A new metric, defects per million opportunities (DPMO), to predict/assess the quality of a business process.
b. A new methodology, “define–measure–analyse–improve–control” (DMAIC), to ensure that very high levels of quality
could be assured in the chosen business processes thereby generating favourable outcomes to both the business and the
customers.
c. An organizational framework for ensuring that these outcomes are generated on a sustained basis.
Six-sigma quality control differentiates itself from the traditional quality control methodology on the basis of
three features:
Assume that we run a hotel business in a tourist destination such as a hill resort in Darjeeling or
the backwaters in Kottayam, Kerala. Several business processes are involved in running the
hotel. Let us consider one business process—the check-in and check-out of guests. The process
may have a few steps and a certain number of potential opportunities to make errors leading to
some defect. Assume that the check-in process has potentially 9 defect opportunities. During one
peak season, the resort dealt with 1,142 guests and the number of defects was 361. From this
data, one can compute the DPMO in the following fashion:
Number of defect opportunities per unit (k) = 9
Number of guests handled (n) = 1,142
Number of defects (d) = 361
Number of defect opportunities = 9 × 1,142 = 10,278
Defects per million opportunities (DPMO) = (361/10,278) × 1,000,000 = 35,123
One of the important requirements of a six-sigma approach quality control is to be able to predict
the DPMO of a business process by observing the process characteristics and linking them to
customer specifications. We shall see later in the chapter how this is possible by obtaining the
voice of the process using control charts and mapping it on to the voice of the customer.
Six-sigma quality levels are achieved using a structured five-step methodology in organizations.
It is usually referred to as DMAIC (define–measure–analyse–improve–control). This resembles
closely the PDCA cycle advocated by Deming. DMAIC provides a mechanism for making
improvement in quality on a project-by-project basis. By the use of several tools available in the
domain of quality management and continuous improvement, the DMAIC methodology
combines group work, organizational focus towards customers, and process improvement into
one. Thus it reaches far and wide compared to traditional quality control using mere statistical
tools and is likely to generate a greater impact. For each project, a team utilizes the DMAIC
methodology to improve quality. The five steps, as shown in Figure 19.1, are detailed here:
The DMAIC methodology combines group work, organizational focus towards customers, and process
improvement.
1. Define: The first step in a six-sigma project is to define the problem and the context for improvement. The activities
involved in this step are:
Defining the problem, the requirements, the scope of the project, and the project charter.
Setting goals for improvement.
2. Measure: Once the problem is defined, the next phase is to obtain data. Six-sigma methodology is a data-driven
employee-centred methodology. Furthermore, the analysis of process variations is critical to improve quality. Therefore,
in the measure phase, the activities include:
Identifying variables to be measured and the type of measurement.
Data collection and synthesis.
This forms the basis for further analysis and quality improvement.
3. Analyse: During this phase, the data collected is analysed to find possible causes for bad quality and identify areas that
need improvement in order to reduce the defects. Several statistical, analytical, and graphical tools could be used in this
stage. The main activities performed include:
Applying graphical tools of analysis
Identifying possible sources of variation and “vital” few root causes.
Exploring means of eliminating them.
Various QC tools described in Chapter 12 are typically put to use in this phase. Often, the analysis is
triggered by the use of control charts and identification of situations where the process seems to be going
out of control.
4. Improve: In the improvement phase ideas for eliminating the “vital” few root causes are further developed and solution
methodologies devised for implementing the same. The activities include:
Generating and validating improvement alternatives.
Creating new process maps for the process.
5. Control: In the control phase, steps are taken to ensure that the process does not drift away from the newly improved
standard. In order to achieve this, the following steps are taken:
Developing control plan
Establishing revised standard measures to maintain performance
Developing relevant training plans to maintain standards
Let us take a look at the people required and their roles in order to achieve this.
The process owner: The process owner is the one who takes responsibility for the various steps of a process that is
responsible for delivering some output to the customer. The process owner could be the supervisor or a manager in a
particular work area where the improvement project has been identified.
The team leader and the members: Six sigma is a small group improvement activity. Therefore, the team leader (the
project leader) and the members will comprise the employees in the chosen work area. They will have day-to-day
operational control of activities and knowledge of the process that they seek to improve.
The six-sigma coach: A coach is a consultant who offers expert knowledge on various aspects of six sigma. This
includes statistical tools, process design and analysis, change management, small group improvement, use of QC tools
for improvement, etc. The coach could be a senior person within the organization or it could be a resource person
external to the organization.
The sponsor: A sponsor is a member of the senior management who oversees the overall progress and implementation
of the improvement project by a six-sigma team. The sponsor helps the team refine the project scope, sorts out issues
cutting across other parts of the organization, approves projects, and provides the necessary support in terms of
resources.
Clearly, this example amply illustrates the fact that choosing an appropriate characteristic for
measurement, monitoring, and control is not an easy task. It is important for a quality
manager to realize that resolving this issue is fundamental to setting up a process control
system.
Source: B. Mahadevan, “Personal Finance Limited” (Case #02-091, Centre for Development of Cases and Teaching Aids, IIM Banga-lore, 2002).
In six sigma, three terminologies are used to indicate these organizational entities. This includes
master black belt, black belt, and green belt. The depth of training and experience differentiates
these three. As soon as one completes a basic six-sigma training programme, he/she could be
designated as a green belt. A black-belt candidate is supposed to have completed a certain
number of improvement projects before earning this designation. In general, a master black belt
is expected to be an expert in statistical tools with a good capability for change management.
Team leaders are typically green belt, although occasionally even a black belt can be a team
member. The coach is either a black belt or a master black belt. The sponsors are sometimes
master black belts.
The depth of training and experience differentiates master black belts, black belts, and green belts in a six-sigma
organization.
19.4 VARIATIONS IN PROCESSES
Any business process is bound to have variations from time to time. The vibration of a machine
used for metal cutting in a machine shop will vary over time. The sharpness of the cutting tool
used in the machine also decreases over time, resulting in variations in the surface finish. A
process industry such as Reliance will experience variations in the temperature of the molten
PVC resin being fed to a spinning machine in the PVC plant. The workers in the final watch
assembly at the Titan plant will also go through variations in their work routines, thereby
impacting the quality of the assembly. Service organizations go through similar problems of
variations in their business processes. An insurance firm may see variations in preparing an
insurance quote to new and existing customers leading to some quality problems. Hospitals
experience considerable variation in processes pertaining to the registration and admission of
patients. In the hospitality industry, customers experience considerable variations in the manner
of the services delivered to them. Setting up a good six-sigma quality control mechanism in an
organization requires a sound footing on the notion of variations in business processes and the
use of well-known statistical tools to assess the nature and quantum of these variations.
The collective set of tools and techniques used to develop a quality assurance system when business processes
exhibit variations is known as statistical process control (SPC).
Variations happen because the operating system tends to deviate from the specifications.
When variations are inevitable, how are we to ensure that the operating system produces quality
output? Statistical techniques play a valuable role in addressing this issue. In this chapter, we
will know more about these variations, the statistical principles that help an operations manager
to develop methods to reduce variations and ensure quality even when variations exist. We will
also see how an organization can set up a process for assuring quality. The collective set of tools
and techniques used to develop a quality assurance system when business processes exhibit
variations is known as statistical process control (SPC).
Two types of variations are normally encountered in operating systems. These variations occur
on account of two types of causes. The first type is known as chance variations. These variations
arise on account of purely random events that are not in the control of the system designer or the
operating personnel. Therefore, the causes of these variations are also known as common causes
of variations. The second set of variations happen on account of assignable causes. Certain
perceptible changes happen in the system, resulting in variations, and the system drifts away
from the desired state. It is possible to identify these changes, take corrective measures, and
restore the system back to its desired state of operation.
Common Causes
Since the common causes of variations stem from random behaviour of the system, one can
utilize the laws of probability to understand, analyse, and predict the state of the system.
Consider a machine used to fill the half-kilogram sachets of Surf Excel, a detergent
manufactured by Hindustan Unilever Limited. Due to a variety of random occurrences in the
system, the machine may not fill exactly 500 g in each sachet. In some cases it could be 499 g, in
some other cases it could be 502 g, and so on. Random events are both unavoidable and
uncontrollable. Therefore, the focus of maintaining quality in this example will shift to the
following:
a. How do we know whether the observed changes are due to random variations or assignable causes?
b. How does one ensure that the random events are indeed rare events?
Chance variations arise on account of purely random events that are not in the control of the system designer or
the operating personnel. The causes of these variations are known as common causes of variations.
We will introduce some principles from probability to help us address the these questions
associated with common causes of variations. More detailed application of these concepts will be
taken up in later sections of the chapter.
Let us consider the detergent-filling machine. If we make n observations of the working of the
machine and note the filled weight of each sachet, we can compute the mean and the standard
deviation based on these observations. This set of observations is nothing but a sample from the
detergent-filling machine. One can compute the standard deviation of these observations as
follows:
Furthermore, we can make use of several such samples of n observations and construct a
distribution of the means of the samples. The distribution of the sample means follows a normal
distribution.1 Therefore, using the samples, we can compute the statisticalattributes (such as the
mean and the standard deviation) of the “population” from which we have drawn this sample. In
our example, the “population” is the half-kilogram sachets continuously coming out of the
detergent-filling machine.
A normal distribution is characterized by the location of the average, the spread of the
distribution, and the shape. The spread of normal distribution is a direct function of the standard
deviation. If the standard deviation is large, the spread is also large and vice versa. Figure 19.2
illustrates these aspects of normal distribution.
It is well known that the area covered under three standard deviations on either side of the
mean of normal distribution is 99.7 per cent. If the observed samples are within ±3σ, then one
can conclude that there is a 99.7 per cent probability that the variations are due to random events.
Conversely, if the observed values are outside ±3σ, one begins to suspect that the variations are
due to random events.
FIGURE 19.2 Normal distribution: spread and shape
Assignable Causes
When the observed variations are not statistically found to be due to random effects, it clearly
points to the existence of some assignable causes. In our detergent-filling machine example, if
we found that the variations are beyond random effects, then several other issues may contribute
to the variations. If the machine requires some operator intervention, then differences in the
practices between two operators could be one cause. Alternatively, the viscosity of the detergent
powder could have changed due to some changes in the previous stage of production. A third
possibility is malfunctioning of the filling machine itself due to some structural disturbances in
the operating conditions of the machine or the operating environment.
When the observed variations are not statistically found to be due to random effects, it clearly points to the
existence of some assignable causes.
From a statistical point of view, assignable causes of variations impact the distribution of the
process in more ways than one. One possibility is that the average would have shifted out of the
desired level. Instead of operating with an average of 500 g, the detergent-filling machine might
have shifted to some other mean (say 497 g or 502 g). The second possibility is that the variance
in the process increases significantly, leading to a wider spread. When the spread is wider, large
variations in the process will be considered to be within the chance category, when it is not the
case in reality. It is also possible that the process is skewed to one side, thereby having a greater
tendency to fill more than 500 g, or less than 500 g, depending on the direction of the skewing.
In all the cases we just discussed, it is possible to trace the causes of the variations and take
corrective actions. Once corrective action is taken, the system will return to normal operating
conditions and variations will be only due to chance. A process having only chance variations is
said to be in a state of control. Therefore, it is clear that when a process is under a state of
control, the associated distribution of the process will not shift the average. Moreover, the spread
of the process will also remain the same and the distribution will be symmetric.
Before we proceed further in our understanding of the elements of process control, it is useful to
know the terms associated with the techniques.
Nominal value: The desired average for a process is known as the nominal value. In our detergent-filling machine
example, the nominal value for the weight of the filled sachets is 500 g.
Specification limits: Specification limits are nothing but the tolerances allowed in every process. These tolerances are
arrived at after making a detailed design based on several considerations. The upper specification limit (USL) provides
the upper limit for the measure under consideration, while the lower specification limit (LSL) provides the lower limit.
In our detergent-filling machine example, the design for the process would have specified that the process could have a
fill weight ranging between 496g and 504g. In this case, USL = 504 g and LSL = 496 g. The specification limits are
nothing but the voice of the customer, pertaining to the process under consideration.
Process average: How the process is operating is independent of what one desires the process to be. By actually
sampling the process, one can establish the average at which the process is operating. This average is known as the
centre line (of the process).
Control limits: As we already saw, the process is considered to be under control when there are only chance variations.
The upper control limit (UCL) is defined by the value of three positive standard deviations from the process average.
Similarly, the lower control limit (LCL) is defined by three negative standard deviations from the process average.
Specification limits are nothing but the tolerances allowed in every process. The upper specification limit
(USL) provides the upper limit for the measure under consideration while the lower specification limit (LSL)
provides the lower limit.
The process average indicates how the process is operating. This average is known as the centre line (of the
process).
The upper control limit (UCL) is defined by the value of three positive standard deviations from the process
average. Similarly, the lower control limit (LCL) is defined by three negative standard deviations from the
process average.
Clearly, as actual observations fall within UCL and LCL, we conclude that there are only chance
variations. The process average and the control limits represent the voice of the process. In our
detergent example, if the process is centred at 500.5 g and the standard deviation is 0.25, then
UCL = 501.25 g and the LCL = 499.75 g.
Setting up a process control system for any process involves a set of decisions that the quality
manager needs to take. Figure 19.3 outlines a six-step methodology for this purpose.
Table 19.1 gives a sample set of characteristics that are applicable for various manufacturing
and service settings. There is no rule that each process should have only one characteristic to be
chosen for process control. As a rule, every process will have at least one equipment related
measure and another measure for the component/product being produced in the process.
Similarly, in a service organization, there should be at least one measure that is customer-related
and another pertaining to the resources involved (primarily the human resource) in generating
customer service in that process.
Attribute measurements are easy to make, quick, and less expensive. However, they reveal very little information
on the actual performance of the process with respect to the characteristic.
A control chart is a graphical representation of the process status in terms of the UCL, LCL, process average,
and the plot of the sample data. It helps to identify when the process goes out of control.
Step 4: Choose the Type of Control Chart
A control chart is a graphical representation of the process status in terms of the UCL, LCL,
process average, and the plot of the sample data. Figure 19.4 is a generalized representation of a
control chart. Irrespective of whether an attribute measure or a variable measure is used, the
control chart structure remains the same. The only difference lies in the manner in which the
process average, UCL, and LCL are computed. The plot of the sample data, once the control
limits are established, is an integral part of setting up the control chart. It helps to identify when
the process goes out of control. A plot outside the control limit is an indication of the possibility
of assignable causes for the variations.
TABLE 19.2 Notations and Computations Used in Control Charts
used for variable measurements while the p chart and the c chart are used for attribute
measurements.
and R charts
Variable measurement based process control systems use and R charts. Using the equations
available in Table 19.3, one can compute the mean for and R charts. However, in order to
establish the UCL and the LCL, the standard deviation of the sampling distribution is required.
One method is to use readily available tables that provide coefficient values for computing the
UCL and LCL for the and R charts. The computational aspects of setting up and R charts are
as follows.
The chart and the R chart are used for variable measurements, while the p chart and the c chart are used for
attribute measurements.
Computations for the chart:
The values A2, D3 and D4 can be read directly from Table 19.3.
TABLE 19.3 Coefficients for Computing the LCL and UCL in and R Charts
EXAMPLE 19.1
compute the average and range of the observations in each subgroup and then the average of
the sample means and ranges. Table 19.5 gives the details:
From this table we find that the process is centred at 12.417, with a value of 0.119.
Using these and the table for coefficient values one can compute the UCL and LCL as
follows:
n = 5, therefore, from Table 19.4 we obtain A2 = 0.577, D3 = 0 and D4 = 2.114
given in Figure 19.5. Since no sample value falls outside the control limits, the process is in a
state of statistical control. All variations observed are due to common causes only.
FIGURE 19.5 The graphical representation of the control charts
p charts for attribute measures
There are numerous occasions on which firms use a two-way classification of products and
services while assessing quality. The products/services are either classified as conforming to
specifications (good) or not conforming to specifications (bad). One way to represent this
information is to compute the proportion of defects (p) in the process and establish a control
chart for this measure. Such a measurement system follows a binomial distribution. Therefore,
one can compute the mean and standard deviation of the population using a sampling distribution
of the proportion of defects using known results for the binomial distribution. The relevant
computations for establishing a p chart are as follows:
There are numerous occasions in which firms use a two-way classification of products and services when
assessing quality. The products/services are either classified as conforming to specifications (good) or not
conforming to specifications (bad).
EXAMPLE 19.2
A manufacturer of an electronic control system (ECS) for musical appliances uses a non-
destructive testing mechanism to assess the quality of ECS. A lot of 100 ECSes is drawn
periodically for testing. After testing, the ECSes will be classified as either defective or good
depending on the outcome of the test. Table 19.6 gives the number of defects for 12 such
samples. Establish a p chart for the process.
Solution
We first compute the proportion of defects in each sample. The computations are in Table
19.7:
The mean and standard deviation of the sampling distribution of the proportion of defects
is computed as follows:
Mean of the sampling distribution =
Therefore, the UCL and LCL values for the p chart can be computed. Further, one can set up
the chart and plot the sample proportions to investigate whether the process is in control. The
computations and the graphical representation of the chart are shown in Figure 19.6.
UCLp = 0.105 + 3 × 0.031 = 0.198
LCLp = 0.105 − 3 × 0.031 = 0.012
EXAMPLE 19.3
A manufacturer of colour televisions has a final inspection before the product is packed for
shipping. The final inspection involves testing the television on several parameters and
verifying if the set performs satisfactorily in all the tests. It is possible that every television
may fail to fulfil all the parameters. Therefore, a c chart is used at this stage to control the
process. Periodically, 10 television sets are drawn for testing. Table 19.8 shows the
performance during one such test. Use this data to set up a control chart.
Solution
As in the previous cases, we first compute the mean and the standard deviation of the
sampling distribution of the number of defects per unit (c).
The graphical representation of the chart is given in Figure 19.7. We note that Sample 6 falls
outside the UCL. Therefore, the process does not appear to be in control. Management must
investigate the reasons for this and decide if any corrective actions are required. If so, the
process needs to be adjusted and a new control chart needs to be established. If no corrective
actions are immediately warranted, then the chart needs to be plotted again by deleting
Sample 6 and recomputing the control limits.
It can be verified that after removing Sample 6, the process average and the control limits
have shifted down. The revised control chart, after deleting Sample 6, is shown in Figure
19.8.
FIGURE 19.7 Graphical representation of the chart
Let us see when a process should be stopped for investigation. In order to specify some simple
rules, we need to split the control chart into three zones. Zone C is the area between one standard
deviation on either side of the centre line. Zone B is the next standard deviation on either side of
Zone C, and Zone A is the next standard deviation on either side of Zone B. The following
guidelines are useful for deciding when to stop the process of investigation:
a. One point beyond Zone A (as in the case of Example 19.3)
b. Nine points in a row in Zone C, or more
c. Six points in a row, steadily increasing or decreasing
d. Fourteen points in a row, alternating up and down [see Figure 19.9(a)]
e. Two out of three points in a row in Zone A or beyond
f. Four out of five points in a row in Zone B and beyond
g. Fifteen points in a row in Zone C [see Figure 19.9(b)]
We know from the elementary probability theory that every one of the occurrences listed is a
rare event and, therefore, the causes are likely to be of the assignable category.
FIGURE 19.9 When to Stop the Process: Some Rules
19.7 PROCESS IMPROVEMENTS IN THE LONG RUN
So far, we have discussed one aspect of the control chart, that is, to ensure that the process is in
statistical control. If a process is in statistical control, it merely indicates that there are no
assignable causes of variations to the process and that the process is stable. Having a statistically
stable process is the foundation on which superior quality improvements could be launched in
any organization. If a process is in control, it provides additional predictive capabilities to the
organization. Before we understand the notion of predictive capability, let us try to differentiate
between accuracy and precision.
A statistically stable process is the foundation on which superior quality improvements can be launched in any
organization. It provides additional predictive capabilities to the organization.
Let us assume that there is a component manufactured using a process. The design
specification is that the diameter of the component should be in the range of 15 ± 0.05 cm. If we
set up a variable-based process control chart for this process, we may get to know the process
average, UCL and LCL. Let us now consider two scenarios of the process. In the first scenario,
the process is centred at 15.3 cm, with a spread of ± 0.04 cm. In the second scenario, the process
is exactly centred at 15, but with a spread of ± 1.2 cm. What do we infer from these two
scenarios? Scenario A is precise but not accurate, whereas Scenario B is accurate but not precise.
Accuracy is a function of how far the process average is from the designed average. On the other
hand, precision is a function of how well the process lends itself to repetition. A process with
high repeatability will have narrow standard deviation and therefore narrow spread.
Which one of the two is better? Clearly, a process with high precision is better than a process
with low precision. The issue of accuracy can be sorted out by shifting the process average
(through some process adjustments) to the desired average. Moreover, the lower spread
characteristic of a precise process provides certain advantages to an organization and enables it
to predictably produce goods of a certain quality. Using these ideas, it is possible to use a control
chart as a useful tool for managers to make long-term and sustained improvements in the process
performance. We shall dwell on these aspects in some detail.
From our discussion, we infer that superimposing the process performance (as evidenced by
the process average and the spread) on the design specifications can throw up some useful
measures for a manager to understand the ability of the process to generate good-quality
components in the long run. Before addressing this issue, let us look at some related definitions.
Process Capability
Process capability is the spread of the process. If the standard deviation of the process is s, then
the process capability is 6s. This merely indicates that the process can produce goods within ±3s.
To understand the notion of process capability, it is important to relate it to the design
specifications, usually indicated by the process centre, USL, and LSL. Let us consider two
processes A and B (see Figure 19.10). Process A has a larger spread and Process B has a smaller
spread. If the 6s of a process lies within the USL and LSL, then the process is capable, as 99.73
per cent of the goods produced will be within the specification, limits. In our example, Process A
is not capable of producing the components as per specifications, whereas Process B is. It is clear
from this example that, in general, processes with narrow spreads are likely to be more capable
than those with larger spreads. This has been one of the motivating reasons for reducing process
variations in practice.
Potential Capability
As we see in the example we just discussed, the process spread alone does not completely
capture the capability of the process. It also needs to be related to the specification limits.
Therefore, an alternative measure is the potential capability of the process. Potential capability
(Cp) is the ratio of the difference in the specification limits to the process spread.
Potential capability (Cp) is the ratio of the difference in the specification limits to the process spread.
As potential capability takes on larger values, the process capability increases. It is customary to
use a measure of parts per million (ppm) to measure defects these days. It can be shown that a
process with a Cp value of 1.33 will produce 64 ppm, whereas a process with a Cp value of 0.67
will produce 45,500 ppm (4.55% defects) and a process with Cp = 1 will produce 3,000 ppm.
The quality improves exponentially as Cp values increase.4 However, improving the Cp value
requires exponential efforts too.
Actual Capability
In our analysis of process capability, we have made a very important assumption about the
process. We have implicitly assumed that the process centre exactly coincides with the nominal
value (centre as per design specifications). In reality, this need not be the case. Let us examine
the situation portrayed in Figure 19.11. We consider two identical processes with the same
spread. However, Process A is centred along the nominal value and Process B is offset from the
nominal value to the left. Process A is capable of producing components without defects as the
spread is well within the specification limits. Although Process B is also capable by virtue of
having an identical spread, it will produce defective components as the LCL is outside the LSL.
It is clear from these two cases that Cp merely indicates that the process is capable of producing
components as per specifications but may not reveal if it indeed does so. That is why Cp is
referred to as potential capability.
One can compute the actual capability of a process by incorporating the additional information
pertaining to the extent to which the process has deviated from the nominal value. Such as
measure is known as actual capability (Cpk). An expression for Cpk is given by:
In Equation 19.5, we compute the offset and identify the side to which the process has drifted
and use it as the basis for computing the process capability. It is easy to infer that if the Cpk value
is equal to Cp, then the process is exactly centred at the nominal value. As in the case of Cp, a
higher value of Cpk denotes fewer defects coming out of the process.
EXAMPLE 19.4
A manufacturer of ball-bearings has the following specifications for the diameter of finished
ball bearings. The nominal value = 245 mm. USL = 245.85 mm, LSL = 244.15 mm. The
process planning department proposed two processes for manufacturing the ball bearings.
After extensive pilot runs, the processes were statistically stabilized. The performance of the
two processes is given in Table 19.9.
The production department wants to know which of these two processes is suitable for
commercialization. Compute the relevant measures and offer appropriate recommendations
to the production department.
Solution
Let us first compute process capability measures for the two processes.
Process capability = 6σ
For Process A, process capability = 6 × 0.25 = 1.50
For Process B, process capability = 6 × 0.27 = 1.62
Process B exhibits greater spread although it is centred close to the nominal value of 245.
Therefore, while Process B is more accurate than Process A, it is less precise.
Process A is more capable of producing components as per specifications (as it has the
potential to produce fewer defects than Process B).
Actual Capability of the Processes
Using Equation 19.5, we can compute Cpk values for the two processes as follows:
When we compute the actual capability, we find that Process B is marginally better than
Process A. This is because Process B is more accurate than Process A. While it may be
tempting to prescribe Process B to the production department, we must remember that a
precise process is always better than an accurate process with less precision. By taking some
additional steps to move the centre of Process A close to the nominal value, one can achieve
better results in the production. Therefore, the following recommendations will be in order.
a. Both the processes exhibit low Cp and Cpk values. There should be efforts to reduce the variations in the process
and improve the process capability. This is a long-term requirement for installing a better process for production.
b. If the costs of the adjusting the centre of the process are not significant in the case of Process A, then it should be
preferred over Process B. Otherwise, immediate action is to be taken to reduce the variations in Process B, if it
needs to be used for the production of ball bearings.
The idea of six sigma comes directly from the understanding of the Cp measure, which we have
already discussed. A process with six-sigma quality level is one that has a Cp value of 2.
Rewriting Equation 19.4, for a Cp value of 2, we get,
Therefore, in order to attain six-sigma quality, we need to ensure that the design specification
width is ± 6σ (see Figure 19.12). Thus, even if the process centre does not coincide with the
nominal values (as per design specifications), a further spread of 4.5σ available on both sides
will ensure that minimal defect opportunities are created in the process. In the worst-case
scenario, the process will have 3.4 ppm non-conformance on either side of the distribution (6.8
ppm total) and in the best-case scenario, it will have 1 non-conforming part per billion (ppb) on
either side of the distribution (2 ppb total). Motorola has made significant progress towards this
goal across most processes, including many office and business processes.
FIGURE 19.12 Graphical representation of six-sigma quality
This requires a two-pronged approach. First, product designers should succeed in creating
designs with the widest possible tolerances. At the same time, process designers should be able
to design processes with the narrowest possible variation.
The greatest advantage of using these measures is the predictive power of these measures with
respect to the number of defects that the process is likely to produce if it is in statistical control.
Multinational companies belonging to several manufacturing and service sectors use these
measures to assess the processes of the vendors for vendor selection and for pegging the level of
incoming inspection required when the selected vendors supply products and services.
Modern-day initiatives in continuous improvements in the long run revolve around the use of
measures such as Cp, Cpk, and six sigma. More and more manufacturing and service companies
are using these principles and setting targets to improve the quality of product and service
offerings.
Earlier approaches to quality control did not rely much on process control. Instead, quality was
measured after the processing was completed and components were ready for use. Let us
consider a simple example of a manufacturer of grey iron castings supplying an automobile
component manufacturer. In this case, the castings manufacturer is the supplier to the automobile
component manufacturer. The question of whether or not the castings supplied were as per
specifications (and hence of right quality) could be answered in two ways. One method is to put
a set of appropriately designed control charts in the casting manufacturing plant at various stages
and monitor the process. We have discussed various issues pertaining to this in earlier sections.
The other alternative is to request the supplier to deliver the finished castings at the automobile
component manufacturer’s stores and use statistical methods to assess the quality of the lot based
on the information obtained from a sample drawn out of the lot. For instance, if a lot consisted of
100 castings, then the automobile manufacturer may choose to draw a sample of 5 castings and
subject it to some tests. Based on the outcome of the test, the fate of the entire lot will be
decided. This approach to quality assurance is known as acceptance sampling as the decisions
with regard to acceptance or otherwise of a lot is based on a sample. The key assumption behind
the approach is that the sample is random and every item in the overall lot has an equally likely
probability to get sampled.
Setting up a system for acceptance sampling requires that the following questions are
satisfactorily answered:
a. What is the size of the sample required for assessing quality?
b. What should be the criteria for acceptance of the lot?
c. Is it enough to have just one sample or is it desirable to go for multiple samples?
d. How do we assess the impact of the choice of the above parameters on the quality level in the long run?
The first three questions are answered by the choice of an appropriate sampling plan. A
sampling plan describes the lot size (N), size of the sample to be assessed (n), the acceptance
number (c), that is, the number of defects permissible in the observed sample, and whether it is a
single sampling plan or a multiple sampling plan. In the case of multiple sampling plans, for
each, the sample size (n1, n2, and so on) and the acceptance number (c1, c2, and so on) needs to
be specified. Furthermore, the plans could be sequential with a stopping criterion. The actual
number of defects detected in the sample is denoted by d.
A sampling plan describes the lot size (N), size of the sample to be assessed (n), the acceptance number (c), that
is, the number of defects permissible in the observed sample, and whether it is a single sampling plan or a
multiple sampling plan.
Acceptable quality level (AQL) is the percentage of defects that the buyer is willing to tolerate in the lot
delivered by the supplier.
Lot tolerance per cent defective (LTPD) is the worst quality beyond which the buyer is not
willing to accept the incoming lot. In Example 19.5, the manufacturer may say, that even in the
worst case, he may not want to have a lot with more than 10 per cent defects.
Lot tolerance per cent defective (LTPD) is the worst quality beyond which the manufacturer is not willing to
accept the incoming lot.
Since the decision making on the entire lot is based on imperfect information (obtained from a
random sample), it becomes inevitable to work with a bandwidth of quality levels for acceptance.
Only if the buyer can manage a 100 per cent inspection is it possible for him to specify a single
parameter. For instance, he may say that he would require lots that are 1 per cent or less
defective, and all lots more defects will be rejected and the rest accepted.
EXAMPLE 19.5
We are able to infer that if the incoming quality level of the electronic component in the
long run is 1 per cent defective, then choosing the sampling plan will ensure that the
organization accepts the lot on 98.6 per cent of the occasions. On the other hand, if the
incoming quality level is 12 per cent defective, the organization will still accept the lot on 5.1
per cent of the occasions. These have important implications for the management in the long
run.
However, specifying a pair of measures introduces an element of risk for both the supplier and
the buyer. In order to understand this, let us return to Example 19.5. Let us suppose that AQL = 1
per cent and LTPD = 10 per cent for the problem. This means that as long as the lot contains up
to 1 per cent defects, the buyer is willing to accept the lot without any hesitation. However, we
noticed that instead of accepting the lot 100 per cent, the sampling plan would have rejected it on
1.3 per cent occasions. Thus, the supplier faces the risk of having his/her lot rejected when it was
indeed conforming to the agreed terms. This is known as producer’s risk (or Type I error in
statistical terms, indicated by a ).
When the supplier faces the risk of getting his/ her lot rejected when it conforms to the agreed terms, it is known
as producer’s risk.
On the other hand, we also notice that when the incoming lot had more than 10 per cent
defects, the sampling plan would have accepted the lot on 11.2 per cent occasions. This is the
risk that the buyer faces when the lot is beyond his/her tolerance level. This is known as
consumer’s risk (or Type II error in statistical terms, indicated by b ). Therefore, an appropriate
choice of sampling plan is important to reduce both the producer’s and consumer’s risks. Figure
19.14 graphically depicts these ideas using the problem in the example.
The risk that the manufacturer faces by accepting a lot when it is beyond his/ her tolerance level is known as
consumer’s risk.
FIGURE 19.14 Producers’ and consumers’ risks in using an acceptance sampling plan
SUMMARY
Six sigma is a new approach to process control based on a set of principles that enables organizations to improve their
quality to near-zero defects level.
Six-sigma quality control differentiates itself from the traditional quality control methodology on the basis of three
features: A new metric, defects per million opportunities (DPMO), to predict/assess the quality of a business process, a
new methodology, DMAIC (“define–measure–analyse–improve– control”), to ensure that very high levels of quality
could be assured in the chosen business processes, and an organizational framework for ensuring that these outcomes
are generated on a sustained basis.
All processes exhibit variations on account of two causes. Common causes of variations happen on account of random
events. Assignable causes introduce variations that can be detected and eliminated.
Setting up a process control system involves choosing a characteristic to control, identifying a measurement method,
and developing a relevant control chart.
A plot of the data helps a quality manager to assess if the process is in control and also helps her to detect any
impending shift in the process parameters.
Superimposing the specification limits on the control chart helps an organization to assess the process capability and the
likely number of defects that the process will produce. It also helps to target for a six sigma quality improvement
programme.
Acceptance sampling is used to accept a lot of items based on the observed number of defects in the sample drawn out
of the lot.
The performance of an acceptance sampling plan could be judged by developing an operating characteristics (OC)
curve for the plan.
An acceptance sampling design results in a certain amount of risk for both the customer and supplier.
FORMULA REVIEW
6.
REVIEW QUESTIONS
1. What do you understand by the term six-sigma quality control?
2. Often several managers comment that six sigma is nothing but an alternative version of TQM, Kaizen, JIT, and other such
continuous improvement initiatives undertaken in an organization. How do you react to this comment?
3. Distinguish between assignable and common causes of variations. Why is this distinction important in quality control?
4. Give three examples for assignable causes of variation.
5. In the broad scheme of TQM, where do you find a role for statistical process control?
6. What are specification limits? Why do we need them in SPC?
7. Are specification limits and control limits the same? Explain.
8. A manufacturer of garments wants to set up a quality control system using control charts for process control. The
manufacturer has three options to choose from:
a. Measure the critical dimensions of the garment for establishing its quality
b. Segregate every batch of production into good quality and seconds quality
c. Estimate the number of defects per bale of cloth issued for production
The manufacturer is not sure about what it means to choose each of the above. Prepare a report explaining the pros and cons
of each of the choices, the nature of efforts required to set up control charts, and the implications of their use.
9. Suppose an organization utilizes a variable-based measurement system for process control. During a period, it was found that
while all the plotted observations fell within the control limits in chart, one point was lying outside the control limits in the
PROBLEMS
1. An insurance service provider monitors the process of handling customer queries for new insurance quotes using a control
chart. The process has five steps and after preliminary analysis a set of parameters was identified as critical. If these
parameters were not properly addressed, then the process is considered defective. Every day, samples are drawn from the
process and defects measured. Each sample is of size 100. Table 19.11 has 30 such samples drawn over a two-month period.
2. A manufacturer of hair oil has an automated process consisting of a filling machine for filling the 200-ml bottles and a sealing
unit. After filling the bottle, it is automatically sealed with a metallic cap. To monitor the quality of the process, samples of
size 5 are drawn periodically from the process and the quantity filled is measured. Table 19.12 has data on 12 such
observations.
TABLE 19.12 Sample Data (Problem 2)
4. Table 19.14 presents some details on the mean and range of 11 samples drawn from a process during routine sampling and
inspection. The readings are the weight of a certain metal deposited on the surface of a component used in an electro-
mechanical device. The lower and upper specification limits for the weight of metal to be deposited are 10.7 g and 13.3 g,
respectively. The standard deviation for the process is found to be 1.28. Assume the sample size to be four.
5. The critical design specification for the manufacture of a component is the length of the component. The current specifications
are 15 ± 0.05 cm. The organization currently employs a manufacturing process with the centre at 14.98 and a standard
deviation (σ) of 0.02.
a. What is the process capability of the process?
b. Are its potential capability and the actual capability the same? What is the ratio of the actual capability to potential
capability? What does this ratio indicate?
c. If the organization is interested in ensuring a quality level of 64 ppms and is unable to shift the process average, what
should the spread be to achieve this quality level?
d. If the process is centred to the nominal value (design specification), what will the actual capability be?
6. A manufacturer of computer accessories sources certain plastic components from a supplier. The manufacturer subjects the
components to an acceptance sampling procedure with a sampling plan in which a sample of 50 components is drawn from the
supply and the lot is accepted if the number of defects detected in the sample is not more than three. Answer the following
questions:
a. If the incoming lot is 1 per cent defective, what is the probability of acceptance?
b. If the incoming lot is 5 per cent defective, what is the probability of acceptance?
c. Make the necessary computations and plot an operating characteristic curve for the sampling plan.
d. If the manufacturer sets an AQL of 1 per cent, what is the implication of the sampling plan to the supplier?
e. Suggest an appropriate LTPD so that consumer’s risk is not more than 10 per cent.
f. If the LTPD is unacceptable to the manufacturer (because it is too high) what changes would you like to propose in the
sampling plan?
g. Suppose the manufacturer wants to tighten the plan by reducing the acceptance number from 3 to 2. What are the long
term implications of this decision?
h. Construct another OC curve for an acceptance number of 4. Based on these computations, what do you think is the
relationship between the alternative sampling plans and the OC curve? What are its implications for practice?
NET-WISE EXERCISES
1. Wikipedia has a collection of articles, references and multiple perspectives on six sigma. Go to
http://en.wikipedia.org/wiki/Six_Sigma to learn more about this. Study the material available on this page carefully, and then
look up the references that it provides.
Based on this, prepare a report for a public-sector bank in India on the issue of implementing
the six-sigma approach in their organization. Indicate the steps that they need to take, the nature
of training that they may need, and the tools and technologies that they need to acquire. Select
one potential pilot project in banking operations as a context for covering the six-sigma
implementation part of the report.
MINI-PROJECT
Visit the local branch of a nationalized or a private-sector bank. Study the working of the system
from a customer perspective and identify three critical measures of quality. Use the six-step
approach to developing process control described in the chapter and establish a method for
process control. Elicit the manager’s views on desired specifications for the three chosen
parameters.
a. Collect sample data and analyse the process. Is the process in a state of statistical control?
b. Share the data with the bank manager. What are the manager’s key observations on your work?
c. How capable is the process now?
d. Identify methods by which the process capability could be increased by 25 per cent.
Prepare a report based on the work done along the lines suggested.
SUGGESTED READINGS
S. E. Daniels, Kristen Johnson and Corinne Johnson, “Quality Glossary,” Quality Progress 35, no. 7 (2002): 43–61.
E. L. Grant, Statistical Quality Control (New York: McGraw-Hill, 1988).
R. W. Hoyer and W. C. Ellis, “A Graphical Exploration of SPC: Part 1 SPC’s Definitions and Procedures,” Quality
Progress 29, no. 5 (1996): 65–73.
J. M. Juran and F. M. Gryna, Quality Planning and Analysis, 3rd edition (New Delhi: Tata McGraw-Hill, 1995).
T. A. Little, “10 Requirements of Effective Process Control: A Case Study,” Quality Progress 34, no. 2 (2001): 46–52.
J. M. Lucas, “The Essential Six Sigma,” Quality Progress 35, no. 1 (2002): 27–31.
D. C. Montgomery, “The Use of Statistical Process Control and Design of Experiments in Product and Process
Improvement,” IIE Transactions 24, no. 5 (1992): 4–17.
G. P. Moore and D. A. Hendrick, “Statistical Process Control in Project Management,” AACE Transactions (1991):
A3(1)–A3(7).
J. Munoz and C. Nielsen, “SPC: What Data Should I Collect? What Charts Should I Use?” Quality Progress 24, no. 1
(1992): 50–52.
P. S. Pande, R. P. Neuman, and R. R. Cavanagh, The Six Sigma Way (New York: McGraw-Hill, 2000).
E. L. Porteus and A. Angelus, “Opportunities for Improved Statistical Process Control,” Management Science 43, no. 9
(1997): 1214–1228.
G. Tennant, Six Sigma: SPC & TQM in Manufacturing and Services (Burlington, VT: Gower, 2001).
C. Wozniak, “Proactive versus Reactive SPC,” Quality Progress 27, no. 2 (1994): 49–50.
Operations Management: Trends and Issues
1. For details, visit http://www.servicetax.gov.in
2. M. Sharfman, R.T. Ellington, and M. Meo, “The Next Step in Becoming ‘green’: Life-Cycle
Oriented Environmental Management,” Business Horizons, May–June 1997, pp.13–22.
3. V. D. R. guide and L. N. Wassenhove, “Managing Product Returns for Remanufacturing,”
Production and Operations Management 10, no. 2 (2001): 142.
Operations Strategy
1. “Precariously Poised,” Business India, 24 March–6 April 1997, pp. 69–72.
2. For a detailed analysis of the performance of automobile manufacturers during 1980–1990 and
its implications for operations management, see J. P. Womack, D. T. Jones, and D. Roos, The
Machine That Changed the World (New York: Rawson Associates, 1990).
3. For more details see P. Kumar, R. Batra, and A. Narain, “State Bank of India: Kohinoor
Banjara Branch”, Indian School of Business Case.
4. For more details, see R. J. Schonberger, World Class Manufacturing: The Lessons of
Simplicity Applied, (New York: The Free Press, 1986); and R. J. Schonberger, World Class
Manufacturing: The Next Decade (New York: The Free Press, 1996).
5. Poka yoke and total quality management are discussed in Chapter 12, and just-in-time
philosophy is discussed in Chapter 13.
6. For an idea on the issues involved see, B. Mahadevan, “Journey to the Deming: Implications
for Corporate Transformation and Competitiveness,” IIMB Management Review 17, no. 2
(2005): 59–70.
7. K. Mitra, “Bloodbath in White Goods,” Business Today, 17 July 2005, pp. 118–122.
8. T. V. Mahalingam, “The China Effect,” Business Today, 25 March 2007, pp. 86–95.
9. S. Srivastava, “Just Click Away”, Businessworld, 26 June 2006, pp. 48–49.
Sustainability in Operations
1. The periodic review inventory control model (known as p-system) is discussed in detail in
Chapter 17. See Section 17.6 for details.
2. For one such example see, B. Mahadevan, D.F. Pyke, and M. Fleischmann, “Periodic Review
Push Inventory Policies for Remanufacturing”, European Journal of Operational Research,
2003, 151, pp. 536–551.
Project Management
1. For the purpose of the discussions in this chapter, we analogously use the terms “a work
package”, an element of a project and an activity in a project. In reality, it is possible to
attribute differences to these terms by alternative definitions. However, such differences do
not matter at the level of treatment of the concepts and techniques in the chapter.
2. For a discussion on the concept of bill of materials, please refer to Chapter 16 of the book.
3. For a detailed analysis of the matrix organization structure and its benefits to a project
organization, readers are referred to standard textbooks in the area of organization, structure,
processes, and design.
4. For a good treatment of the fundamentals of project management tools, the reader is
encouraged to refer to J. D. Weist and F. K. Levy, A Management Guide to PERT/CPM, 2nd
edition (Englewood Chiffs, NJ: Prentice-Hall, 1988).
5. Software packages for project management, such as MS Projects, could depict both versions.
However, they use AON more often.
6. We use AOA for the purpose of constructing the network in this example and in all the others
that follow in this chapter. However, interested readers can develop the equivalent network in
AON. The analysis and interpretations, however, do not change.
7. The process of computing early schedules by proceeding from the “start” node to the “end”
node is known as forward pass. Similarly, the process of computing late schedules by
proceeding from the “end” node to the “start” node is known as reverse pass.
8. There are two types of slack; total slack and free slack. In this book, our definition of slack
pertains to total slack. Free slack, on the other hand, is the amount of time by which an
activity’s earliest finish times could be altered without affecting the early start times of its
successors. We refer readers to any standard textbook on project management for finer
differences and implications of these two definitions of slack.
9. A discerning reader will note that many of the alternatives proposed for altering the supply
capacity of an organization, discussed in Chapter 15 on aggregate planning, could be applied
for crashing a project.
10. The assumption here is that (tp − t0) spans 6σ in the distribution. By Chebyshev’s inequality,
this will cover at least 89 per cent variations in the time estimate.
11. This assumption is statistically reasonable in several real life examples due to the central
limit theorem, which states that the distribution of a random variable constructed using the
sum of random variables from an unknown distribution follows a normal distribution when the
sum involves a large number of random variables from the parent distribution. For practical
purposes, any number above 10 is considered satisfactory to apply central limit theorem.
12. A good description of the general methodology of studying risks applicable in a capital
budgeting situation is available in D. B. Hertz, “Risk Analysis in Capital Investment,”
Harvard Business Review 57, no. 5 (1979): 169–181. Readers are encouraged to go through
the article for a good understanding of how risks and uncertainties could be incorporated in a
managerial decision making framework.
Supply Chain Management
1. Product markdown is the term used to denote the downward revision of the selling price at the
end of a season to clear leftover stocks. We see frequent examples of this in the textile
industry.
2. This is due to what is known as the risk pooling effect. In simple terms, if we accumulate
multiple data points having uncertainty, the resultant variance will be lesser than the sum of
individual variances.
3. For details see, H. L. Lee and C. Billington, “Evolution of Supply Chain Management Models
and Practice at Hewlett-Packard Company,” Interfaces 25, no. 5 (1995): 542–63.
4. For an excellent introduction to this subject, see, H. L. Lee, V. Padmanabhan and S. Whang,
“The Bullwhip Effect in Supply Chains,” Sloan Management Review 38, no. 3 (1997): 93–
102.
5. The term order-to-order is frequently used in business process reengineering literature to
denote the elapsed time between a customer placing an order and the time when the order was
finally fulfilled by the organization.
6. M. L. Fisher, “What is the Right Supply Chain for Your Product?” Harvard Business Review
(March– April 1997): 105–116.
7. We discuss these issues in detail in Chapter 17.
Facilities Location
1. For more details, refer to A. Pal, “ABB Global Factories,” Business Today, 22 June 2003, pp.
122–126.
2. Michael Porter argued this issue in greater detail by proposing that industrial clusters can add
a competitive dimension to location. For more detail see, M.R. Porter, “Clusters and the New
Economics of Competition,” Harvard Business Review 76 no. 6 (1998): 77–90. For some
Indian experiences of benefiting from clusters, see R. Dubey, “Clustering for
Competitiveness,” Business Today, 22 April 1999, pp. 50–63.
3. This figure is adapted from the model proposed by the World Economic Forum to assess the
competitiveness of various countries during the year 1996. For more details about this model
refer to “Why Competitiveness Matters,” Business Today, 22 June 1996, pp. 71–73. The
model used for assessing the competitiveness rating has undergone some changes since then.
4. See H. A. Taha, Operations Research: An Introduction, 7th edition (New Delhi: Pearson
Education, 2003), or A. Ravindran, D. T. Philips, and J. J. Solberg, Operations Research:
Principles and Practice, 2nd edition (New York: Wiley, 1987) for more details on the
structure and solution methodologies for the transportation problem. Microsoft Excel has an
add-in utility called “Solver.” One can also use solver to obtain the solution for this problem.
Sourcing and Supply Management
1. For more details on this, see J. P. Womack, D. Jones and D. Roos, The Machine that Changed
the World (New York: Rawson Associates, 1990).
2. For details see T. Minahan, “Is this the Future of Purchasing?” Purchasing, 12 March 1998,
pp. 42.
3. Paul Snell, “Sourcing Hotspots,” Supply Management 12, no. 8 (2007): 24–26.
4. For details see Sue Welch, “The 5 Stages of Global Sourcing Step-by-Step,” World Trade 19,
no. 12 (2006): 60–63.
5. Throughout this chapter, we use terms such as “supplier”, “vendor”, and “supply chain
partner” in an interchangeable fashion. All of them essentially mean suppliers of goods and
services for an organization.
6. For a good introduction to the use of the Web for procurement, see J. Hazra, “Supply Chain
Management in the Internet Era,” IIMB Management Review 16, no. 4 (2004): 54–74.
7. In the chapter on inventory planning and control, we show that long lead times will result in
high investment in pipeline inventory. For more details, see Section 17.2 in Chapter 17.
8. Just-in-time (JIT) philosophy is discussed in detail in Chapter 13 of this book.
9. A good source to know how single sourcing will successfully handle such calamities is T.
Nishiguchi, “The Toyota Group and the Aisin Fire,” Sloan Management Review 40, no. 1
(1998): 49–59.
10. For more details see J. H. Dyer, “Dedicated Assets: Japan’s Manufacturing Edge,” Harvard
Business Review 72, no. 6 (1994): 174–178.
11. For a detailed set of criteria that could be used for vendor rating, see G.W. Dickenson, “An
Analysis of Vendor Selection Systems and Decisions,” Journal of Purchasing 2, no. 1 (1966):
5–17. Even after 30 years, several vendor rating systems are using only a subset of the factors
identified in this study.
12. For a good description of the concept of total cost of ownership, refer to L. Ellram,
“Purchasing: The Cornerstone of the Total Cost of Ownership Concept,” Journal of Business
Logistics 14, no. 1 (1993): 163–184. Also refer to Ideas at Work 11.5 for an actual
implementation of the total cost of ownership in an auto-component manufacturing unit.
13. For more details see
http://en.wikipedia.org/wiki/Business_process_outsourcing_in_India#cite_note-feout-0,
accessed on 5 March 2010.
14. For a detailed treatment of this subject, refer to B. Mahadevan, “Making Sense of the
Emerging Market Structures in Business-to-Business E-Commerce,” California Management
Review 45, no. 3 (2003): 86–100.
Process and Capacity Analysis
1. M. Hammer and J. Champy, Reengineering the Corporation (London: Nicholas Brealey
Publishing Ltd, 1993). For a good description of BPR in the Indian context, refer to M.S.
Jayaraman, G. Natarajan and A. V. Rangaramunajam, Business Process Reengineering, (New
Delhi: Tata McGraw-Hill, 1994).
2. For a five-step methodology to implement BPR, see Chapter 2 of M. S. Jayaraman, G.
Natarajan and A. V. Rangaramunajam, Business Process Reengineering, (New Delhi: Tata
McGraw-Hill, 1994).
3. Practitioners widely use an alternative term known as value stream mapping, which is the
same as process mapping. This is evident from the following definition of value stream
mapping: Value stream mapping allows the user to see his/her organization’s core processes
and understand how to best satisfy customer needs. This perspective allows managers to
evaluate which activities add value for the customer and to identify waste as well as areas of
opportunities for improvement.
4. For a detailed discussion of the notion of economies and diseconomies of scale, refer to any
standard textbook on managerial economics.
5. No attempt has been made in this book to define each source of waste and methods of
eliminating them in detail. Interested readers are referred to C. E. Dyer and A. T. Campbell,
Canon Production System: Creative Involvement of the Total Workforce (Cambridge, MA:
Productivity Press, 1987).
6. There is a well-known Japanese management concept known as total productive maintenance
(TPM), which addresses issues related to eliminating wast-ages arising out of breakdowns and
poor equipment maintenance.
7. Detailed discussions on alternative forecasting techniques and methods of computing the
forecast requirements are available in Chapter 14.
8. Instead of a simple yes or no, organizations can also introduce a five-point scale to reflect the
proficiency of the skill acquired by each employee. In such a situation, the method of
computing the multi-skilling index needs to be slightly modified. It may be desirable, in any
case, to start with this simple scheme before graduating to other forms.
Design of Manufacturing Processes
1. The first Indian company to obtain the PM Prize, an award from Japan for excellence in
maintenance management, is a process-industry organization (a cement-manufacturing unit).
Furthermore, a number of PM prize winners in the country belong to the process industry.
Given the importance of maintenance in process industries, this is hardly surprising.
2. Refer to Chapter 13 for more details on JIT.
3. The idea of the process–product matrix was first proposed by Hayes and Wheelright. For
details on this, refer to R. H. Hayes and S. C. Wheelright, “Link Manufacturing Process and
Product Life Cycles,” Harvard Business Review 57, no.1 (1979): 133–140.
4. For a good description of the various problems than an organization faces while adopting a
process layout, see Chapter 3 of B. Mahadevan, The New Manufacturing Architecture (New
Delhi: Tata McGraw-Hill Ltd, 1999).
5. The Russians pioneered the concept of group technology. As early as 1948, nearly 800
factories in the USSR were organized on the basis of group technology. The credit of
popularizing the concept of group technology among the non-Russian speaking nations goes
primarily to the late Professor J. L. Burbidge.
6. The layout problem described here belongs to a class of problems known as quadratic
assignment problems (QAP). For details on QAP, refer to S. Nahmias, Production and
Operations Analysis, 3rd edition (Englewood Cliffs, NJ: McGraw Hill International Editions,
1997), pp 580–581.
7. This is just an expectation and not a definite outcome as we are merely computing average
numbers for each workstation. The precedence relationships among tasks sometimes may not
allow combining tasks into the desired number of workstations for a targeted cycle time.
Therefore, unless we assign tasks and compute cycle time, we may not know the exact picture.
8. Several heuristic solution techniques have been developed to solve this problem. For example,
there is a rank positional weight method. For details on this method readers are referred to S.
Nahmias, Production and Operations Analysis, 3rd edition. (Englewood Cliffs, N. J.: McGraw
Hill International Editions, 1997), pp. 447–452.
9. For an exhaustive coverage on the methods available for the cell formation problem, readers
are encouraged to refer to B. Mahadevan and S. Venkataramanaiah, “Realigning Research
Objectives In Cellular Manufacturing System Design: A Users’ Perspective,” Technology,
Operations and Management, (2007), 17-29.
Design of Service Systems
1. In terms of the probability theory, a finite source corresponds to sampling without replacement
and an infinite source corresponds to sampling with replacement. As is known, there are
significant differences between these two cases with respect to the arrival rate and probability
distributions pertaining to arrival.
2. We do not intend to elaborate on the properties of probability functions and methods of
deriving cumulative distribution functions based on the probability-density function. We
assume that the readers are conversant with these basic details pertaining to probability
distributions.
Interested readers are referred to any standard textbook on the theory of probability.
3. For an alternative and detailed view on the demand versus capacity management issue in
operations, readers are referred to Chapter 15 of the book.
4. The idea was first published in the form of an article in the Journal of Marketing. Interested
readers may refer to A. Parasuraman, V. A. Zeithhaml and L. L. Berry, “A Conceptual Model
of Service Quality and Its Implications for Future Research,” Journal of Marketing, no. 49
(Fall 1985): 41–50.
PRODUCT DEVELOPMENT PROCESS
1. A good description of this is available in the book that was brought out at the culmination of this work. Interested readers can
refer J. P. Womack, D. T. Jones and D. Roos, The Machine That Changed the World (New York: Rawson Associates, 1990).
2. For more details, see S. C. Wheelright and K. Clarke, “Competing Through Development Capability in a Manufacturing-based
Organization,” Business Horizons 35, no. 4 (1992): 29–43.
3. For more details, please visit the Web site: http://tatanano.inservices.tatamotors.com/tatamotors/index.php?
option=com_content&task=view&id=215&Itemid=207
4. This is merely a calendar representation of the duration of the product development process. In reality, the requirement in each
stage is calculated either in man-months or man-hours: In this example, if eight people work on the concept generation stage,
then the requirement is 56 man-months.
5. Multi-dimensional scaling (MDS) is a standard statistical technique that helps an organization analyse the competitive
offerings in the market on any number of dimensions. Knowledge of this technique may help the design team in assessing the
strength of their proposed product offering in comparison with the existing ones in the market. For more details on MDS, refer
to any standard textbook on multivariate statistical analysis.
6. For a very good description of quality function deployment and its applications, readers may consult Y. Akao, Quality
Function Deployment: Integrating Customer Requirements into Product Design (New York: Free Press, 1990).
7. For more details, see “The Story Behind the Nano,” Business Week, 9 May 2008.
8. DFM principles were first developed by Professors Geoffrey Boothroyd and Peter Dewhurst during the early 1980s. For more
details on their work, a useful source is G. Boothroyd and P. Dewhurst, Product Design for Assembly (Wakefield, R.I.:
Boothroyd Dewhurst Inc, 1987).
9. For a good description of the design for the localization concept, its application, and the impact that it is likely to have on an
organization, refer to the paper, H. L. Hee, C. Billington, and B. Carter, “Hewlett-Packard Gains Control of Inventory and
Service Through Design for Localization,” Interfaces 23, no. 4 (1993): 1–11.
10. For a good discussion on the various aspects of mass customization, readers are referred to J. H. Gilmore and B. J. Pine II,
“The Four Faces of Mass Customization,” Harvard Business Review 75, no. 1 (1997): 91–101. Mass customization also
demands certain organizational structure changes. Refer to the paper by B. J. Pine II, B. Victor, and A.C. Boynton, “Making
Mass Customization Work,” Harvard Business Review 71, no. 5 (1993): 108–117 for more details.
11. A study reported that a Dell Inspiron 8100 G1200VT notebook computer can be configured online into 2,080,899,072 unique
alternatives using the options available for various system components such as the base system, the OS, optical drives (fixed
and removable), the carry case, battery, zip drive, video card, mouse, keyboard and hardware maintenance option. For details,
see B. Mahadevan, “Making Sense of Emerging Market Structures in B2B E-Commerce,” California Management Review 46,
no. 1 (2003): 86–100.
12. For a good introduction to product platforms, readers are encouraged to consult D. Robertson and K. Ulrich, “Planning for
Product Platforms,” Sloan Management Review 39, no. 4 (1998): 19–31; and M. H. Meyer and A. P. Lehnerd, The Power of
Product Platforms (New York: The Free Press, 1997).
13. Interested readers can refer to R. S. Pressman, Software Engineering: A Practitioner’s Approach (New York: McGraw-Hill,
1997) for more details on these models.
TOTAL QUALITY MANAGEMENT
1. For more details on this study, see D. Garvin, “Quality on Line,” Harvard Business Review 61, no. 5 (1983): 64–75.
2. Visit the website http://www.juse.or.jp/e/deming/index.html to know more about the Deming Prize and other categories of
prizes instituted by the Union of Japanese Scientists and Engineers (JUSE).
3. These 14 points were originally published in W. E. Deming, Out of the Crisis (Cambridge, MA: MIT Centre for Advanced
Educational Services, 1986).
4. A detailed description of process control charts follows in Chapter 19.
5. For details, see D.A. Garvin, “What Does Quality Really Mean?” Sloan Management Review 26, no. 1 (1984): 25–43.
6. In this chapter, we do not cover all QC tools in detail. This is partly because some of them are fairly simple and straightforward
to use. Moreover, there is plenty of published literature available on this topic. For a sample, readers are referred to K.
Ishikawa, Guide to Quality Control (Tokyo: Asian Productivity Organization, 1991). Another useful source is K. Suzaki, The
New Shop Floor Management (New York: Free Press, 1993). Control charts are discussed in detail Chapter 19.
7. The details pertaining to construction of a histogram have not been included in this write-up. This is because, it is not only a
widely known and a simple procedure, but also available in any standard elementary textbook on statistics. On the other hand,
we focus on actual examples where the tool has been used and explain its relevance to quality management. We take a similar
approach to several other tools described in the book.
8. N. Ravichandran, “A Framework for Creating A Competitive Advantage Through Throughput Time Reduction” (MPT Report,
IIM Bangalore, 1996).
9. For more details on CEDAC, readers are encouraged to refer to R. Fukuda, CEDAC–A Tool for Systematic Improvement,
(Cambridge, MA: Productivity Press, 1990). Another useful source is K. Suzaki, The New Shop Floor Management (New
York: Free Press, 1993).
10. This illustration is based on an unpublished report submitted in the Management Programme for Technologists at the Indian
Institute of Management Bangalore during 1996.
11. For a good description of QFD, readers may refer to J. R. Hause and D. Clausing, “The House of Quality,” Harvard Business
Review (May-June, 1988): 63–73.
12. For more details, see Y. A. Akao, Quality Function Deployment: Integrating Customer Requirements into Product Design
(New York: Productivity Press, 1990).
13. A detailed discussion on the house of quality is available in Chapter 11. Therefore, we merely provide a contextual
description of the tool here.
LEAN MANAGEMENT
1. For a detailed treatment of the notion of waste, see The Canon Production System: Creative Involvement of the Total
Workforce (New York: Japanese Management Association, Productivity Press, 1997).
2. M. Hammer and J. Champy, Reengineering the Corporation (London: Nicholas Brealey Publishing Ltd, 1993). pp39-44.
3. For a very good description of the productivity descriptions and improvement profiles of Japanese and U.S. automobile
manufacturers, see J. P. Womack, D. Jones, and D. Roos, The Machine that Changed the World (New York: Rawson
Associates, 1990).
4. B. Mahadevan, “Journey to the Deming: Implications for Corporate Transformation and Competitiveness,” IIMB Management
Review 17, no. 2 (2005): 59–70.
5. There are a number of studies that deal with the various factors required for JIT. For a representative paper on this, refer to B.
Mahadevan, “Are Indian Companies Ready for JIT?” IIMB Management Review 9, no. 2&3 (1997): 85–92.
6. For a detailed treatment of the subject of manufacturing architecture, refer to B. Mahadevan, The New Manufacturing
Architecture (New Delhi: Tata McGraw-Hill, 1999).
7. For a detailed explanation of issues related to focused manufacturing, refer to Chapter 3 of R. J. Schonberger, Building Internal
Chain of Customers (New York: Free Press, 1990), pp. 34–64.
8. Therefore, in this chapter we use set-up time to also denote changeover time. This is obvious because any changeover from one
variation to another requires a new set-up to be made for the manufacturing process.
9. For a detailed analysis of SMED, refer to S. Shingo, A Revolution in Manufacturing: The SMED System (Cambridge, Mass:
Productivity Press, 1985).
10. Please note that we have discussed the key elements of push type scheduling at great length in Chapters 14 to 18.
11. The fixed order quantity (FOQ) system is described in detail in Chapter 17 of this book. We merely use some of the results
from that discussion here to motivate the computation of number of Kanbans.
12. Various QC tools utilized for reducing the defects in an operating system are discussed in detail in Chapter 12 of this book.
13. For a good understanding of various issues related to Kaizen, readers are encouraged to visit the web site of the Kaizen
Institute at http://www.kaizen.com (Web site last accessed on June 25, 2014).
14. For example, see B. Mahadevan, “Are Indian Companies Ready for Just-in-Time,” IIMB Management Review 9, nos. 2&3
(1997): 85–92.
DEMAND FORECASTING
1. For more details, see S. Makridakis, A. Anderson, R. Carbone, R. Fildes, M. Hibon, R. Lewandowski, J. Newton, E. Parzen,
and R. Winkler, “The Accuracy of Extrapolation (Time Series) Methods; Results of a Forecasting Competition,” Journal of
Forecasting 1 (1982): 111–153.
OPERATIONS PLANNING
1. The term “heat” refers to one cycle of charging a furnace with all the required material so that molten steel is ready for pouring
into the required applications.
2. Organizations carry inventory for a variety of reasons. In this example, we build inventory in anticipation of the future surges
in demand. See Chapter 17 for a discussion on the other reasons for carrying inventory.
3. In the terminology of IT firms, it is referred to as “sitting on the bench”. There are cases of billable benches and non-billable
benches. A non-billable bench represents the UT costs arising out of lost productivity.
4. We do not make an effort to describe the solution methodologies available for solving transportation problems in this book.
Readers can refer to any standard textbook in operations research for details. However, we have briefly demonstrated the use
of the transportation model in Chapter 6. Refer to Section 6.3 and Example 6.4 for details.
5. LDR is also known as the HMMS method (from the first letter of the names of four authors). A detailed description of LDR is
available in C. C. Holt, H. Modigliani, J. F. Muth, and H. A. Simon, “Planning, Production, Inventories and Workforce,”
(Englewood Cliffs, NJ: Prentice Hall, 1960).
RESOURCES PLANNING
1. For a detailed treatment of this subject, readers are advised to consult Chapter 17 of this book.
2. Chapter 15 has a discussion on master production schedules (MPS) and the manner in which
information contained in the MPS could be put to use for the purpose of planning.
3. For a good treatment of this subject, readers are referred to the book, O. W. Wight,
Manufacturing Resource Planning: MRP–II (Essex Junction, VT: Oliver Wight Publishers,
1984).
INVENTORY PLANNING AND CONTROL
1. In Chapter 15 of the book, we show how build-up of this type of inventory is planned through
an aggregate planning exercise, and its cost implications.
2. It is easy to note that this level of inventory represents the reference for initiating action for
repeatedly replenishing the cyclic inventory. This is referred to as the reorder point (ROP).
3. The number of orders need not be strictly an integer, as it indicates an average value. In this
case, it merely means that 25 orders would have been placed every two years. However, we
round off the number of orders to the nearest integer just for convenience.
4. For instance, if the “in-house” manufactured item has a demand rate of d/unit time and the
production facility is capable of a production rate of p/unit time (p > d), then the economic run
length is given by:
For more details, please see Russell, R. S. and B. W. Taylor III (1998), Operations Management, 2nd ed. Prentice-Hall
International Inc. pp 581–584.
5. Note that variance is an unbiased estimator, whereas standard deviation is not. Therefore,
when the demand distribution data and lead time data pertain to different time periods, the
standard deviation needs to be converted to variance. After mathematical manipulations,
finally one can obtain the standard deviation. In simple terms, if the demand attributes (mean
and standard deviation) are in days and the lead time is L days, then an expressions for μ(L) and
σ(L) are given by μ(L) = μ × L and σ(L) = × σ where μ is the mean daily demand and σ is the
where TC(Q) is the total cost of stocking Q units and f(x) is the probability density function for demand. The first term on the
RHS corresponds to the total cost of overstocking and the second term corresponds to the total cost of understocking. By
taking the first derivative of the total cost function with respect to Q, and equating it to zero, one can arrive at an equivalent
expression for a continuous density function for demand. For more details see, Nahamias, S. (1997), Production and
Operations Analysis, 3rd ed., McGraw-Hill International ed. pp 272–275.
10. For a relevant discussion on this topic, please see, N. Venkatraman, “Multiplier Effect in
Inventory Management,”, Management Accountant, September 1986, pp 548–549.
OPERATIONS SCHEDULING
1. There is a plenty of literature available on the mathematical formulation and solution for the
loading problem. For an illustration of this, readers are referred to M. Berrada and K. E.
Stecke, “A Branch and Bound Approach for Machine Load Balancing in Flexible
Manufacturing Systems,” Management Science 32, no. 10 (1986): 1316–1335.
2. In this chapter, we demonstrate the use of the assignment method for loading using this
example. We assume that readers have a basic understanding of the solution methodology of
the assignment method. In any case, one can use a standard OR software such as LINDO to
obtain these solutions by appropriate input of the data. Interested readers may want to practice
the problems at the end of the chapter which deal with the assignment method.
3. Readers are referred to Chapter 9 of this book, which has more details on these alternative
configurations. Other configurations are also available, but we focus only on these two for the
purpose of discussing scheduling methodologies.
4. Panwalker and Iskander have reported nearly 300 scheduling rules. For more details, readers
are referred to S. S. Panwalker and W. A. Iskander, “A Survey of Scheduling Rules,”
Operations Research 25, no. 1 (1977): 45–61.
5. For Johnson’s original paper, please refer to S. M. Johnson, “Optimal Two and Three Stage
Production Schedules with Set-Up Time Included,” Naval Research Logistics Quarterly 1
(1954): 161–168.
6. Interested readers are referred to J. R. Jackson, “An Extension of Johnson’s Results on Job-
Lot Scheduling,” Naval Research Logistics Quarterly 3 (1956): 201–203.
7. The CDS method derives its name from the authors who proposed this variation of Johnson’s
rule. Interested readers are referred to H. C. Campbell, R. A. Dudek, and M. L. Smith, “A
Heuristic Algorithm for the n Job m Machine Sequencing Problem,” Management Science, 16,
no. 10 (1970): B630–B637.
8. “TAKT” is a word of German origin that means rhythm. The choice of this word is therefore
very appropriate. However, Taiichi Ohno, the father of the Toyota Production Systems, spells
this as “TACT”. Therefore, both spellings are used.
9. References of some of the books written by Goldratt are given at the end of the chapter.
10. As discussed in Chapter 16 of this book, the MRP logic does this job.
11. E. M. Goldratt and J. Cox, The Goal, (Madras: Productivity Press India Pvt. Ltd., 1993).
Six-Sigma Quality Control
1. This is based on the well-known central limit theorem. We skip the statistical aspects
pertaining to these details in this writing. However, interested readers can refer to any standard
book on business statistics for more details.
2. Alternative measurement scales are available. The nominal scale is a simple categorization of
observed values (such as good and bad). An ordinal scale is a method of ranking the
observations (for example from poor to excellent on a 5-point scale). A third option is to use a
ratio scale where exact measurements are possible (such as the length of a rod is 23.54 cm and
the length of another rod is 22.78 cm). Interested readers may refer to any standard textbook
on business statistics for more details. In this chapter, we deal only with nominal and ratio
scales.
3. If the standard deviation of the process (s ) is known, then the standard deviation of the sample
means is given as , where n is the size of the sample. The three stand-n ard deviations of
the sample means on either sides of the mean of the sample means provides the upper and
lower control limits for the chart.
4. There are several other benefits of higher Cp values. For instance, Juran and Gryna show that
the nature of inspection efforts required will be vastly reduced at higher Cp values. For details,
see J. M. Juran and F. M. Gryna, Quality Planning and Analysis (New Delhi: Tata McGraw-
Hill Co. Ltd, 1995), pp 393–396.
5. The probability of finding x defects in a sample of n drawn from a lot of N follows hyper-
geometric distribution— one of the well-known discrete probability distributions. However,
when the population is large compared to the sample, it can be approximated to a binomial
distribution. Therefore, the expression for the probability of finding x defects in a sample of n
with a proportion of defects in the lot as p is given by: P(x|n, p)5 nCx pxqn−x Since in a
sampling plan, up to c defects will be accepted, the probability of acceptance of a lot base on
probability for more details on the hyper-geometric distribution and its approximation to
binomial.
Copyright © 2015 Pearson India Education Services Pvt. Ltd
No part of this eBook may be used or reproduced in any manner whatsoever without the publisher’s prior written consent.
This eBook may or may not include all assets that were part of the print version. The publisher reserves the right to remove any
material in this eBook at any time.
ISBN 9789332547520
ePub ISBN 9789332541092
Head Office: A-8(A), Sector 62, Knowledge Boulevard, 7th Floor, NOIDA 201 309, India
Registered Office: 11 Local Shopping Centre, Panchsheel Park, New Delhi 110 017, India