See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281267267

Low Carbon Comprehensive Mobility Plan - Visakhapatnam

Book · November 2014

CITATIONS READS

8 2,122

4 authors, including:

Anvita Arora Ravi Gadepalli

Innovative Transport Solutions Pvt Ltd. International Association of Public Transport
14 PUBLICATIONS   82 CITATIONS    11 PUBLICATIONS   27 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Bicycle Partnership Program View project

Transportation View project

All content following this page was uploaded by Anvita Arora on 26 August 2015.

The user has requested enhancement of the downloaded file.

PROMOTING LOW CARBON TRANSPORT IN INDIA

LOW-CARBON
COMPREHENSIVE
MOBILITY PLAN
VISHAKHAPATNAM

2011 - 30
PROMOTING LOW CARBON TRANSPORT IN INDIA

Low – carbon Comprehensive Mobility Plan:

Vishakhapatnam

Authors
Lead author:

Dr. Anvita Arora

Contributing Authors:

Mr. Ravi Gadepalli, Mr. Parvesh Kumar Sharawat, Ms. Anusha Vaid, Mr. Abhishek Keshri

November 2014
UNEP DTU Partnership
Technical University of Denmark

This publication is part of the ‘Promoting Low-Carbon Transport in India’ project.

ISBN: 978-87-93130-25-8

Photo acknowledgement

Front cover photos: Authors’ own photographs

Disclaimer:

The findings, suggestions and conclusions presented in the case study are entirely those of
the authors and should not be attributed in any manner to UNEP DTU Partnership or the
United Nations Environment Programme, nor to the institutions of individual authors.
4
Acknowledgements
Innovative Transport Solutions Pvt. Ltd. (iTrans) expresses its sincere thanks to the UNEP DTU
Partnership on Energy, Climate and Sustainable Development, Technical University of Denmark, for
awarding the work of preparation of the “Low-Carbon Comprehensive Mobility Plan for Visakhapatnam
city” under the United Nations Environment Programme (UNEP) sponsored project on Promoting Low-
Carbon Transport in India.

Sincere thanks to Ms. Janaki Yuvraj Commissioner, Greater Visakhapatnam Municipal Corporation
(GVMC), Mr. MV Satyanarayana, former Commissioner GVMC, Mr. KVN Ravi, Executive Engineers and all
GVMC staff members for their guidance and support to the study team.

Special thanks are due to all our project partners, Prof. Dinesh Mohan and Prof. Geetam Tiwari of the
Transport Research and Injury Prevention Programme (TRIPP) at the Indian Institute of Technology (IIT),
Delhi, the Indian Institute of Management (IIM) Ahmedabad, Center for Environmental Planning and
Technology (CEPT), Ahmedabad and the Urban Mass Transit Company (UMTC), New Delhi for their
continued advice and support throughout this endeavour.

We acknowledge the contribution of Dr. Sarath Guttikunda from Urban Emissions for providing insights
into carbon emissions and SIM-air modelling-related issues.

iTrans wish to express their sincere thanks to Dr. Subash Dhar, UNEP DTU Partnership, Denmark for his
valuable suggestions and support during the preparation of this mobility plan.

Warm regards from the project team:

Dr. Anvita Arora Team Leader and CEO iTrans

Mr. Ravi Gadepalli Manager (Transport Engineering)
Mr. Parvesh Kumar Sharawat Manager (Projects)
Ms. Anusha Vaid Project Associate
Mr. Abhishek Keshri Administrator

5
Executive summary
Visakhapatnam is the second-largest urban agglomeration in Andhra Pradesh with a population of about
1.73 million (Census, 2011). It has been and continues to be a hub of economic activity in the region,
thereby leading to more and more people migrating to the city. Travel is derived demand of this
population, which engages in activities like work, education, recreation, health etc. for the sake of their
improved socio-economic wellbeing. Increased transport activity can however lead to an increase in
pollution, congestion and accidents in the city. The 534km2 under the jurisdiction of the Greater
Visakhapatnam Municipal Corporation (GVMC) is the planning area under consideration.

Out of the total planning area, only 31 per cent is currently built up and the rest is covered by
agriculture, forests and hills. As a result, the population densities are higher than 27,000 people per
hectare in the core city, and the outskirts are very sparsely populated. Also the city has 686 slums, which
together comprise 44 per cent of its total households. Therefore the city needs to plan for a transport
system that caters to its high density, low-income development. A Low-Carbon Mobility Plan (LCMP)
supported by the United Nations Environment Program (UNEP) has been prepared for the city, the
findings and recommendations of which are presented in this document.

A low-carbon vision for the city has been articulated as developing sustainable, low-carbon and safe
urban transport systems that provide access to the required goods, services and activities for all citizens.

As a part of this project, secondary and primary data was collected for the travel behaviour and other
parameters in the city, the results of which summarise the current situation and the key challenges in
the city. Current mode shares in the city are comprised of 52 per cent walk trips, followed by 18 per cent
bus usage, while car and 2-wheeler trips together contribute only to 17 per cent of the total trips in the
city. For females the private mode usage is even smaller, with 92 per cent trips made by walking, bus or
auto-rickshaw (an intermediate public transport). The city has a bus system with good network
connectivity and a fleet of 670 buses that cater to around 500,000 trips per day. Forty-two kilometres of
Bus Rapid Transit (BRT) corridors are also being developed, out of which 21km are currently operational.
Vehicle free Zones (VFZ) have been implemented on two stretches in the city from 5am to 7am everyday
where free cycles are provided to people during these hours. The city currently has up to 450 fatalities
annually, i.e. around 26 per lakh1 of the population, which is amongst the highest in the country. The city
only has 78km of footpaths out of the total arterial road length of 430km, and even these footpaths are

1
Lakh – 100,000. A unit in the Indian numbering system.

6
not continuous, universally accessible or properly lit at night. The funds allocated to walking and cycling
are less than one-hundredth of the total funds allocated for transport. Pedestrians are most vulnerable
to traffic fatalities, resulting in up to 200 fatalities per year by high speed trucks. The average household
size in Visakhapatnam is 4.0 people, the per capita trips per day is 1.6 and the average trip length in the
city is 4.1km. Based on this data, the Travel Demand Model for the city has been prepared and
calibrated for the base year using TransCAD 5.0 and the mode-wise vehicle kilometres travelled are
derived from that. These vehicle kilometres are further used as the input to estimate the aggregated
CO2 emissions in the city. The SIM-air tool is used for emissions and dispersion modelling, and the values
are calibrated based on the Ambient Air-Quality Measures (AAQM) in the city.

After the development of the baseline, a scenario analysis was carried out to visualise the mobility
situation in the Business as Usual (BAU) scenario and the Low-Carbon Development scenario

In the BAU scenario, the past trends of development are assumed to continue, and their likely impacts
on the congestion, emissions and traffic fatalities in the city are estimated. The year 2030 is taken as the
horizon year, the likely population growth is estimated as per United Nations projections, and the future
urban transport scenario as proposed in the master plan is taken as the input for travel demand
modelling. The likely mode shifts, vehicle kilometres travelled, and the resultant emissions have been
modelled. It is observed that the BAU scenario induces huge mode shifts away from public transport (PT)
and non-motorised transport (NMT) towards cars and 2-wheelers. This results in an increase of CO2
emissions in Visakhapatnam from 5.3 million tonnes in the base year to 7 million tonnes in 2030, i.e. a
32 per cent increase. This is due to the fact that the existing trends of development like segregated land
use, hostile environments for walking and cycling, and poor quality public transport encourage a shift
towards cars and 2-wheelers.

To mitigate such a high carbon pattern of growth, various sustainable transport initiatives have been
explored for the city, and those providing the maximum impact in terms of mobility, emission reduction
and safety have been identified. These include:

 Land use interventions

 Public transport interventions
 Non-motorised transport interventions
 Technology transfer interventions

7
Land use interventions include changes in policy norms relating to the usage type of various upcoming
developments, encouraging mixed land use and other such policy-related measures. These interventions
are to be implemented by the Visakhapatnam Urban Development Authority (VUDA).

The other three interventions involve the creation of new infrastructure in the city.

Public transport intervention, with an improved supply of public transport, is modelled for the likely
population and city structure in 2030 to observe the various high demand corridors in the city. It was
found that the best way to reduce the vehicle trips is to shift them from private modes of transport to
city buses as far as possible, and to auto-rickshaws if bussing is not feasible for the trip. Also, the
emissions per passenger kilometre are the least in buses followed by auto-rickshaws. Therefore, both
the carbon footprint and congestion on roads would decrease. Base year traffic fatality data shows that
travelling by bus or an auto-rickshaw is much safer than a 2-wheeler. In summary, shifting trips from
cars and 2-wheelers to buses and auto-rickshaws will be beneficial in terms of congestion, emissions and
also traffic safety. The additional fleet required in such a scenario is observed to be 1100 buses by 2030.
The fleet augmentation should also be linked with the following measures for it to be most useful:

 Cleaner vehicle technologies like buses running on compressed natural gas (CNG). Hybrid buses
need to be explored while determining the future fleet type.
 Better routing and scheduling of vehicles to improve their efficiency. The routes need to be updated
regularly, and dynamic scheduling systems should be used to optimise the available fleet.
 Buses need to be universally accessible, i.e. children, women, elderly people and wheelchair users
should also be able to use them. As many low-floor buses as possible should be added to the fleet to
make them more accessible.
 Supporting infrastructure, such as bus stops and access facilities at bus stops, needs to be provided.
Adequate bus depots and terminals need to be provided to cater to the increased fleet required for the
horizon year. VUDA needs to allocate the land for upcoming depots to the Andhra Pradesh State Road
Transport Corporation (APSRTC) for it to be functional.
The mass transit interventions needed on various key corridors in the city have been identified based on
the public transport demand. The current population of the GVMC is 1.7 million, the average trip length
is 4.1, and peak hour per direction trips (PHPDTs) exceed 15,000 only in two stretches, both of them less
than 5km in length, even in 2030. The guidelines suggest that these demands should be achieved in
2020 for the city to plan for a metro. Hence metro is not a feasible option for the city. However, based
on the available demand and existing population, Visakhapatnam qualifies in two of the three criteria
for a BRT system. Also, the city has a successfully operating BRT system, which should be expanded to
improve network connectivity. In addition to the mass transit corridors, a few other corridors have been

8
identified as their feeder routes in terms of promoting a regular city bus system, and also various other
measures which will help in increasing the ridership of the proposed BRT corridors.

Non-motorised transport (NMT) interventions focus on improving the condition of walking and cycling
by investing in safer walking and cycling infrastructure in the city. For the pedestrians, 430km of arterial
and sub-arterial roads in the city are proposed to have safe and adequate footpaths and crossings.
However, 41km of priority corridors for footpaths and safe crossings that need to be taken up
immediately are identified. All the footpaths to be developed need to follow the ‘urban road design
guidelines’ released by the Ministry of Urban Development. Currently, cycles cater to 3 per cent of the
total trips in the city. One of the major reasons for such limited usage is the lack of safety on roads for
the cyclists, and also the lack of parking for cycles in the city. On the arterial roads with right of way
(ROW) exceeding 30m, segregated cycle tracks are needed to separate the cyclists from high-speed
motorised traffic. On roads with shorter ROW, traffic-calming measures need to be taken up so that the
traffic speed is not too high compared to the speed of the cyclists, thereby increasing their safety. All
roads with a proposed ROW of 30m or higher, as per the VUDA master plan, are identified for the
provision of segregated cycle tracks. The remaining roads will also be made cycle-friendly by applying
measures like traffic-calming, better crossing facilities and other amenities like street lighting. The
immediate priority cycle tracks are already covered as part of the priority BRT corridors listed above, and
separate investment is not needed along these corridors. The GVMC has also identified a public bicycle
sharing (PBS) scheme as a means of promoting cycling further in the city. Two networks have been
identified that can act as a pilot for the implementation of PBS in the city.

The technology improvement scenario is considered in addition to the land-use, public transport and
non-motorised transport scenarios explained above. This scenario explains the likely impact of
improving future vehicle and fuel technologies on the city’s level of emissions. This includes improving
the energy efficiency of existing vehicle and fuel types, and introducing new fuel types as listed below:

 The existing fuels, i.e. petrol and diesel, are assumed to be upgraded to Bharat Stage IV by 2015
and Bharat Stage V by 2020. The resultant reduction in emission factors of vehicles are
incorporated in the SIM-air model for the 2030 scenario with land use, public transport and non-
motorised transport initiatives.
 The introduction of bio-fuel as an alternative to fossil fuels like petrol and diesel is explored in
this scenario. As per the projections of the International Energy Agency (IEA), the share of bio-
fuel in transport is taken as 9.5 per cent.

9
 The emissions in this scenario are seen to reduce to 0.40 (equal to base scenario) from 0.56 million
tonnes in the NMT improvement scenario, i.e. a 28.5 per cent reduction. The significance of
implementing all the scenarios is evident when compared with the BAU emission of 1.1 tonnes per
capita. By implementing all the scenarios, the total reduction in per capita CO2 emissions is 63.63 per
cent.

The likely benefits to the city if the above interventions are implemented are estimated and are
summarised in the following table.

Benefit analysis of proposed interventions

Base year BAU BAU+LU BAU+LU+PT BAU+LU+PT+N BAU+LU+PT+NMT
Scenario
MT +Technology
Horizon year 2011 2030 2030 2030 2030 2030
Population 2,946,000
1,730,320 2,946,000 2,946,000 2,946,000 2,946,000
Total trips 4,831,440
2,438,130 4,831,440 4,831,440 4,831,440 4,831,440
Car 2% 8% 4% 3% 3% 3%
2W 15% 25% 21% 12% 10% 10%
Mode share

Bus 19% 10% 11% 29% 25% 25%

Auto 9% 19% 22% 13% 8% 8%
Walk 52% 36% 38% 38% 45% 45%
Cycle 3% 2% 4% 4% 7% 7%
Car 572,149
181,982 1,442,475 728,319 572,149 572,149
2W 2,202,096
1,414,115 4,670,392 3,849,064 1,924,532 1,924,532
Veh-km travelled

Bus 506,251
166,768 173,932 188,943 437,523 437,523
Auto 718,725
250,779 1,049,113 1,191,440 464,099 464,099
Walk 1,291,196
887,479 1,217,523 1,291,196 1,520,230 1,520,230
Cycle 580,795
234,060 309,212 580,795 1,087,356 1,087,356
Total vehicles on 957,723
road 504,418 1,983,481 1,511,377 859,966 859,966
CO2 emissions 0.94 4.34 2.76 2.39 2.28 1.5
(million tonnes)
Per capita CO2 0.40 1.07 0.68 0.59 0.56 0.40
emissions
(tonnes)
Traffic 425 1,600 1,050 800 250 250
fatalities/year

10
The study shows that the CO2 emissions in the city can be brought down from 4.34 million tonnes in the
BAU scenario to 1.5 million tonnes in 2030 by implementing all the proposed measures, which would
mean that CO2 emissions will remain constant as in existing scenario in spite of the increase in the
population. It is also evident that the total vehicles added on to the roads in the best case scenario is
much less than the current trend, thereby reducing the huge investments required for road building and
maintenance to support the increased vehicles on the road. The traffic fatalities, which are a serious
public health hazard in the city, can even be brought down to half the existing number. In the long term,
the entire road network of the city should be universally accessible by all modes as per the code of
practice for urban roads. However, these are segregated into long-term, medium-term and short-term
recommendations to take them up in a phase-wise manner. The road network development of the city
forms the long-term intervention, and the mass transit corridors and their and feeder routes form the
medium-term intervention in the city. Also, a few priority walking and cycling projects are identified that
need to be implemented in the short term. Other than these, various urban transport-related policies
adopted by the Ministry of Urban Development (MoUD) at the national level are recommended to be
implemented as well.

11
Abbreviations
APPCB Andhra Pradesh Pollution Control Board
APSRTC Andhra Pradesh State Road Transport Corporation
ARAI Automobile Research Association of India
BAU Business As Usual
BCI Bicycling Compatibility Index
BRT Bus Rapid Transit
CBD Central Business District
CDP City Development Plan
CEPT Center for Environmental Planning and Technology
CGG Centre for Good Governance
CI Confidence Interval
DPR Detailed Project Report
FIR First Information reports
GIS Geographical Information System
GVMC Greater Visakhapatnam Municipal Corporation
HHI Household Information
HHI Household Interview
HPCL Hindustan Petroleum Corporation Ltd.
HPEC High Powered Expert Committee
ICI International Climate Initiative
IIMA Indian Institute of Management, Ahmedabad
IITD Indian Institute of Technology, Delhi
IPT Intermediate Public Transport
IRC Indian Road Code
iTrans Innovative Transport Solutions Pvt. Ltd.
KPH Kilometres per hour
LCD Low-Carbon Development
LCMP Low-Carbon Comprehensive Mobility Plan
LU Land use
MCA Multiple Classification Analysis

12
MIN Minutes
MoU Memorandum of Understanding
MoUD Ministry of Urban Development
NAPCC National Action Plan for Climate Change
NCRB National Crime Research Bureau
NGO Non-Government Organisations
NHAI National Highway Authority of India
NMSH National Mission for Sustainable Habitat
NTDPC National Transport Development Policy Committee
NUTP National Urban Transport Policy
OD Origin-Destination
PA Production-Attraction
PBS Public Bicycle Sharing
PCU Passenger Car Units
PHPDT Peak hour per direction trips
PPP Public Private Partnership
PT Public Transport
PUC Pollution under Control
ROW Right Of Way
RTA Road Transport Authority
TAZ Traffic analysis zones
TMC Turning Movement Counts
UMTA Metropolitan Transportation Authority
UN United Nations
UNEP United Nations Environment Programme

UTF Urban Transport Fund

Vizag Visakhapatnam
VMC Visakhapatnam Municipal Corporation
VUDA Visakhapatnam Urban Development Authority

13
Table of Contents
LIST OF TABLES ........................................................................................................................................ 15

LIST OF FIGURES ..................................................................................................................................... 18

1 Introduction ........................................................................................................................................ 21

2 City Status as of 2012 .......................................................................................................................... 24

2.1 Delineation of Planning Area ......................................................................................................... 24

2.2 City Profile ...................................................................................................................................... 24

2.3 City Structure ................................................................................................................................. 25

2.4 Data Collection ............................................................................................................................... 27

2.5 Base Year Transport Indicators ...................................................................................................... 53

3 Base Year Scenario .............................................................................................................................. 71

3.1 Base Year Travel Demand Model ................................................................................................... 71

3.2 Base Year Emissions Model............................................................................................................ 89

4 Scenario Analysis – Business As Usual ................................................................................................ 92

4.1 Population and Trip Rate Projections ............................................................................................ 92

4.2 BAU Scenario.................................................................................................................................. 94

4.3 BAU Scenario Summary ............................................................................................................... 105

5 Low-Carbon Development Scenarios ................................................................................................ 107

5.1 Land Use (LU) Scenario ................................................................................................................ 108

5.2 Public Transport (PT) Scenario ..................................................................................................... 114

5.3 Non-Motorised Transport (NMT) Scenario .................................................................................. 118

5.4 Technology Improvement Scenario ............................................................................................. 122

5.5 Summary of Scenario Analysis ..................................................................................................... 124

6 Recommendations ............................................................................................................................ 126

6.1 Long-Term Intervention: Road Network Development ............................................................... 126

14
 6.2 Long-Term Intervention: Land Use Interventions........................................................................ 127

6.3 Long-Term Public Transport Interventions .................................................................................. 127

6.4 Medium-Term Intervention: Mass Transit Corridors................................................................... 130

6.5 Medium-Term: Feeder Route Network Development ................................................................ 133

6.6 Short-Term NMT Interventions.................................................................................................... 134

6.7 Policy Recommendations............................................................................................................. 137

6.8 Implementation Plan ................................................................................................................... 138

6.9 References ................................................................................................................................... 139

Annexure-I: Survey Formats ..................................................................................................................... 140

Household Survey Part I: (Revealed Preference Survey) ...................................................................... 140

Household Survey Part II: (Stated Preference Survey) ......................................................................... 144

Traffic Volume Count Survey Format .................................................................................................... 149

Origin-Destination (OD) Survey Format ................................................................................................ 150

Road Inventory Survey .......................................................................................................................... 151

Auto-Rickshaw Operator Survey ........................................................................................................... 157

Petrol Pump Survey Format .................................................................................................................. 159

LIST OF TABLES

Table 1 Land use patterns in Visakhapatnam ............................................................................................. 26

Table 2 List of primary surveys and secondary data collected for the LCMP ............................................. 27

Table 3 Travel behaviour – household information (to be collected through household survey) ............. 28

Table 4 Income categorisation of households ............................................................................................ 33

Table 5 Income categorisation of households zone-wise ........................................................................... 33

Table 6 List of intersections for turning movement counts........................................................................ 34

Table 7 PCU values (Source: IRC SP 41) ...................................................................................................... 37

15
Table 8 Sample size of trips captured in OD surveys .................................................................................. 47

Table 9 Major outcomes from IPT operator survey .................................................................................... 51

Table 10 Summary of petrol pump survey data ......................................................................................... 52

Table 11 Trip purposes in the city ............................................................................................................... 53

Table 12 Mode shares in Visakhapatnam ................................................................................................... 54

Table 13 Trip purpose-wise mode shares ................................................................................................... 55

Table 14 Trip purpose-wise mode shares disaggregated by gender .......................................................... 55

Table 15 Trip length distribution in Visakhapatnam ................................................................................... 57

Table 16 Mode-wise average trip length .................................................................................................... 58

Table 17 Income-wise and purpose-wise travel time (in minutes) ............................................................ 59

Table 18 Average speeds of various modes along arterial roads ............................................................... 60

Table 19 Access-egress trip times for PT & IPT ........................................................................................... 60

Table 20 Risk exposed to, and risk imposed by, each mode ...................................................................... 62

Table 21 Length and width of available footpaths in Visakhapatnam........................................................ 63

Table 22 Quality of footpath infrastructure ............................................................................................... 64

Table 23 Mode-wise cost of commute ....................................................................................................... 65

Table 24 Proportion of household income spent on travel for various modes .......................................... 66

Table 25 Distribution of land allocated for transport activity .................................................................... 66

Table 26 Item-wise transport infrastructure investments.......................................................................... 67

Table 27 Mode-wise tax burden ................................................................................................................. 67

Table 28 Fuel prices in the city.................................................................................................................... 68

Table 29 Modelling components and input sources................................................................................... 71

Table 30 Trip rate table used for trip production ....................................................................................... 79

Table 31 Regression statistics for work attraction...................................................................................... 83

Table 32 Regression statistics for education attraction ............................................................................. 83

16
Table 33 Regression statistics for other trips attraction............................................................................. 83

Table 34 Calibrated deterrence function for trip distribution .................................................................... 86

Table 35 Mode-wise average occupancy .................................................................................................... 86

Table 36 Mode-wise summary of base year model .................................................................................... 89

Table 37 Base year SIM-air inputs............................................................................................................... 91

Table 38 Base year transport emissions ..................................................................................................... 91

Table 39 Population and trip projections for horizon years ....................................................................... 93

Table 40 Trip length vs mode share in Visakhapatnam: base year............................................................. 99

Table 41 Household income vs mode share in Visakhapatnam: base year .............................................. 100

Table 42 BAU scenario income projections .............................................................................................. 100

Table 43 BAU scenario trip length range .................................................................................................. 100

Table 44 BAU scenario – 2030 aggregate mode shares............................................................................ 101

Table 45 SIM-air inputs for BAU scenario: 2030 ....................................................................................... 103

Table 46 2030 BAU scenario emissions .................................................................................................... 103

Table 47 2030 BAU scenario traffic fatalities ............................................................................................ 105

Table 48 Land use scenario trip length distribution ................................................................................. 110

Table 49 Land use scenario mode shares ................................................................................................. 111

Table 50 SIM-air inputs for the land use scenario .................................................................................... 112

Table 51 2030 land use scenario emissions .............................................................................................. 113

Table 52 2030 land use scenario traffic fatality projections ..................................................................... 113

Table 53 2030 mode shares – PT scenario................................................................................................ 116

Table 54 SIM-air inputs for the PT scenario .............................................................................................. 117

Table 55 2030 PT scenario emissions ....................................................................................................... 117

Table 56 2030 traffic fatalities – PT scenario ............................................................................................ 118

Table 57 2030 mode shares – NMT scenario ............................................................................................ 120

17
Table 58 SIM-air inputs for the NMT scenario .......................................................................................... 121

Table 59 2030 NMT scenario emissions.................................................................................................... 121

Table 60 2030 traffic fatalities – NMT scenario ........................................................................................ 122

Table 61 SIM-air inputs for the NMT scenario .......................................................................................... 123

Table 62 2030 NMT scenario emissions.................................................................................................... 124

Table 63 Summary of scenario analysis .................................................................................................... 125

Table 64 Selection criteria of mass rapid transit modes .......................................................................... 130

Table 65 Salient features of selected BRT corridors ................................................................................. 132

LIST OF FIGURES

Figure 1 Visakhapatnam – pockets of development .................................................................................. 22

Figure 2 Planning area and road network of the GVMC ............................................................................. 24

Figure 3 Age-sex pyramid of Visakhapatnam ............................................................................................. 25

Figure 4 Zonal segregation adopted for household interviews .................................................................. 30

Figure 5 Location of inner cordon points for TMC ...................................................................................... 35

Figure 6 Vehicular mode shares at inner cordon points ............................................................................. 38

Figure 7 Hourly variation of passenger vehicles (in PCUs) at inner cordon points ..................................... 40

Figure 8 Hourly variation of goods vehicles (in PCUs) at inner cordon points ........................................... 40

Figure 9 Hourly variation of total non-motorised vehicles at inner cordon points .................................... 41

Figure 10 Hourly variation of pedestrians at inner cordon points.............................................................. 41

Figure 11 Location of outer cordon points for TMCs .................................................................................. 42

Figure 12 Vehicular mode shares at inner cordon points ........................................................................... 43

Figure 13 Hourly variation of passenger vehicles (in PCUs) at outer cordon points .................................. 44

18
Figure 14 Hourly variation of goods vehicles (in PCUs) at outer cordon points ......................................... 45

Figure 15 Hourly variation of total non-motorised vehicles at outer cordon points.................................. 45

Figure 16 Hourly variation of pedestrians at outer cordon points ............................................................. 45

Figure 17 Links selected for road inventory data collection ....................................................................... 48

Figure 18 Inadequate footpath infrastructure along the BRT corridor ...................................................... 49

Figure 19 Inadequate bus stop facilities ..................................................................................................... 49

Figure 20 On-street parking observed on most arterials ............................................................................ 50

Figure 21 Mode-wise fuel type distribution ............................................................................................... 53

Figure 22 Income group-wise mode share ................................................................................................. 56

Figure 23 Mode-wise trip length distribution (in km)................................................................................. 59

Figure 24 Traffic fatality – victim and impacting vehicle types .................................................................. 61

Figure 25 Inadequate street infrastructure along the NH-5 ....................................................................... 63

Figure 26 Gender-wise safety perception................................................................................................... 65

Figure 27 Annual transport expenditure pattern ....................................................................................... 69

Figure 28 TAZ map showing the 97 zones .................................................................................................. 73

Figure 29 Property data overlapped with the TAZ map ............................................................................. 74

Figure 30 Road network used for modelling (including connectors) .......................................................... 75

Figure 31 Multiple classification analysis trip rate charts for a combination of three variables................ 78

Figure 32 Multiple classification analysis trip rate charts for combination of two variables ..................... 78

Figure 33 Multiple classification analysis trip rate charts for one variable ................................................ 79

Figure 34 Residential buildings across the GVMC area .............................................................................. 80

Figure 35 Commercial developments across the city ................................................................................. 81

Figure 36 Industrial developments across the city ..................................................................................... 82

Figure 37 Calibrated network flows for the base year travel demand model ............................................ 88

Figure 38 Base year cell-wise vehicle kilometres travelled ........................................................................ 90

19
Figure 39 Land use plan according to the master plan – 2030, Visakhapatnam ........................................ 95

Figure 40 2030 BAU population densities (km2) ......................................................................................... 96

Figure 41 Proposed ROWs as per the master plan ..................................................................................... 97

Figure 42 Trip production and attractions for BAU – 2030 ........................................................................ 98

Figure 43 BAU Scenario – 2030 network loads ......................................................................................... 102

Figure 44 Emission pathways to reach global temperature targets ......................................................... 107

Figure 45 TAZ-wise growth factors for land use scenario population growth.......................................... 109

Figure 46 Land use scenario trip productions and attractions ................................................................. 110

Figure 47 Land use scenario – 2030 network loads .................................................................................. 111

Figure 48 PT scenario corridor loads ........................................................................................................ 116

Figure 49 Proposed ring-radial pattern of growth for the road network ................................................. 127

Figure 50 Location of supporting infrastructure for buses ....................................................................... 129

Figure 51 Existing and proposed BRT corridors in the city ....................................................................... 131

Figure 52 Feeder route network in addition to the existing and proposed BRT network ........................ 133

Figure 53 Priority footpaths in the city ..................................................................................................... 135

Figure 54 Links with ROW>30m for segregated cycle tracks .................................................................... 136

Figure 55 PBS networks identified in the GVMC....................................................................................... 136

20
1 Introduction
Visakhapatnam is the second largest urban agglomeration in Andhra Pradesh with a population of about
1.73 million (Census, 2011). The city has been, and continues to be, a hub of economic activity in the
region, and hence its development is critical to the overall development of the region. This urban
development has led to more and more people migrating to the city. Travel is a derived demand of this
population, which engages in activities like work, education, recreation, health etc., for the sake of
improving their socio-economic wellbeing. Therefore, providing adequate transport infrastructure is an
integral part of the development of the city. Increased transport activity can however lead to an
increase in pollution, congestion and accidents in the city. Hence, there is a need for appropriate
planning, which can ensure safe mobility and accessibility to people irrespective of their socio-economic
background in a way that does not compromise the health of the environment.

The United Nations Environment Programme (UNEP) is implementing a project on ‘Promoting Low-
Carbon Transport in India’ as a part of which ‘Low-Carbon Comprehensive Mobility Plans’ (LCMP) are
being prepared for three cities across the country. This initiative aims to address transportation growth,
development challenges and climate change issues in an integrated manner. Visakhapatnam has been
selected as one of the case study cities for the preparation of LCMP, and UNEP has signed a
Memorandum of Understanding (MoU) with the Greater Visakhapatnam Municipal Corporation (GVMC)
for the same. The project is supported by the International Climate Initiative (ICI) of the German
Government as well as the implementing partners including UNEP, the UNEP DTU Partnership in
Denmark (UDP), Indian Institute of Technology, Delhi (IITD), Indian Institute of Management,
Ahmedabad (IIMA) and CEPT University, Ahmedabad2. Innovative Transport Solutions (iTrans) Pvt. Ltd.3
carried out the study in partnership with the Greater Visakhapatnam Municipal Corporation (GVMC).

Data has been collected through primary surveys and secondary sources in order to understand the
existing urban transport scenario in Visakhapatnam and to develop the indicators of sustainable
transport, which were developed as a part of the LCMP methodology. The 534km2 area under the
jurisdiction of the Greater Visakhapatnam Municipal Corporation (GVMC) has been selected as the study
area for the LCMP. The following figure shows the major pockets of development of Visakhapatnam,
also known as ‘Vizag’.

2
UNEP (2014). The future of low carbon transport in India. From http://www.unep.org/transport/lowcarbon
3
iTrans (2010). Homepage. www.itrans.co.in

21
 Figure 1 Visakhapatnam – pockets of development

Based on the findings of the data collected, travel demand and air pollution modelling has been carried
out for the city. These models are developed for the base year, i.e. the existing scenario, and for the
horizon year in a Business As Usual (BAU) scenario, i.e. if the growth pattern of the city continues as it
exists now and no low-carbon interventions are made. A long-term planning horizon of 20 years is
adopted for the study, and hence 2030 is considered as the horizon year for planning the future
scenarios of development. The drawbacks in such a scenario are identified and various interventions
needed to lead the city in a low-carbon growth path are tested for the likely mode shift they may cause
and the resultant reduction in the carbon footprint of the city. These interventions include changes in
land-use patterns, public transport improvement, non-motorised transport, i.e. better walking and
cycling, technology and policy related interventions.

22
Chapter 2 presents the city status as of 2012, i.e. the baseline scenario for Vizag. Chapter 3 presents the
likely transport and emission scenario in 2030 in a Business as Usual (BAU) scenario of development.
Chapter 4 presents the various alternative scenarios of development for the low-carbon development of
the city, and the project recommendations identified from these scenarios are presented in Chapter 5.
Chapter 6 provides recommendations

23
2 City Status as of 2012
This chapter presents the existing situation of the city’s transportation systems, considering 2012 as the
baseline population of the city.

2.1 Delineation of Planning Area

The Greater Visakhapatnam Municipal Corporation (GVMC) area of 534km2 represents the urban
agglomeration area of the city, and is considered the study area of the Low-Carbon Mobility Plan
(LCMP). The city has large pockets of land covered by hills, forests, the port and industries. Excluding
these areas, the urban built-up area of the city is concentrated in 166km2 spread across the 534km2 of
the total city area. Figure 2 shows the planning area, i.e. the GVMC boundaries and its road network.

Figure 2 Planning area and road network of the GVMC

2.2 City Profile

From 1858, when its ‘Municipal Association’ was formed, to 2012, the city has developed into a
significant economic, educational, health and tourism hub for the people of north-eastern Andhra

24
Pradesh and southern Orissa. The city is well connected nationally and internally by the National
Highway (NH-5) passing through it. It is also well connected by railways, the airport and sea port. The
city is also an education, health and tourism hub for the region, and hence attracts a daily floating
population to access these services.

The demographic profile of the city shows the presence of a largely young population, with 68 per cent
under 40 years of age. The age-sex pyramid of the city shown below gives the proportion of men and
women in the city in various age categories. The 1.73 million population of the city comprises of 875,000
males and 855,000 females, i.e. 977 females per 1000 males in the city. (Census 2011)

Figure 3 Age-sex pyramid of Visakhapatnam

85+
80-84
75-79
70-74
65-69
60-64
55-59
50-54
Age class

45-49
40-44
35-39
30-34
25-29
20-24
15-19
10-14
5-9
Under 5

12% 10% 8% 6% 4% 2% 0% 2% 4% 6% 8% 10% 12%

% of Male % of Female

(Source: Household Interview Surveys)

2.3 City Structure

Greater Visakhapatnam Municipal Corporation (GVMC) was formed in 2005 by merging the erstwhile
Visakhapatnam Municipal Corporation (VMC), Gajuwaka municipality and 32 other villages. The
following sections explain the existing city structure.

25
2.3.1 Land use patterns

Table 1 shows the existing land use pattern in the city. As explained above, only 166km2 out of 534km2
in the city is currently built up, while the rest is covered by hills, forests and water bodies. The usage
pattern of the built-up area is shown below. It is observed that industries occupy 57 per cent of the city
area, forming the predominant land use type. Among other land uses, residential use is the major land
use type while commercial and public spaces only occupy 1 per cent each. This is disproportionately low
compared to the residential areas. Even these commercial areas are all concentrated in the core city
area, thereby inducing long-distance trips from the people residing in the outskirts of the city. There is a
need to encourage mixed land use and increase the commercial areas in all outgrowths of the city.

Table 1 Land use patterns in Visakhapatnam

Land use type Built up area (in km2) % area occupied

Industrial 95 57%

Residential 38 23%

Commercial 2 1%

Public spaces 1 1%

Roads 17 10%

Railways+ airport + bus terminals 11 6%

Other 2 1%

Total 166 100%

Data Source: VUDA, April, 2007, Revised Master Plan for Visakhapatnam Metropolitan Region – 2021

2.3.2 Regional linkages

Situated almost midway between Chennai in the south (762km) and Kolkata in the north (879km), this
region occupies an important place in the development of modern Andhra Pradesh. To the region’s east
is the Bay of Bengal, while in the north is Srikakulam district, in the southeast is Godavari district of
Andhra surrounds. In the region’s western side lies the rest of Visakhapatnam district, which is
surrounded by the state of Orissa in its extreme western limits. There are five major urban centres:
Visakhapatnam, Gajuwaka, Anakapalli, and Bheemunipatnam in Visakhapatnam district, and
Vizianagaram in Vizianagaram district. The principal city in this region is Visakhapatnam. The state

26
capital of Hyderabad is 637km from Vizag, well connected with road, railway and air. (Source: VUDA,
April, 2007, Revised Master Plan for Visakhapatnam Metropolitan Region – 2021)

2.3.3 Population density

With a population of 1.73 million (Census 2011) and a total area of 530km2, the average population
density of the city is around 3,300 people per square kilometre. However, since the effective built up
area is only around 166km2, the effective population density in the city is around 10,400 people per
square kilometre. Within this, 50km2 of the core city area holds up to 50 per cent of the city population
and has a very high population density of 27,000 people per square kilometre. Some wards in the core
city even have a density of 60,000 people per square kilometre. In summary, the population density is
concentrated mostly in the core city area while the outgrowths have a sparse density.

2.4 Data Collection

Data has been collected from primary surveys and secondary sources in order to understand the existing
urban transport scenario in Visakhapatnam, and to develop the indicators of sustainable transport,
which were developed as part of the LCMP methodology. The following table gives a brief summary of
the primary surveys and secondary data collected. The data collection exercise was carried out between
August 2012 and March 2013.

Table 2 List of primary surveys and secondary data collected for the LCMP4
Source Data collected Sample size/source
Household interviews 3,100 households
Primary data

Traffic volume counts 19 intersections

OD surveys 8 locations
Road inventory Arterial roads
IPT driver profile 500 Autos
Petrol pump survey 600 Vehicles
CDP, master plan and BRT DPR GVMC, VUDA
Secondary data

Ambient air quality measures APPCB

Property tax data for individual households GVMC
Annual budget and accounts data GVMC
Bus route and operations data APSRTC

4
GVMC: Greater Visakhapatnam Municipal Corporation. VUDA: Visakhapatnam Urban Development Authority.
APPCB: Andhra Pradesh Pollution Control Board. APSRTC: Andhra Pradesh State Road Transport Corporation.
NCRB: National Crime Research Bureau. RTA: Road Transport Authority. HPCL: Hindustan Petroleum Corporation
Ltd.

27
 Safety data Traffic Police, NCRB
Vehicle inventory data RTA, APSRTC
Energy consumption HPCL
This section explains the detailed methodology and primary surveys. The formats followed for each of
these surveys are provided in the Annexure.

2.4.1 Household interview (HHI) surveys

For the current study, a household survey is designed keeping in mind the need to capture different
social groups effectively and understand the perception of people towards different modes of transport
in terms of time, cost, comfort, safety and security. The questionnaire is divided into two parts:

1. General or household questionnaire

2. Individual questionnaire

The household information (HHI) questionnaire is designed to capture general information about
household demographics (individual age, gender, education, and occupation), income, expenditures and
assets, vehicles owned and existing intra-city travel behaviour. The questionnaire is designed to capture
information that is useful in identifying social groups and accordingly distinguish travel behaviour, mode
choice preferences and needs. However, it is also required that the household questionnaire does not
create so many subgroups that the analytical results become statistically insignificant.

Information on the individuals is also included in the household questionnaire survey. It is so designed
that it may or may not be answered by him/her directly, and may be answered by the head of the
household as per the situation. Based on the identified indicators for LCMP, it is required to collect
information regarding the existing use of modes, its availability and the criteria related to safety,
security and cost. Also, the survey was designed to collect trip chain data and capture details for multi-
modal use, and include information like access and egress mode, distance, travel time and cost.

Table 3 Travel behaviour – household information (to be collected through household survey)
Data required Description
Personal information Age
Gender
Occupation
Monthly income
Migration status
Caste & religion
Vehicle ownership and age of vehicle

28
 Monthly expenditure on transport
Trip-making information for the entire day Trip purpose
Trip origin
Trip destination
Travel distance
Mode used
Access mode
Egress mode
Distance to access public transport stop
Distance of egress public transport stop
Travel time to access
Travel time to egress
Average waiting time for public transport
Total travel time
Total travel cost
Average mileage if car/two-wheeler is used
Fuel used
Reason for using the current mode used
Transport infrastructure rating for different modes Perception about safety
Perception about security
Perception about comfort
Perception about cost

The household interview (HHI) surveys capture the socio-economic and trip-making characteristics of
each individual in the household during the course of the whole day. They form the basis for developing
the mobility and accessibility indicators, and also the future planning and modelling exercise to be
carried out for the study. Hence, it is critical for the HHI sample to be representative of the households
in the entire city and for the quality of data to be accurate.

2.4.1.1 Household survey sampling

The city of Visakhapatnam has grown organically over the years based on its economic activity. The
initial development of the city started from its port, and the rest of the city has grown from there over a
period of time. It is necessary to collect data from all wards to capture the difference between wards,
and also different income groups within each ward to cater to the income level heterogeneity that exists
within each ward. Consequently 65 out of the total 72 municipal wards in the city have been selected for
the survey. The other seven wards are located at the outskirts of the city and have very little

29
development and hence have been excluded from the survey. However, the horizon year population
projections can be carried out based on the growth trends of the rest of the city, and subsequent travel
and infrastructure needs can be modelled.

These 65 wards are segregated into nine different zones, based on their proximity to the central
business district (CBD). This is to ensure that different trip-making characteristics that exist in different
parts of the city are all captured. The figure below shows the various zones that are selected for the
study, and the number of wards within each zone. While this segregation won’t have an effect on the
HHI sampling, it will help in the HHI data analysis and modelling exercise to establish the heterogeneity
in trip-making characteristics within the city.

The sample has to address three types of heterogeneity that exists in the city:

 Different types of land use settlements, i.e. the older parts of the city having low-rise, high density
development, and newer areas developed in the last two decades comprised of high rise buildings.
Also, because of the vast area of the city, various pockets have different building characteristics.
 Different trip-making characteristics, i.e. people near the CBD having shorter trip lengths, people
living in the outgrowths having to travel longer distances to reach their workplaces and educational
institutions concentrated in the CBD, and the industrial area having separate travel patterns and so
on.
 Income heterogeneity that exists between different households within each area.

Figure 4 Zonal segregation adopted for household interviews

30
The actual sample size within each ward has been selected based on a three-level hierarchy as explained
below.

City-level statistical representativeness of sample

The sample that is collected should be statistically representative of the total number of households in
the city. The municipal corporation maintains a GIS-based property tax database of the city, which gives
the ward-wise list of each and every property that exists in the city, its land-use pattern and also the
number of assessments, i.e. individual households within each property. The total number of properties
in the city under residential land use has been derived from the total data base, and it is observed that
the total number of households in each ward add up to 419,343 households in the city. A total sample
size for the city is determined, initially assuming a normal distribution, which is later distributed into
each ward. The minimum sample size for statistical representativeness at 95 per cent confidence
interval is 1,885 households.

Ward-wise distribution of sample size

The total sample size is distributed across all wards in proportion to their total households, derived from
the GIS-based property tax data. However, 1,885 is only 0.44 per cent of the total households in the city,
and when distributed across all wards is too low to be statistically significant in many wards. The GIS-

31
based data is used to observe the total number of households in each ward, out of which a statistically
significant sample size of households is estimated for survey. On the basis of sample theory, the sample
size for each ward is calculated and for a 95 per cent confidence level. Hence, the total sample has been
increased in such a way that all the wards have a statistically significant sample. Such a sample needs
3,100 households, i.e. 0.7 per cent of total households in the city to capture the heterogeneity.

Representing the income-level heterogeneity within each ward

The next task is to distribute the ward-wise sample within each ward among different income groups.
The property tax of each household is considered to be an indicator of household income, i.e. assuming
that with increasing income people live in bigger houses paying higher property tax. Property tax paid by
households is segregated into six different slabs, and the total number of households in each slab has
been calculated for all the wards. The proportion of households in each of these slabs is an indicator of
the proportion of households in each income group for the ward.

2.4.1.2 Household survey methodology

A professional survey agency with more than five years of experience in urban transport surveys was
engaged to carry out the household interviews. A team of 25 graduates well-versed with the city of
Visakhapatnam and the local knowledge of Telugu were recruited for the surveys. A five day training
program, which included pilot surveys, was carried out for the enumerators. The questionnaire format,
as given by the United Nations Environment Program (UNEP) toolkit for the preparation of a Low-carbon
Comprehensive Mobility Plan (LCMP), was translated to Telugu and followed for the survey. Surveyors
were sent to the field every day with information on the number of different types of households to be
covered in each ward. The surveyor would sit with the available members of the household and collect
the required data of all members.

The enumerators dispersed to various localities within the ward and selected households for interview
based on random sampling. The questionnaire had two parts: a revealed preference and stated
preference survey. The revealed preference part comprised of questions regarding the socio-economic
and trip-making characteristics of each individual in the household for the entire duration of a typical
working day. The overall activities performed by each individual in the household on the day before the
survey was captured. The stated preference part of the questionnaire showed the various future
scenarios for transport in the city, and asked the people about their willingness to shift to different
modes in such a scenario. The total questionnaire had 106 entries for each individual. The surveys were

32
carried out between July 2012 and September 2012. The data collection site was later coded into
spreadsheets for further analysis. Three levels of cross-checking were carried out to minimise the
manual errors made during data entry. Data for some of the households were discarded due to
incomplete survey forms, and the remaining 3,021 households with complete data were used for further
analysis.

The household income data revealed in the surveys did not match with the household assets data,
showing that people did not share the correct income details. Therefore to understand the income-wise
variation in travel patterns, the total households surveyed have been divided into three income
categories based on the type of household they live in and the assets owned in each household. The
household interview captured the type of household and the various assets owned in the household
including a car, 2-wheeler, mobile phone, refrigerator, LPG stove, air cooler, air conditioner, television
and a computer. The type of household is taken as the primary indicator of the income of the
household. Within the lower income range households, the ones owning assets similar to higher income
households are put in the next highest range. Care was also taken to have an adequate number of
households in each income class.

Table 4 Income categorisation of households

S. No. Range Household type Households

interviewed
1,252
1 Low income i) Huts or ordinary tiled roof households who
don’t own a refrigerator
2 Middle income i) Masonry terraced or masonry tiled buildings 1,263
who don’t own either an air conditioner or a
car
ii) Low income households owning a refrigerator

3 High income i) Apartments or RCC posh buildings 542

ii) Middle-income households owning an air-
conditioner or a car

Total Households 3,057

Table 5 Income categorisation of households zone-wise

Range Zones

33
 1 2 3 4 5 6 7 8 9

Low income 159 70 120 170 202 223 69 218 22

Middle income 115 23 110 243 252 187 112 204 16

High income 55 23 57 102 112 46 40 103 5

2.4.2 Traffic volume counts

Traffic volume counts have been carried out at various locations of the city to understand the existing
traffic situation using a manual method of counting. Turning movement counts (TMCs) have been
carried out at 20 intersections, out which seven were outer cordon points and 13 were screen line
points. Combining data from each of these points, the internal and external movement in the city has
been estimated. Furthermore, these 20 intersections are comprised of 76 arms, which in effect give the
traffic data at 76 mid-blocks in the city. Also, pedestrian movement counts have been carried out at all
the above intersections to derive the demand for footpaths and crossing facilities at these intersections.
The TMCs were carried out using a manual method of survey with adequate number of enumerators
and supervisors posted at each survey location. The surveys were carried out between August 2012 and
September 2012.

Out of the 20 intersections, 4 have been selected at the outer cordon points of the city, i.e. the four
entry points to the study area, and the rest of the 16 are locations within the city. The 16 locations
within the city are finalised in consultation with the GVMC engineers in such a way that together they
represent the major intersections in the city, and they also represent imaginary screen lines that divide
the traffic movements in the city from north to south, east to west, between two major regions, etc. The
following table gives the list of intersections selected for turning movement counts (TMCs).

Table 6 List of intersections for turning movement counts

Name of intersection Location Type of intersection
Palm Beach Inner Cordon 3 armed
Seven Hills Hospital Inner Cordon 3 armed
Purna Market (To Police Barracks) Inner Cordon 3 armed
Gopalapatnam Inner Cordon 3 armed
Hanumanthavaka Inner Cordon 4 armed
Maddilapalem Inner Cordon 4 armed

34
Siripuram Inner Cordon 4 armed
Asilmetta Inner Cordon 4 armed
Diamond Park Inner Cordon 4 armed
Jagadamba Inner Cordon 4 armed
Dolphin Hotel Inner Cordon 4 armed
Convent Junction Inner Cordon 5 armed
Collector's Office Junction Inner Cordon 4 armed
NAD Junction Inner Cordon 4 armed
Dabagardens-Ambedkar Statue Inner Cordon 5 armed
Gurudwara Inner Cordon 4 armed
Bheemunipatnam Outer Cordon 3 armed
Thagarapuvalasa Outer Cordon 3 armed
Lankelapalem Outer Cordon 4 armed
Gajuwaka Circle Outer Cordon 4 armed

2.4.2.1 Inner cordon TMCs

Sixteen locations in the inner cordon of the city have been selected for TMCs, which represent the major
intersections in the city and they also represent imaginary screen lines, which divide the traffic
movements in the city from north to south, east to west, between two major regions, etc. Sixteen-hour
TMCs have been carried out at each of these intersections on a typical working day. These 16 locations
are well spread out throughout the city, as can be seen from the following figure showing the locations
of the survey points on the map.

Figure 5 Location of inner cordon points for TMC

35
The mode shares, hourly variation of passenger vehicles (including two-wheeler, car, taxi, bus, auto-
rickshaw), goods vehicles (2, 3 and multi-axel trucks, LCV, goods-rickshaw, tractors) and non-motorised
vehicles (including bicycle, cycle-rickshaw, animal-drawn) at each of the screen line points are presented
in the following figures. The salient features observed from the traffic flow trends are:

 Auto-rickshaws form the majority of the total vehicles at most intersections, followed by two-
wheelers and cars. Buses, in spite of carrying the maximum number of passengers, are observed to
be in the minority except at intersections along the NH-5, which are observed to have a significant
proportion of buses. On average, buses occupy 10 per cent of road space while they cater to 37 per
cent of the vehicular trips in the city, as observed from household interviews. This indicates the
efficiency of space utilisation of the city-bus system.
 The peak hour volume of pedestrians reaches up to 3000 at some of the intersections, whereas the
infrastructure provided for crossing facilities is minimal.
 The vehicular traffic flow at various times of the day follows a similar pattern at various points in the
city, with only the magnitude of traffic volume varying between locations.
 A clear distinction exists between the peak hour observed for motorised and non-motorised modes.
The morning and evening peaks for non-motorised vehicles occur from 8am to 9am, and 5pm to

36
 6pm respectively, while for passenger motorised modes it is observed to be from 9am to 10am, and
6pm to 7pm respectively.

The following figures show the relative proportion of various modes at different locations in the city.
Passenger car units (PCUs), as given in IRC SP: 41, which provides the ‘Guidelines on Design of At-Grade
Intersections in Rural & Urban Areas’, are used to convert different vehicles into a common unit. The
following table gives the PCU values adopted for the study.

Table 7 PCU values (Source: IRC SP 41)

Type of vehicle PCU values

1
Car/jeep/taxi
Two-wheeler 0.5
Auto-rickshaw 1
Cycle-rickshaw 1.5
Bicycle 0.5
Bus 3

37
 Figure 6 Vehicular mode shares at inner cordon points*

*NMV includes bicycles, cycle-rickshaws, hand carts, and animal-drawn vehicles

38
39
 Figure 7 Hourly variation of passenger vehicles (in PCUs) at inner cordon points

80000
70000
Passenger Vehicles, PCU

60000
50000
40000
30000
20000
10000
0
09 - 10
07 - 08

08 - 09

10 - 11

11 - 12

12 - 13

13 - 14

14 - 15

15 - 16

16 - 17

17 - 18

18 - 19

19 - 20

20 - 21

21 - 22

22 - 23
Time of the day

Ambedkar Junction Collector's Office Convent Junction Diamond Park

Dolphin Hotel Gopalapatnam Gurudwara Hanumanthawaka
Jagadamba Maddilapalem NAD Palm Beach
Purna Market Seven Hills Siripuram

Figure 8 Hourly variation of goods vehicles (in PCUs) at inner cordon points

8000
Goods Vehicles, PCU

6000

4000

2000

0
11 - 12
07 - 08

08 - 09

09 - 10

10 - 11

12 - 13

13 - 14

14 - 15

15 - 16

16 - 17

17 - 18

18 - 19

19 - 20

20 - 21

21 - 22

22 - 23

Time of the day

Ambedkar Junction Collector's Office Convent Junction

Diamond Park Dolphin Hotel Gopalapatnam
Gurudwara Hanumanthawaka Jagadamba
Maddilapalem NAD Palm Beach
Purna Market Seven Hills Siripuram

40
 Figure 9 Hourly variation of total non-motorised vehicles at inner cordon points

1000
900
800
700
600
NMT

500
400
300
200
100
0
08 - 09

22 - 23
07 - 08

09 - 10

10 - 11

11 - 12

12 - 13

13 - 14

14 - 15

15 - 16

16 - 17

17 - 18

18 - 19

19 - 20

20 - 21

21 - 22
Time of the day

Ambedkar Junction Collector's Office Convent Junction Diamond Park

Dolphin Hotel Gopalapatnam Gurudwara Hanumanthawaka
Jagadamba Maddilapalem NAD Palm Beach
Purna Market Seven Hills Siripuram

*Non-motorised vehicles include bicycles, cycle-rickshaws, hand carts, and animal-drawn vehicles

Figure 10 Hourly variation of pedestrians at inner cordon points

41
 30000
Number of pedestrians
25000
20000
15000
10000
5000
0

15:00-16:00
10:00-11:00

11:00-12:00

12:00-12:30

13:00-14:00

14:00-15:00

16:00-17:00

17:00-18:00

18:00-19:00

19:00-20:00

20:00-21:00

21:00-22:00

22:00-23:00
9:00-10:00
7:00-8:00

8:00-9:00

Time of the day

Ambedkar Circle Collector office Convent Junction

Diamond Park Dolphins Junction Gopalapatnam Junction
Gurudwar Junction Hanumanthwaka Jagadamba Junction
Maddilapalem NAD X Road Palm beach Junction
Purnamarket Sevenhills hospital Siripuram Junction

2.4.2.2 Outer cordon TMCs

The four locations selected for outer cordon TMCs represent the four entry points to the city. The
following figure shows the location of these points on the map of the GVMC. Sixteen-hour TMCs have
been conducted on a typical working day for each of these intersections.

Figure 11 Location of outer cordon points for TMCs

42
 (Source: www.maps.google.com)
The hourly variation of passenger vehicles, goods vehicles and non-motorised vehicles at each of the
outer cordon points are shown in the following figures. The following are the salient features observed
from the data:

 Auto-rickshaws are again observed as the predominant mode among passenger vehicles even
at the outer cordon points, followed by two-wheelers and cars.
 Significant numbers of pedestrians use the intersections even at the outer cordon locations of
the city.
 The peak hour for both motorised passenger and non-motorised vehicles is observed to be from
8am to 9am in the morning, while it differed in the evening with 5pm to 6pm being the peak for
non-motorised vehicles and 6pm to 7pm for passenger vehicles.
 The movement of goods vehicles remains relatively constant throughout the day.

Figure 12 Vehicular mode shares at inner cordon points*

43
 *NMV includes bicycles, cycle-rickshaws, hand carts, and animal-drawn vehicles

Figure 13 Hourly variation of passenger vehicles (in PCUs) at outer cordon points

10000
9000
Passeneger Vehicles, PCU

8000
7000
6000
5000
4000
3000
2000
1000
0
12 - 13

22 - 23
07 - 08

08 - 09

09 - 10

10 - 11

11 - 12

13 - 14

14 - 15

15 - 16

16 - 17

17 - 18

18 - 19

19 - 20

20 - 21

21 - 22

Time of the day

Bheemili Gajuwaka Lankelapalem Thagarapuvalasa

44
 Figure 14 Hourly variation of goods vehicles (in PCUs) at outer cordon points

3500

3000
Goods Vehicles, PCU

2500

2000

1500

1000

500

0
07 - 08

08 - 09

09 - 10

10 - 11

11 - 12

12 - 13

13 - 14

14 - 15

15 - 16

16 - 17

17 - 18

18 - 19

19 - 20

20 - 21

21 - 22

22 - 23
Time of the day

Bheemili Gajuwaka Lankelapalem Thagarapuvalasa

Figure 15 Hourly variation of total non-motorised vehicles at outer cordon points*

300

250

200
NMT

150

100

50

0
21 - 22
07 - 08

08 - 09

09 - 10

10 - 11

11 - 12

12 - 13

13 - 14

14 - 15

15 - 16

16 - 17

17 - 18

18 - 19

19 - 20

20 - 21

22 - 23

Time of the day

Bheemili Gajuwaka Lankelapalem Thagarapuvalasa

*Non-motorised vehicles include bicycles, cycle-rickshaws, hand carts, and animal-drawn vehicles

Figure 16 Hourly variation of pedestrians at outer cordon points

45
 4500
4000
3500
Number of pedestrians

3000
2500
2000
1500
1000
500
0

22:00-23:00
9:00-10:00

10:00-11:00

11:00-12:00

12:00-12:00

13:00-14:00

14:00-15:00

15:00-16:00

16:00-17:00

17:00-18:00

18:00-19:00

19:00-20:00

20:00-21:00

21:00-22:00
7:00-8:00

8:00-9:00

Time of the day

Tagarapuvalasa Bheemili Junction Lankelapalem Gajuwaka Junction

2.4.3 Origin-Destination (OD) surveys

OD surveys have been conducted at eight locations to capture the travel characteristics of trips
originating outside the city. The formats suggested in the CMP toolkits by the Ministry of Urban
Development (MoUD) have been adopted for the survey. These trips could either be destined to the city
or just be passing through the city. Four of these locations are the outer cordon locations where TMCs
were carried out, two were entry points to the core-city area, i.e. NAD Junction and Hanumantawaka.
The other two were the Visakhapatnam Railway Station and the regional bus depot, from which external
trips using public transport enter the city. Their sample size, out of the total trips entering the city, shall
be calculated and the total of such trips per day shall be stated at the modelling stage of the project. The
following table gives the sample size of trips captured at each location through the OD surveys.

46
 Table 8 Sample size of trips captured in OD surveys
Name of intersection Sample size
Passenger Goods
vehicles vehicles
New Gajuwaka 1689 690
Tenneti Park 905
Tagarapuvalasa 1468 784
Lankelapalem 1112 826
NAD Junction 3542 1681
Railway Station 3705
RTC Complex 3317
Hanumantawaka 3259 949

2.4.4 Road inventory survey

The main objective of the road inventory survey is to evaluate the existing street infrastructure available
for various modes. While the municipal agencies are expected to maintain the database on their existing
infrastructure, no such data exists with the GVMC, and hence a sample of the total road network is
selected for the survey. The major arterials, both in the core-city area and other built up areas like
Gajuwaka and Pendurti have been collected. The exact roads to be surveyed in the city have been
identified in consultation with the GVMC, and existing road inventory for various modes has been
collected. The following figure shows the total roads covered among the major roads in the city.

47
 Figure 17 Links selected for road inventory data collection

It was observed that only 77km of the 1,100km of total urban roads have footpaths constructed or being
constructed along them. But their quality is questionable as they are not observed to be continuous,
with encroachments like parked vehicles, crossings and entries to properties along the road blocking the
free movement of pedestrians. Also, at many locations concrete blocks covering the sewerage lines are
being used as footpaths. They have high kerb heights, gaps between blocks and electric poles, street
lights and trees planted in the middle making them inaccessible for many pedestrians.

48
 Figure 18 Inadequate footpath infrastructure along the BRT corridor

Adequate bus stop infrastructure is also observed to be lacking at most places in the city, and bus users
are made to stand on open roads even during extreme weather conditions like the summer heat and
rain.

Figure 19 Inadequate bus stop facilities

49
The available data on carriageway widths for vehicle users have been collected for all major roads, and
shall be used during the modelling phase of the study. Organised off-street parking is very minimal in
the city, but on-street parking is observed predominantly on many arterial roads.

Figure 20 On-street parking observed on most arterials

2.4.5 IPT operator survey

Auto-rickshaws form the intermediate public transport (IPT) system of the city, acting as an end-to-end
taxi service in some areas, and also as a public transport system on some corridors where they operate
as shared services shuttling between fixed origins and destinations. The total fleet of auto-rickshaws is
around 25,000 out of a total of 500,000 vehicles registered in the city. They cater to a significant
proportion of trips in the city and operate throughout the day. But the infrastructure provided for them
is negligible. Therefore a questionnaire survey has been conducted with the drivers of auto-rickshaws,
both shared and normal autos, to understand their operational details and the economics involved. The
survey format is provided in the Annexure.

A sample size of 450 out of the 25,000 auto-rickshaws is observed to be significant at a 95 per cent
confidence interval (CI) using non-probability sampling. Hence, 450 drivers have been interviewed at
various locations all over the city, out of which data for 418 autos is observed to be accurate and hence
selected for further analysis. The data from this survey helps in carrying out the comparative analysis

50
between the formal city bus system and the informal shared auto-rickshaw system to integrate them
into the proposed low-carbon and sustainable transport system in the city.

The following table gives the major trends emerging from the auto-rickshaw survey:

Table 9 Major outcomes from IPT operator survey

IPT operator details Salient features of data
Type of 3-wheelers Currently equal number of 4 & 7-seaters, but increasing trend
of 7-seaters
Vehicle ownership 80 per cent are driver-owned rickshaws, while 20 per cent are
rented
Driver details 75 per cent from localities in Vizag, while 25 per cent are
migrants
Hours of operation 10 hours/day
Distances operated 100km/day
Income of auto-rickshaw drivers 80 per cent of drivers operate with incomes less than Rs.
10,000/month
Major items of expenditure Loan instalments, fuel and maintenance costs
Major issues in operation Lack of finance, benefits like pension and insurance, parking
facilities, basic amenities like drinking water and toilets

The following major corridors have been identified as having intense shared auto-rickshaw movements,
and can point towards future improvement requirements of the city bus system:

 Maddilapalem to NAD
 Maddilapalem to Madhavadara
 RTC Complex to Waltair
 RTC Complex to Gajuwaka
 RTC Complex to Sindia
 RTC Complex to Pendurti
 Jagadamba to Waltair
 Jagadamba to Arilova
 Gajuwaka to Aganampudi
 Gajuwaka to NAD
 Pendurti to Maddilapalem

2.4.6 Petrol pump surveys

Data regarding the vehicle’s make, age, fuel type, odometer reading, fuel efficiency (mileage per litre of
fuel), annual mileage of the vehicle, and pollution under control (PUC) certification has been collected
through the petrol pump surveys. A questionnaire survey has been carried out for 1000 cars spread
across two petrol pumps (Isakathota and Siripuram), and 1000 two-wheelers, 200 auto-rickshaws and

51
100 LCVs and buses at one petrol pump (Isakathota). It is preferred to collect all the data at a single
petrol pump for a longer duration to capture all vehicle age categories. But as the sample size was
smaller, and there were fewer cars observed in Vizag, the data was collected at two petrol pumps to
attain a better spread of age of vehicles in the city. Enumerators were positioned at the fuel-filling
stations, and the vehicle drivers were interviewed in the order of their arrival. The survey was conducted
during March 2013. The key findings from these surveys are explained in this section. The following
table and figure give the summary of data collected through petrol pump surveys.

Table 10 Summary of petrol pump survey data

Type of vehicle Type of fuel Average age Fuel efficiency Annual

(in years) (km/litre) mileage
(km/year)

Car Diesel 3.6 16 22,662

Petrol 5.1 14 10,891

Petrol+LPG 4.2 14 16,832

2-wheeler Petrol 4.1 49 11,018

3-wheeler Diesel 3.0 29 31,689

Bus Diesel 3.5 5 80,890

Mini-bus Diesel 3.8 7 13,081

Goods auto Diesel 1.0 23 19,901

LCV Diesel 2.5 17 50,315

Truck Diesel 8.4 7 26,179

52
 Figure 21 Mode-wise fuel type distribution

100%
100%

100%

100%

100%

100%

100%
58%
31%

10%

0%

0%
0%

0%

0%
0%

0%

0%

0%

0%

0%

0%

0%

0%
TRUCK CAR 2- 3- BUS MINI BUS GOODS LCV
WHEELER WHEELER AUTO

PETROL DIESEL Gas/Petrol

These inputs are used at various stages of travel demand and emission modelling as explained in the
following chapters.

2.5 Base Year Transport Indicators

2.5.1 Trips in the city

The per capita trip rate in the city is observed to be 1.66 trips per day, with males having a higher trip
rate of 1.81 trips per day compared to females who make 1.52 trips per day. This leads to a total of 2.88
million trips being made in the city everyday by the 1.73 million people using various modes. These trips
are disaggregated by trip purpose and gender in the following table. It is observed that education trips
are made by both men and women in equal proportion. However, in work trips the proportion of female
trips is very low, reflecting the low workforce participation rates among women.

Table 11 Trip purposes in the city

Trip purpose Male Female Proportion of trips Number of trips

Work
33% 6% 39% 1,123,200

53
Education
16% 16% 32% 921,600
Social/recreational
4% 23% 27% 777,600
Other
2% 0% 2% 57,600
All purposes
54% 46% 100% 2,880,000

(Source: Household Interview Data)

2.5.2 Mode share

The aggregate mode share for various trips is shown in the following table. It is observed that up to 55
per cent of the total trips are made by non-motorised modes like walking and cycling. Cars cater only to
2 per cent of the total trips made in the city. Among vehicular trips, buses are the dominant mode of
transport followed by 2-wheelers and auto-rickshaws. There is a clear difference in the mode shares
between males and females. Up to 70 per cent of the trips made by women are by walking, followed by
bus and auto-rickshaw, which cater to about 23 per cent of their trips. The rest of the modes form a
minor share of the trips. However, in the case of men, two-wheelers also form a significant proportion
of the mode-share, covering 22 per cent of the trips.

Table 12 Mode shares in Visakhapatnam

Mode Including walk trips Vehicular trips

Total Male Female Total Male Female

Walk 52% 37% 69% NA NA NA
Car 2% 2% 1% 3% 4% 3%
2-wheeler 15% 22% 6% 30% 35% 19%
Bus 18% 22% 14% 38% 35% 46%
Auto-rickshaw 9% 10% 9% 20% 16% 28%
Cycle 3% 5% 1% 7% 9% 3%
Others 1% 1% 0% 1% 1% 1%
Total 100% 100% 100% 100% 100% 100%

(Source: Household Interview Data)

These mode shares vary between different trip purposes, income groups and the trip lengths of the
users, as shown in the tables below. Walking remains the dominant mode for all trip purposes. Among

54
vehicular trips, 2-wheelers are the most preferred mode for work trips, while education trips are mostly
made by bus and auto-rickshaw. It is also to be noted that cycle usage is higher in work trips than in
education trips. This shows that the cycle users are captive to that mode and can’t afford the cost
involved even in using bus or auto-rickshaw.

Table 13 Trip purpose-wise mode shares

100% 1%
6% 3%
90% 2%
80%
32% Others
70% 46%
43%
73% Cycle
60%
9%
Walk
50%
40% 21% 15% Auto-Rickshaw
15%
30% Bus
6% 13%
20% 29% 2-Wheeler
28% 8%
10% 15% Car
11%
0% 3% 5% 1% 3%
Work Education Shopping/ Other
Recreational

(Source: Household Interview Data)

Further disaggregate analysis of trip purpose-wise mode shares have been carried out gender-wise, and
the results are presented in the table below. It is observed that even among work trips that have high
car and 2-wheeler usage, females in the city make the majority of their trips by walking and by bus.
Therefore 2-wheelers are the exclusive domain of males making work trips. Improving public transport
and non-motorised transport facilities will benefit the majority of the population.

Table 14 Trip purpose-wise mode shares disaggregated by gender

Trip purpose Gender Type of mode

Car Two-wheeler Bus Auto Cycle5 Walk Total

Work Male 3% 33% 14% 8% 7% 35% 100%

Female 2% 18% 33% 11% 2% 34% 100%

Sub total 3% 29% 19% 9% 6% 35% 100%

5
No cycle-rickshaw trips were observed either in the household interviews or in the traffic volume counts.

55
Education Male 0% 6% 26% 14% 4% 50% 100%

Female 0% 5% 28% 16% 2% 49% 100%

Sub total 0% 6% 27% 15% 3% 49% 100%

Social Male 5% 10% 3% 3% 1% 78% 100%

Female 1% 3% 2% 2% 0% 93% 100%

Sub total 1% 4% 2% 2% 0% 90% 100%

Recreational Male 1% 15% 19% 10% 2% 55% 100%

Female 2% 14% 18% 7% 0% 60% 100%

Sub total 1% 14% 18% 8% 1% 57% 100%

Total Trips 2% 15% 17% 9% 3% 54% 100%

The following figure shows the mode shares of trips made by various income groups identified from the
household interviews. It is observed that the walk mode share reduces with increases in household
income, while the 2-wheeler mode share increases. The mode share of the public transport trips, i.e. the
city bus and intermediate public transport (IPT) like auto-rickshaws, remain relatively the same across all
income groups. Even among the high income groups, the car mode share is only 3 per cent of the total
trips.

Figure 22 Income group-wise mode share

56
 100%
90%
80%
46% Others
70% 52%
66% Cycle
60%
Walk
50%
10% Auto
40% 9% Bus
30% 19%
2W
9% 19%
20% Car
15% 19%
10% 14%
0% 6% 1% 3%
Low-Income Middle Income High Income

(Source: Household Interview Data)

2.5.3 Trip lengths

The trip length distribution of all trips in the city are shown in the table below. Seventy-one per cent of
trips are less than 3km in length, which explains the high proportion of walk trips in the city. The average
trip length in the city is observed to be 4.1km.

Table 15 Trip length distribution in Visakhapatnam

Trip length % of trips

<1 km 54%

1-3 km 17%

3-5 km 9%

5-10 km 10%

>10 km 9%

The trip length patterns show a clear difference between usages of different modes. The following table
gives the average trip length of various modes. It can be observed that each mode is catering to a
particular trip length category, and the choice of mode is determined significantly by the trip length.

57
Non-motorised modes like walking and cycling are used mostly for short trips. The average trip length in
the city is only 4.1km in spite of its large area, showing that people prefer to live near their destinations.

Table 16 Mode-wise average trip length

Mode Average trip length (in km)

Car 9.3

2-wheeler 5.8

Bus 11.7

Auto-rickshaw 5.9

Walk 0.7

Cycle 3.2

City average 4.1

(Source: Household Interview Data)

These average trip lengths are further disaggregated to observe the trip length distribution within each
mode. It is observed that 87 per cent of walking trips, the most used mode, are within 1km, which shows
the existing mixed land use pattern of development in the city. Also, contrary to popular perception, the
city bus and auto-rickshaw are not in competition with each other. They are catering to public transport
trips of varying trip lengths. The high frequency, low occupancy auto-rickshaws serve the smaller trip
lengths, while the high occupancy, low frequency but more comfortable buses are being used for longer
trip lengths.

58
 Figure 23 Mode-wise trip length distribution (in km)

100

90

80

70

60
% trips

50

40

30

20

10

0
0 2 4 6 8 10 12 14 16 18 20
Car Two-wheeler Bus Auto Walk Bicycle Trip Length (Km)

2.5.4 Travel times

The average time spent travelling by various income groups for various trip purposes is presented in the
following table. It was observed that even though the mode choice is different, the travel times are
similar across all income groups within a trip purpose. This shows that people across income groups
have fixed travel time budgets. The choice of mode therefore depends on the trip purpose, trip length
and availability of access to various modes.

Table 17 Income-wise and purpose-wise travel time (in minutes)

Income range Work Education Others Grand total

Low income 27 23 14 21

Middle income 28 25 14 23

High income 30 27 16 25

Average travel time 29 26 15 24

59
2.5.5 Infrastructure quality: average speed of different modes

The following table shows the average speeds observed for various modes from the road inventory
survey along the arterial roads. All modes except buses have average speeds in excess of 25kph,
indicating a good quality infrastructure for motorised modes of transport. The average speed of buses is
lower due to the delay caused at the bus stops.

Table 18 Average speeds of various modes along arterial roads

Mode Average speed

Car 30 kph
2-wheeler 28 kph
3-wheeler 25 kph
Bus 18 kph

2.5.6 Households within 10min walking distance of public transport

The access and egress trip time for public transport and IPT stops has been collected as a part of the
travel diary during the household interview. Based on the trip purpose, the access and egress trips of
public transport and IPT trips have been summarised in the following table. It is observed that about 70-
75 per cent of the PT & IPT trips originate and end within 15min from the service.

Table 19 Access-egress trip times for PT & IPT

Travel time (in min) % Access trips % Egress trips

<=5 28 29
5-10 45 46
10-15 15 15
15-30 10 9
30-60 2 1
>60 0 0
Total 100 100

2.5.7 Accessibility for disabled people by different modes

It is observed that of the few footpaths and bus stops that are available in the city, none have been built
for disabled-friendly access. The only means of access available for the disabled is to use their private

60
modes of transport, i.e. cars as even the 2-wheelers are not disabled-friendly. Considering the existing
mode share of 2 per cent for cars, it can be concluded that most of the city is inaccessible for the
disabled.

2.5.8 Traffic safety

More than 400 people are being killed annually in road traffic crashes (source: National Crime Research
Bureau), which is around 24 traffic fatalities per lakh of population. This is among the highest in the
country and needs urgent attention from the agencies involved, i.e. the GVMC to provide better
infrastructure, the Traffic Police for better enforcement of traffic speeds and helmet laws, and also the
National Highway Authority of India (NHAI) since a significant proportion of the fatalities are being
caused by trucks along the 70km of NH-5 that passes through the city and is currently functioning as a
major arterial road. This road however has speed limits and infrastructure specifications similar to a
highway, thereby imposing a risk on its users. The following figure shows the victim and impacting
vehicles based on FIR data collected from the Traffic Police.

Figure 24 Traffic fatality – victim and impacting vehicle types

180
160
140 Unknown
120 Others
No. of fatalities

100
Cycle
80
Truck
60
Bus
40
20 Auto Rick

0 Two Wheeler
Car

Victim Road User Type

(Source: FIR Data from Visakhapatnam Traffic Police, IIT Delhi)

It is clear that pedestrians are the most vulnerable road users and are exposed to the maximum risk,
followed by two-wheelers. Providing safe walkways and crossing facilities, street lighting for pedestrians
and better enforcement of helmet usage in the city are the immediate measures that need to be

61
implemented in the city. The following table shows the risk of users of each mode, and the risk caused
by each mode on the other modes as a function of 100,000 users of that mode. It is observed that 2-
wheelers have the highest risk of a fatality followed by cycles. This shows the pressing need for a better
implementation of the helmet law. Pedestrians, even though they have less risk per user, form 43 per
cent of the total fatalities. Such high pedestrian fatalities and risk to the cyclists shows the high speed
differential that exists in the city between motorised and non-motorised modes, thereby causing so
many deaths. The need for better helmet law enforcement and better traffic-calming in the city is clear
from the traffic fatality data.

Table 20 Risk exposed to, and risk imposed by, each mode
Mode Fatal accidents per 100,000 Number of accidents caused
users of the mode by the mode on other road
users per 100,000 of respective
mode users
Car 0.28
0.05
2-wheeler 0.03
0.07
Bus 0.02
0.00
Auto 0.03
0.03
Walk 0.00
0.02
Cycle 0.00
0.05
Truck 0.45
0.02

2.5.9 Safety: roads with speed limit greater than 50kph

The GVMC area has a total of 70km of National Highway (NH)-5 that passes through the city. The speed
limit on this road varies from 80kph to 100kph at different locations within the city, causing heavy risk to
the pedestrians, city bus and auto-rickshaw users who need to cross these roads. These roads also lack
basic amenities like footpaths and bus shelters, forcing the users to share road space with high-speed
motorised traffic, thereby inducing significant risk.

The Traffic Police have taken a few initiatives like putting temporary physical barricades at various
locations for traffic-calming, as shown in the figure below. But permanent measures like raised crossings

62
and speed humps along the road are absent. This is because the road is under the jurisdiction of the
National Highway Authority of India (NHAI), which doesn’t permit the creation of such infrastructure.

Figure 25 Inadequate street infrastructure along the NH-5

2.5.10 Safety: quality of footpath infrastructure

The road inventory survey revealed that the entire city has approximately 77km of footpaths that either
exist or are being constructed. Out of this, 40km are part of the two BRT corridors in the city, which are
under construction. None of these footpaths have been provided with disabled-friendly access. The
following table gives the summary of footpath infrastructure in the city.

Table 21 Length and width of available footpaths in Visakhapatnam

Width of Footpath (in Length (km)
m)

1.5 67
2.0 6

63
 3.0 4
Total 77

Most of the footpaths built in the city are as per the earlier Government Order (GO) issued by the state
government to maintain a width of 1.5m. The more recent footpaths built along the inner ring of the city
have a width of 2.0m, and they add up to a total of 6km, which is insignificant when calculated as a
proportion of the total road network in the city. The 3m-wide footpaths are built along the beach road
and are not a phenomenon replicated anywhere else in the city.

Table 22 Quality of footpath infrastructure

Quality of footpath Proportion of footpaths

No encroachments 30%
Discontinuous 50%
Mostly unusable 20%
Total 100%

It is observed that 50 per cent of the footpaths in the city are discontinuous, due to various reasons like
the presence of electric poles and trees on the footpath, entry to properties along the road, and breaks
in footpaths at crossings and on-street parking. 20 per cent of the footpaths are mostly encroached by
parking and shopkeepers along the road extending their shops to the footpaths. Only 30 per cent of the
footpaths are without any encroachments. However, even these footpaths are not completely
pedestrian-friendly because they are mostly concrete blocks placed as a cover to the sewer lines below
also acting as footpaths. They are 180-200mm above the road surface, and also have occasional gaps
between concrete blocks making them uncomfortable to the users.

2.5.11 Security: street lighting

Most of the arterial road network studied for the road inventory survey had adequate street lighting for
the carriageway, but no separate lighting exists for the footpaths. However, some arterial roads have
shops adjacent to the road, thereby providing adequate lighting for people to walk. 20km of the total
37km of footpaths have such active edges, which are also perceived to be more secure by the
pedestrians. The rest of the 20km of footpaths can be considered as insecure for the pedestrians

64
2.5.12 Security: pedestrian safety perception

The percentage of people feeling safe to walk in the city has been derived from the household interview
data and is presented in the following table. It is observed that the majority of the population perceives
the existing roads to be secure for walking.

Figure 26 Gender-wise safety perception

User groups Opinion
Good OK Bad
Females 2% 83% 15%
Males 5% 76% 19%
Overall 4% 79% 17%

2.5.13 Affordability: cost of commuting and affordability for various modes

The monthly amount of money spent on commuting for various modes has been captured in the HHI
survey, and the summary is presented in the following table. It is observed that the average amount of
money spent is higher for a bus user even compared to a 2-wheeler user. This is due to the lack of a
proper parking policy in the city, which results in free parking being provided for 2-wheelers in most
locations of the city.

Table 23 Mode-wise cost of commute

Mode Average Cost of commute/month (in Rs.)

Car 680
2-wheeler 220
Bus 410
3-wheeler 400
Average for the city 360

Also, the proportion of household income spent on transport has been calculated for various modes and
income groups, and is presented in the following table. It is to be noted that these numbers are
calculated as the proportion of individuals’ travel cost out of the total household income. The data
suggests that each trip maker spends around 5 per cent of the household income on travel. In the
households with income less than Rs. 5000 per month, each trip maker spends around 12 per cent of
the household income, which prohibits them from spending money on trips like educational trips for
children.

65
 Table 24 Proportion of household income spent on travel for various modes
Mode Income range (in Rs.)

<5000 5000-10000 10000-20000 20000-50000 >50000 Average

2-wheeler 6% 3% 1% 1% 0% 2%
Auto 14% 6% 2% 1% 1% 6%
Bus 13% 6% 3% 1% 0% 6%
Car 16% 3% 3% 3% 1% 3%
Average 12% 5% 2% 1% 1% 5%

2.5.14 Consumption of land for transport activity

It is observed that out of the 530km2 of area under the GVMC, only 166km2 is built-up area, with the rest
being covered by hills, forests, agricultural and barren land. Within the built-up area of the city, 24km2,
i.e. 15 per cent of the area, has been allocated to various transport activities, out of which exiting
carriageways of roads occupy around 13km2, i.e. 8 per cent of the city area. This includes both intra-city
and inter-city transport infrastructure like railways and the airport in the city, as shown in the following
table.

Table 25 Distribution of land allocated for transport activity

Type of land use Area (in km2) Proportion of transport activity

Roads 12.8 53%

Railways 6.9 28%
Airport 4.6 19%
Total 24.3 100%

2.5.15 Economic indicators: investment trends for various modes

The annual budget and accounts data of the GVMC for the past five years has been studied for
investment trends for various modes. Since 2008, all municipalities in Andhra Pradesh including the
GVMC have been following the Centre for Good Governance (CGG) manual for accounting. This
segregates the money spent on transport infrastructure further into item-wise subheads like road
maintenance, footpaths, etc. Both the planned and actual expenditure has been studied, and it was
observed that the actual expenditure on various items is significantly different from the budget

66
allocated for that subhead. Hence the actual expenditure made on various transportation
infrastructures has been studied and is summarised in the following table.

Table 26 Item-wise transport infrastructure investments

Infrastructure Year-wise expenditure (in Rs. lakhs)
2008-2009 2009-2010 2010-2011
Road building 4910 2221 2313
Road maintenance works 477 602 1110
Street furniture 922 725 904
Footpaths 41 28 42
Bus stops 1 1 1
Total expenditure
6350 3577 4370

It is observed that while the total investment into transport infrastructure remains high, i.e. 64 crores,6
36 crores and 44 crores in successive years, the amount of money spent on various areas is a matter of
concern. The largest amount of money is spent on road-building activities, but the amount spent on
footpaths is only a fraction of the total budget, and no money is spent on bus stop infrastructure. Even
the road inventory surveys reveal that the existing infrastructure assists only private modes of transport
like cars and 2-wheelers, which cater to only 17 per cent of the total trips in the city. Hence there is an
urgent need to realign the spending patterns in the city towards the actual mode-wise users.

2.5.16 Tax burden mode-wise

The following table gives the tax imposed by the Road Transport Authority (RTA) for various modes of
transport in the city. It is observed that private modes like cars and 2-wheelers pay a lifetime tax on
their vehicles, while PT and IPT modes like 3-wheelers and buses pay a recurring tax.

The city bus system is considered as a commercial entity, and tax is levied on their gross-revenue. This
policy has to be reviewed because under the current system of taxes, public transport users pay more
than the private mode users.

Table 27 Mode-wise tax burden

6
Crore – 10,000,000. A unit in the Indian numbering system.

67
 Type of vehicle Mode of tax 1st vehicle 2nd vehicle

Car Life time 11% 14%

2-wheeler Life time 12% 14%
3-wheeler Annual Rs. 420 per annum Not allowed
Bus % of gross revenue 5% of revenue

The number of auto-rickshaws in the VMC area was restricted to 5000 before the formation of the
GVMC. After 2005, the area has expanded vastly from 110km2 to 530km2, but this legislation has not
been changed, making a large number of registered auto-rickshaws illegal to operate in the city. This has
to be reviewed and the ban needs to be lifted. It is clear from the mode share and safety data that auto-
rickshaws cater to a large section of society and are safer than most other modes. Also, if the number of
3-wheelers is restricted, the corresponding trips will shift to cars or 2-wheelers, adding to the road
congestion. Currently, auto-rickshaws are part of the sustainable transport solution for the city, and
need to be encouraged further with more suitable policies and regulations.

2.5.17 Fuel prices at pumps

The following are the existing prices of petrol and diesel in the city.

Table 28 Fuel prices in the city

Fuel type Price per litre (in Rs.)

Petrol (unleaded) 72.4

Diesel 50.4

2.5.18 Other charges levied in the city

Apart from the tax burden mentioned above, the other charges levied in the city are minor. Even the on-
street parking provided for cars and 2-wheelers is free at most locations of the city. Only one toll road
exists at the outskirts of the city, but it is mostly used by the trucks accessing the port and hence the
charge on intra-city trips is miniscule.

2.5.19 Subsidies granted

The state government provides subsidies to students using the city bus system through discounted bus
passes. While a normal city bus pass costs Rs. 550/month, the student bus passes cost only Rs. 240 for 3

68
months. The loss in revenue for the APSRTC due to these discounted passes is covered by the state
government and it amounts to Rs. 2 crores per month.

2.5.20 Population owning passes

On average, a total of 77,000 bus passes are in operation at any point of time in the city, i.e. about 4 per
cent of the population own bus passes in the city. The following are the various types of passes issued in
the city:

 General bus passes at Rs. 550/month

 Student bus passes costing Rs. 240 for 3 months
 Free passes to all students aged less than 12 years
 Free passes to all girl students up to Xth standard
 Route-wise passes for students

Free passes for employees of Non-Government Organisations (NGO) in the city

2.5.21 Transport investments

The budget data of the GVMC has been analysed to identify the trends in investments incurred on
transport infrastructure in the city. Around Rs. 40-50 crores is spent on transport annually. The following
figure shows the proportion of funds that are spent in various sub-areas within transport. It is observed
that up to 80 per cent of the funds go into road building or maintenance works, and around 20 per cent
into street furniture like street lighting, signal systems and signs. Pedestrians and bus users, who
together contribute to 70 per cent of the trips made in the city, are only allocated 1 per cent of the total
transport expenditure. It is to be noted that the amount spent on the Bus Rapid Transit (BRT) corridor
being developed in the city, under JNNURM, is not considered for this analysis as it is a corridor/project-
based expenditure, and does not reflect the overall spending pattern of the GVMC.

Figure 27 Annual transport expenditure pattern

69
 100% 1% 1% 1%
90% 15% 20% 21%
80% 8%

Proportion of investment
70% 17%
25%
60% Bus Stops
50% Footpaths
40% 77% Street Furniture
30% 62% Road Maintenance works
53%
20%
Road Building
10%
0%
2008-2009 2009-2010 2010-2011
Year

(Source: Budget Data, GVMC)

70
3 Base Year Scenario
The data collected and the indicators developed give a thorough understanding of the existing
transportation scenario of Visakhapatnam. To estimate the likely scenario of development of the city
and its transportation system in the horizon year of planning, i.e. 2030, a travel demand model and
emissions model for the city was developed. These models are calibrated for the base year using the
indicator outcomes, and are extrapolated for the horizon years using the calibrated base year models.
These scenarios include a Business as Usual (BAU) scenario and various low-carbon development
scenarios. The BAU scenario assumes that the development trend in the city continues in the current
pattern without any low-carbon interventions being taken up. Low-carbon scenarios include specific
interventions like land use changes, public transport development, non-motorised transport
development and technology transition scenarios. The base year model development and calibration,
and the likely impact of a Business As Usual (BAU) scenario of development are explained in the current
chapter, and the low-carbon scenarios are explained in Chapter 4.

3.1 Base Year Travel Demand Model

The base year travel demand model is required to replicate the road network and travel patterns of the
city in modelling software, and to test for various short-term measures that can be taken to improve the
existing transportation systems. The following table gives the input parameters and their data sources
used for developing the base year model.

Table 29 Modelling components and input sources

Model component Input source

Traffic analysis zone map Derived from ward map

Road network Derived from property tax data
Primary data collected for road inventory and link speeds
Secondary data on road widths
Trip production patterns Household interview data
Trip attraction patterns Land use data from the master plan
Building-wise usage type from property tax database
Trip distribution Trip length distribution patterns from household interview data to
calibrate the gravity model
Base mode shares Household interview data

71
 Traffic assignment Traffic volume counts used for network calibration

The outcomes from the various stages of the modelling procedure are explained in the following
sections.

3.1.1 Traffic analysis zones

Traffic analysis zones (TAZs) were created within the city, and were assumed as the units of production
and attraction for modelling. The available literature suggests that trip production is dependent on the
socio-economic characteristics of households within the TAZ, while trip attraction depends on the land
use type of the TAZ. The trip production and attraction calculation methodology is explained in the
following sections.

The entire planning area delineated for the study, the GVMC area in this case, is divided into various
TAZs, which act as units of disaggregation for trip productions and attractions generated from various
parts of the city. The accuracy of the travel demand model depends heavily on how accurately the TAZs
and the road network replicate the actual scenario in the city. TAZs are identified in such a way that the
land use type and trip-making characteristics of all households in a particular TAZ are assumed to be
uniform throughout the TAZ. TransCAD 5.0, a GIS-based travel demand modelling software programme,
was used to create the TAZ map of Visakhapatnam.

Area-level disaggregation for various input data like the census data (2011) and property tax database
(2012) were available at the ward level for the 72 wards in the GVMC. It was observed that the wards
within the core area, i.e. wards within the erstwhile Visakhapatnam Municipal Corporation (VMC) have a
smaller area, while the outgrowths of the city, which have been added while forming the GVMC, have
much larger areas. The ward population densities showed a skewed pattern of development in the city,
with densities in the core city area as high as 60,000 people per square kilometre, while the outgrowths
have much lower densities, i.e. less than 1000 people per square kilometre. This is because of the sparse
development in the outgrowths. Since the TAZs need to be as homogeneous as possible for the
purposes of study, their areas were reduced. Most of the wards in the core area were already small, i.e.
less than 1km2, and hence have been retained as they are, whereas the larger wards with areas greater
than 1.5km2 and two or three pockets of development were split into smaller TAZs. However, a few
large TAZs with negligible development in the existing scenario were not split into smaller zones. In this
process, the 72 wards in the city were split into 97 TAZs. The following figure shows the created TAZ
map.

72
 Figure 28 TAZ map showing the 97 zones

The next task was to extract the data for each of the TAZs. The property tax data of all the properties
within the limits of the GVMC was created in GIS. This data is overlapped with the TAZ map to extract
the data of each TAZ, and the data was used for further modelling. Figure 28 shows the property data
overlapped with the TAZ map.

73
 Figure 29 Property data overlapped with the TAZ map

3.1.2 Creating the road network

The road network for the base year model was created using the GIS property tax data base prepared by
the GVMC, as a part of which various roads in the city have already been identified. The total road
lengths within the GVMC limits are observed to be 3,469km, including all minor access roads. Since the
road layer is used in conjunction with the TAZ layer, to assign the inter-zonal trips, the interior access
roads within each TAZ have been removed and only the roads catering to inter-zonal trips are retained.
This way, the road network is prepared for inter-zonal trips, and for intra-zonal trips connectors act as
proxies and are connected to the nearest road network to make sure there are no errors. Moreover, as
explained in the section above, TAZ sizes are kept small so as minimise errors. However, this is a
drawback in the travel demand modelling process, as it eliminates some of the very short duration trips,
which is to be kept in mind. The total road network on these roads, i.e. the highway, arterial and sub-
arterial roads, is observed to be 430km in length, and is shown in the following figure. TAZs are

74
represented in the road layer using centroids, i.e. trips between TAZs are assumed to be made between
centroids of the TAZs. Centroid connectors are created in the road network layer to connect the TAZ
centroids to nearby links. The number of centroid connectors for each zone is determined in such a way
that all the major directions of movement from the TAZ are covered. On average, four centroid
connectors are given per each TAZ so that the trips starting or ending in a zone are distributed in all
directions. The following figure shows the road network used for modelling, including the centroid
connectors.

Figure 30 Road network used for modelling (including connectors)

Various network attributes required for modelling, like the link speed, travel time, road widths, effective
capacity, and availability of public transport are all built into the network model based on various
primary and secondary data sources. This network is used further for the four-stage modelling.

3.1.3 Trip generation

This step involves estimating the number of trips produced and number of trips attracted to each TAZ.
The methodology and the results of this exercise have been presented below.

75
3.1.3.1 Trip production

Disaggregate multiple classification analysis (MCA), which gives the number of daily trips produced per
household, was prepared using household-level data of size, income and motorised vehicle ownership,
i.e. cars and 2-wheelers. The variable household size includes both males and females in the household.
In that model, gender-wise trip production variation is explored for the cross-classification analysis, but
the variable was not statistically significant in explaining trip production. Even from the available
literature, these are identified as the likely variables that best explain the trip production characteristics
in Indian cities. The mode choice can however be different for different genders and income groups. The
base year model is only for the existing scenario, developed from our surveys. This already includes the
various user groups. The mode choice model prepared for the horizon year showed trip length and
household income as the critical variable for mode choice, across genders.

Household interview (HHI) data for the city, which captures the socio-economic information and number
of trips made in each household, is taken as the input for the MCA tables. The average of the trips
produced per household for each combination of variables is calculated from the total household data
and placed in the various cells of the table7. The variable boundaries were stratified in such a way that
there are enough observations to accurately describe each cell of the MCA table. Separate tables were
prepared for the four categories of trip purposes considered. An MCA table shows the number of daily
trips produced per household whose characteristics are stratified according to a combination of
household attributes that are most appropriate to describe its trip productions. The variables are
categorised such that the households are divided into a limited set of combinations. The household level
variables identified from the literature and their categories made for Visakhapatnam are explained
below:

 Household size: This is the most influential variable for trip production. Based on the frequency
distribution, households (HH) are categorised into HH sizes <=2, 3, 4, 5, >=6, i.e. HHs with size 1 or 2
are grouped into one category, and all HHs with size >=6 are grouped into a single category.
 Household income is the second variable that is observed to affect the trip rates of a household,
since higher income households are likely to make more recreational trips. Even though direct
income data is collected in the HH interviews, estimating income based on the assets owned by a

7
Guevara, C. A., & Thomas, A. (2007). Multiple classification analysis in trip production models. Transport Policy,
14, 514-522

76
 household is considered to be a more robust way of estimating income. Based on their assets, the
households have been divided into three categories in increasing order of income:
o Inc. category 1: Households with no assets, or owning a mobile/stove/television
o Inc. category 2: Households owning a 2-wheeler/fridge/HHs owning AC but neither car/2W
o Inc. category 3: Households owning both a 2-wheeler and an AC, and HHS owning cars.
 Motorised vehicle ownership: Vehicle ownership gives more options of making a trip and is an
important variable to consider for trip production. In the case of Visakhapatnam, only 3 per cent of
trips are made by cycles, and motorised vehicle ownership is observed to be having a larger
influence on trip production. Therefore two categories of motorised vehicle ownership are
considered for analysis, i.e. HHs owning no motorised vehicles and HHs owning at least one
motorised vehicle, i.e. either a car or a 2-wheeler. However, vehicle ownership can be a dependent
on the income of the household.

Various combinations of these variables were considered for different trip purposes. The trip rate trends
observed in the three more significant MCA tables are shown in the following figures. Figure 31 shows
the trip rate trends considering all the three variables explained above. It is observed that the trip rates
do not follow a clear trend for the combination of these three variables. Since vehicle ownership is a
function of the income of the household, these two variables can be correlated. Hence a separate MCA
chart considering only two variables, i.e. the household size and income, is prepared and presented in
Figure 32.

77
Figure 31 Multiple classification analysis trip rate charts for a combination of three variables

Figure 32 Multiple classification analysis trip rate charts for combination of two variables

78
It is observed from the above two figures that a broad trend emerges for household size, but within
each household size category trip rates between income groups do not follow a clear trend. Hence a
separate MCA chart is prepared considering just household size as the variable and is presented in the
following figure.

Figure 33 Multiple classification analysis trip rate charts for one variable

It is observed from the above figure that a clear trend emerges where the number of trips, for all trip
purposes, increases as household size increases. This shows that in the case of Visakhapatnam, HH size
is the only parameter having significant influence on trip rates. This illustrates that increases in income
do not change trips rates but only the mode of travel may change. Therefore household size is used
further to estimate the trips produced in each TAZ based on the average HH size observed in the TAZ.
The following table shows the trip rates observed for various Household sizes.

Table 30 Trip rate table used for trip production

HH size Work trips/HH Education trips/HH Other trips/HH Total trips/HH

<=2 1.58 0.09 1.64 3.31

79
 3 1.89 0.57 1.93 4.39

4 2.33 1.63 1.76 5.71

5 2.59 2.46 1.81 6.85

>=6 3.09 2.68 2.18 7.95

Average 2.53 1.97 1.91 6.41

The total number of households in each traffic analysis zone (TAZ) was derived by overlapping the
property tax GIS layer with the TAZ layer. Household survey data was used to derive the average
household size in each TAZ, and the total trips produced in each TAZ were derived by multiplying trip
rate from the MCA table with the number of households. This results in zone-wise and purpose-wise
trips produced for the entire city.

3.1.3.2 Trip attraction

The purpose-wise number of trips attracted to each TAZ was estimated in this step. The attractiveness of
a zone is a function of the type of land use of that zone. For example, residential land uses produce trips
while commercial, institutional and industrial areas typically attract trips. Hence the existing land use
mix is considered as the critical variable in determining the trips attracted to each TAZ. Land use data at
the city level was provided by the master plan of the city, but they are only indicative as the land use
allocation in the master plan and the actual land use is observed to vary widely in practice.

The property tax data from the GVMC provides building-wise land use type and its building
footprint/plinth area in GIS. Types of land use in the buildings include: residential, commercial,
educational, industrial, public use, shops, hospital, cinema/pub entertainment, and others. Except
residential, all other land use types attract trips. Hence, the total plinth area of each type of attracting
land uses was derived using layer analysis in TransCAD 5.0, and was used as a measure of attractiveness
of the TAZ. The following figures show the spread of buildings of various land use types of the city on the
GIS map overlapped on to the TAZ map.

Figure 34 Residential buildings across the GVMC area

80
Figure 35 Commercial developments across the city

81
 Figure 36 Industrial developments across the city

Purpose-wise trips attracted to each zone from the household interview survey was used as the
dependent variable, and the total plinth area in the zone of each land use type were used as the
independent variables. However, the MCA method is used for disaggregated analysis, but the in case of
attraction no such data is available at the building level, and therefore TAZ-level attractions are derived
from the land use pattern in the city, for which regression is used. Multiple linear regression was used to
observe the relation between the trips attracted and the land uses of the TAZ. The regression equation
derived for each trip purpose as a function of land use are shown from Equation 1 to Equation 3, and
their respective coefficients are shown in Table 311 to Table 333

Work attraction:

[WorkAttr = 15.97 + 0.003 * Commercial + 0.005 * Industrial] (1)

(Coefficient of determination (R2) = 0.55, Standard error = 67.18)

82
 Table 31 Regression statistics for work attraction

Standardised
Unstandardised coefficients coefficients

Model B Std. error Beta t-value tcritical Sig.

(Constant) 15.973 10.474 1.525 1.96 0.134

Commercial
0.003 0.000 0.435 5.635 1.96 0.000
Industrial
0.005 0.001 0.564 7.300 1.96 0.000

Education attraction:

[EduAttr = -0.628 + 0.005 * Commercial + 0.001 * Educational] (2)

(Coefficient of determination (R2) = 0.98, Standard error = 65.34)

Table 32 Regression statistics for education attraction

Standardised
Unstandardised coefficients coefficients

Model B Std. error Beta t-value tcritical Sig.

(Constant) -0.628 8.216 -0.076 1.96 0.939

Commercial
0.005 0.000 0.982 41.809 1.96 0.000
Educational
0.001 0.000 0.055 0.976 1.96 0.357

Other trips attraction:

[OthAttr = 31.882 + 0.001 * Commercial] (3)

(Coefficient of determination (R2) = 0.35, Standard error = 32.94)

Table 33 Regression statistics for other trips attraction

Unstandardised Standardised
Coefficients Coefficients

Model B Std. error Beta t-value tcritical Sig.

83
 (Constant) 31.882 4.605 6.923 1.96 0.000
Commercial
0.001 0.000 0.350 2.989 1.96 0.004

It was observed that both commercial and industrial land use are significant variables in explaining work
trips, and the equation has a coefficient of determination. Even for the education trips commercial land
use has good significance, while educational land use has a lower significance. This is because the
municipal corporation registers some of the schools that operate for profit as commercial land use.
Commercial land use is the only significant variable to explain other trips, which include trip purposes
like shopping, recreation, health and social trips.

Based on these equations, the number of trips attracted to each zone for various trip purposes is
recalculated using the equations. This will however only give the number of trips at the scale of the
sample size of data, since the coefficients of the equation were derived based on sample trips from
household interview data. These attractions were used as an indicator of the purpose-wise
attractiveness of each zone. These trip purpose-wise attractions were scaled up proportionally for each
zone in such a way that the total attractions match the total trips produced for that purpose. This was
done using the PA-Balance technique available in TransCAD. The same procedure is applied for all
purposes to derive the total trips attracted to each zone for each purpose. Combining the results of the
above two steps, the trip production and attraction numbers for all zones in the city was derived, which
completes the trip generation component of the travel demand model.

Hence the Production-Attraction (PA) table is prepared for the total trips made in the city for each trip
purpose. The peak hour PA table is derived based on the hourly variation of the trips of each purpose.
8am to 9am is identified as the peak hour for the city. This is confirmed from the traffic volume count
surveys, and hence this is taken as the peak hour for the demand model. The PA table for each purpose
is derived for the peak hour according to the proportion of trips observed.

3.1.4 Mode share and trip distribution

For the base year model, the mode share split is carried out before the trip distribution:

 The purpose-wise peak hour trips are added up to get the total trips produced and attracted to each
TAZ
 The TAZ-wise mode share values have been derived from the HH interview data, and applied to the
PA table to get the mode-wise PA table for all zones

84
 One of the features of the four-stage demand modelling process is that only the inter-zonal trips are
considered for assignment. Hence, the proportion of intra-zonal trips in each TAZ are calculated
from the HH interview data, and these trips are excluded from the demand modelling process
 Hence the mode-wise PA table for inter-zonal trips during peak hour is derived from the trip
generation output. This is used as the input for trip distribution.

3.1.5 Trip distribution

Trip distribution is used to derive the Origin-Destination (OD) matrix from the PA table. The gravity
method is adopted for trip distribution in the current study, and the steps followed are outlined below:

Gravity application

 In this step, PA tables of each mode are considered separately, and the gravity method is
applied for each of them
 The speed and travel time for each link are defined separately for each mode. These speeds are
based on the mode-wise speeds observed on various links, from the road inventory and speed-
delay survey
 Mode-wise impedance matrices using travel time skims are generated for these speeds
 Using the PA table and the mode-wise impedances, separate OD matrices are derived for each
mode.

Gravity calibration

 The above gravity application considered an inverse power function as the deterrence function
using its default parameters
 The trip time distribution for this output varies from the actual trip time distribution taken from
the HH interview data, and needs to be calibrated
 Hence the parameters of the inverse power function are calibrated in such a way that the
gravity application output matches the actual trip time distribution.

The following table shows the calibrated values of the inverse power function for various modes.

85
 Table 34 Calibrated deterrence function for trip distribution

Mode Calibrated α for inverse power

Car -0.42

2-wheeler -0.46

Bus -0.35

Auto-rickshaw -0.47

Cycle -0.52

Using these updated values for the inverse power function and the mode-wise PA tables and travel time
skims, the calibrated OD matrix for each mode is derived.

3.1.6 Traffic assignment and calibration

The trip distribution output gives the OD matrix of person trips originating within the city. These person
trips are converted to vehicle trips based on the average occupancy observed in each mode from the
occupancy survey carried out in the city.

Table 35 Mode-wise average occupancy

Mode Average occupancy

Car 2.5

2-wheeler 1.5

Bus 30

Auto 4.9

Cycle 1

*Source: Occupancy Surveys

86
However, the floating population coming into the city through the numerous entry points have also
been captured from the OD surveys at these locations. These sample surveys are scaled up based on the
traffic volume counts at those locations. The OD matrices from these surveys are added to the OD from
the trip distribution to develop the overall OD matrix of the city.

The mode-wise calibrated OD matrices derived from the above step are assigned on to the road network
using the user equilibrium method in TransCAD, which assumes that each user selects the route that
gives the shortest travel time to him or her. The travel time used is the congested travel time calculated
as a function of free flow travel time on the link, its capacity and traffic demand. The Bureau of Public
Roads (BPR) function is used to define the volume-delay function:

Default values of the software for α and β, i.e. 0.15 and 4, are used for the model. Since the travel time
skim matrix is already taking mode-wise speeds based on survey data, the BPR function has not been
changed from default parameters. The link flows observed from traffic assignments are compared with
the actual traffic flows observed from traffic volume counts conducted at 20 intersections across the
city. This translates to 76 mid-blocks, considering each intersection as having 3/4/5 arms. It was
observed that the link flows from the traffic assignments varied from the actual traffic volume counts.
Some missing links in the road network were identified through this procedure. However, the larger
contributing factor to this difference is the OD matrix derived from trip distribution. The OD matrix had
to be recalibrated for it to match the traffic volume counts. For this, an iterative process available in
TransCAD called the OD matrix estimation was used. The OD matrix estimation procedure considers the
observed traffic volume counts and updates the OD matrix based on them iteratively until the resultant
OD matrix matches the traffic volume counts at all locations, within a permissible degree of error. Using
this method, the final calibrated OD matrix and the calibrated traffic flows at all locations of the city are
derived. The calibrated network flows after OD matrix estimation are shown in the following figure.

87
 Figure 37 Calibrated network flows for the base year travel demand model

With this step the base year travel demand model is calibrated and can be used for further analysis.

3.1.7 Key statistics from the base year model

In summary, the following are the key statistics for the calibrated base year model:

Population of the city, 2011: 1,730,328

Total city area: 534km2.

Built-up area: 166km2.

Average household size: 4.0

Average trip rate: 6.41 trips/HH

Peak hour factor (person trips): 22%

Total road length: 3,470km

Length of arterial + sub-arterials: 430km

88
The mode-wise statistics are shown in the following table. The trip length data presented here is derived
from the model calibrated based on data from the household interviews, and hence is slightly different
to the data from the surveys. But it is within an error range of 10 per cent.

Table 36 Mode-wise summary of base year model

Mode-wise details Mode share (%) Trips/mode Average trip Vehicle km – Passenger km
length (in km) Peak hour – Peak hour

Car 2 55,716 9.6 534,876 1,337,190

2-wheeler 15 417,872 5.7 2,381,872 3,572,808

Bus 18 501,447 10.5 5,265,190 157,955,700

3-wheeler 9 250,723 5.0 1,253,617 6,142,723

Walk 52 1,476,482 0.7 1,033,537

Cycle 3 83,574 3.1 259,081 259,081
Total for the city 100 2,785,815 4.0 11,143,260 170,301,039

This base year model is used for further analysis for scenarios for the horizon years.

3.2 Base Year Emissions Model

SIM-air8 models have been used to estimate the base year and future year emissions in the city from
transport and other sources like industries, households and road dust. SIM-air calculates base year
transport emissions based on the vehicle ownership numbers in the city and the average vehicle
kilometres travelled. The vehicle kilometres travelled is derived from the travel demand model. For the
base year, the travel demand model is calibrated based on the average trip length for various modes
estimated from the household interviews. Petrol pump surveys provide the annual vehicle kilometres
travelled in the base year. Therefore this data is directly used in SIM-air. For the horizon years however,
travel demand model results are used to estimate the annual kilometres travelled for various modes.
Some of the inputs for the base year SIM-air model are explained in this section.

An imaginary 44x44 square grid with each cell having an area of 1.18km2 is placed over Visakhapatnam
to cover the entire GVMC area and the nearby industries that cause pollution within the city. Results

8
Urban Emissions Info. (2014). SIM-air. http://urbanemissions.info/model-tools/sim-air.html

89
from the travel demand model developed earlier are used to distribute the emissions within these cells
in the following steps:

 The vehicle kilometres travelled on each link are derived from the calibrated base year model
 Using layer analysis in TransCAD 5.0, the grid layer is overlaid with link layer, and the cell-wise
vehicle kilometres are derived for the entire grid
 These vehicle kilometres are used as an indicator of the vehicular emissions in that cell, and are
used to distribute the total emissions calculated above into each cell.

The following figure shows the grid-wise distribution of vehicle kilometres travelled in the city.

Figure 38 Base year cell-wise vehicle kilometres travelled

The following are the other key transport related inputs used in the SIM-air model:

90
 Table 37 Base year SIM-air inputs
S.No Vehicle type Total no. Of vehicles Km travelled/year % fuel type
km Petrol Diesel Gas
1 2Ws 378,090 8,383 100% 0% 0%
2 Cars 66,737 13,107 46% 47% 7%
3 MUVs 13,813 9,788 0% 100% 0%
4 Taxis 4,736 13,107 46% 47% 7%
5 3Ws 25,862 18,656 0% 100% 0%
6 Buses 1,286 76,285 0% 100% 0%
7 HDVs 11,384 20,775 0% 100% 0%
8 LDVs 2,510 30,590 0% 100% 0%
9 Total 504,418

The total emissions caused by various modes have been calculated based on the vehicular emission
factors given by the Automobile Research Association of India (ARAI). Dispersion modelling is used to
find the ambient air pollution based on these emissions. These values are compared with the ambient
air quality measures (AAQM) data collected from the Andhra Pradesh Pollution Control Board (APPCB),
and the SIM-air model is calibrated for the base year. Based on these inputs, the aggregate transport-
related emissions in the city are estimated and shown in the following table. This calibrated model is
used for estimating the emissions for horizon years.

Table 38 Base year transport emissions

Type of pollutant Unit Total emissions
PM2.5 tonnes 609.8
PM10 tonnes 677.6
SO2 tonnes 71.0
NOx tonnes 7364.0
CO tonnes 16509.5
VOC tonnes 11164.2
CO2 mil tonnes 0.9
CO2 tonnes per capita 0.40

91
4 Scenario Analysis – Business As Usual
The objective of the Low-Carbon Comprehensive Mobility Plan (LCMP) is to plan for a sustainable and
low-carbon development of the future transport systems in the city. 2030 is the horizon year
considered, as around 20 years is a sufficiently long-term target to plan for and implement the low-
carbon interventions proposed in the LCMP. To understand the likely transport system in the horizon
year and its impacts, various scenarios of development have been analysed. These include:

 Business As Usual (BAU) scenario where the current pattern of development is assumed to
continue, and its likely impacts on the congestion, emissions and traffic fatalities in the city are
estimated.
 Low-Carbon Development (LCD) scenarios where various interventions are tested to make the
future development more sustainable and lower carbon than the BAU scenario. A methodology for
these scenarios has been prepared by UNEP and its project partners, including IIT Delhi, CEPT
University Ahmedabad and IIM Ahmedabad, and the same is adopted for the current study. Four
major areas of intervention have been identified for LCD scenarios:
o Land use intervention
o Public transport intervention
o Non-motorised transport intervention
o Technology intervention

Each of these scenarios are modelled both in terms of the transport planning impacts they have on
the city, like mode shifts, corridors of usage, demand projections and also the likely reduction in
emissions in the city. Likely traffic fatalities caused in each of the scenarios have also been
estimated.

The current chapter explains the outcomes of the BAU scenario, while the LCD scenarios are explained in
the next chapter.

4.1 Population and Trip Rate Projections

Population projection forms the first step in forecasting the travel demand for the future years, and it
remains the same across all the scenarios being analysed for the LCMP. The future population estimates
of the city are comprised of the following two major factors:

 Natural growth in population, considering the future birth and death rates

92
  Additional population growth due to migration resulting from the economic growth of the city.

The population growth for Visakhapatnam has been adopted from the forecasts of the UN Population
division9. The Population division of United Nations (UN), Department of Economics and Statistics,
carries out population projections based on these factors, and the projections have been carried out
until the year 2025. The same growth rates have been adopted for the current study until 2025 and
between 2025 and 2030, furthermore the growth rate between the years 2020 to 2025 is assumed to
continue. Even though the five-year period between the years 2020 to 2025 showed a smaller growth
rate than 2015-20, 2025-2030 considered the same growth rate to continue to be on the safer side.

The per capita trip rate of 1.61 is assumed to continue even for the horizon years, since available
literature shows that trip rates do not change significantly over time.10 This is because the work and
education trips remain constant even in the horizon year, and any increase in trips will be for social and
recreational purposes. These trips are made during off-peak hours and are far fewer than the work and
education trips. Hence these trips can be taken care of if the work and education trips are planned for.
Based on this assumption, the total trips in the city have been calculated for the horizon years. The
following table shows the growth rates, population projections and total trips made in the city for
various horizon years.

Table 39 Population and trip projections for horizon years

Year Annual growth rate Population (in Lakhs) Total trips/day (in lakhs)
(%)*

2011 2.65 17.46 28.11

2015 2.81 19.39 31.22

2020 2.77 22.27 35.85

2025 2.77 25.70 41.38

9
United Nations, Department of Economic and Social Affairs, Population Division. (2011). On-line data: Urban
agglomerations. http://esa.un.org/unup/unup/index_panel2.html
10
Mohan, D. (2008, January). Mythologies, metros & future urban transport. Transportation Research and Injury
Prevention Programme, Indian Institute of Technology, IIT, Delhi. From
http://tripp.iitd.ernet.in/delhibrts/metro/Metro/Metro%20Mythology08.pdf

93
 2030 29.46 47.43

(*Source: UN Dept. of Economics and Social Welfare)

4.2 BAU Scenario

The same four-stage modelling procedure as applied to the base year is applied to the horizon year, i.e.
2030. The master plan of a city is the guiding document for its pattern of growth. Hence, the master plan
of Visakhapatnam prepared by the Visakhapatnam Urban Development Authority (VUDA) is taken as the
basis for estimating the BAU development trends in the city. The following steps explain how the BAU
scenario is analysed.

4.2.1 City growth – 2030

To accommodate the increase in population, the built-up space in the city is expected to grow in the
future. Currently only 166km2 out of the 534km2 of the GVMC area is built-up. Hence there is ample
scope within the city to accommodate the increase in population. However, the pockets of the city
where this increased population is accommodated needs to be examined. The master plan land use plan
for the future is shown in the following figure.

94
 Figure 39 Land use plan according to the master plan – 2030, Visakhapatnam

It is observed that the proposed land use plan is not very different from the existing land use in the city,
i.e. even in the future the commercial developments are proposed for the core-city area, and industrial
developments are also proposed for the current industrial areas. Residential land use is concentrated
mostly in the core-city area, like it is happening now. Hence in the BAU scenario where the master plan
gets implemented, it is assumed that the future population growth in each TAZ will be the same across
the entire city. The 2030 populations of each TAZ in such a scenario are estimated and the resulting
population densities are shown in the following figure.

95
 Figure 40 2030 BAU population densities (km2)

It can be observed that the core area will have many zones with population densities in excess of 60,000
people per square kilometre, while the outskirts of the city continue to have densities less than
10,000km2. The travel demand and supply requirements are modelled, identifying the drawbacks and
recommendations to be implemented in the city for its low-carbon growth.

4.2.2 Road network – BAU 2030

The master plan provides the proposed Rights Of Way (ROWs) for various roads across the city. It is
assumed that these ROWs will be built in the future, and they’ll be developed according to the ‘urban
road design guidelines’ of India. The number of road lanes and free-flow speeds in such a scenario are
derived based on each road’s ROW and the corresponding lane configuration according to the codes.

96
The travel times and capacities of various roads are estimated based on these inputs. The following
figure shows the proposed ROWs that are adopted for the BAU scenario.

Figure 41 Proposed ROWs as per the master plan

4.2.3 Trip generation and distribution

Based on the population projected for each TAZ, the number of households is estimated from the
horizon year household size given in UNEP’s ‘Promoting low carbon transport in India’11. The trip
production equations derived for the base year are assumed to hold good, even for the horizon year.
This is because it is established in the literature that the number of trips made per person, which is just a
function of household size, will generally remain constant over time, and only the choice of modes

11
UNEP (2013, August). Promoting low carbon transport in India. New Delhi: Magnum.
http://www.unep.org/Transport/lowcarbon/Pdf's/LowCarbonCity_Guidebook.pdf

97
varies (Gadepalli et. al., 2013)12. Purpose-wise trips for 2030 are hence derived based on these
assumptions.

The land use plan does not envisage any different locations of growth for commercial and industrial
areas. Hence the relative attractiveness between various TAZs is assumed to be the same as the current
scenario, even for 2030. With this the trip purpose-wise PA table for the BAU scenario is prepared using
the same variables as the base year, i.e. household size for trip production, and land use pattern for trip
attraction. A total of 4,743,060 daily trips per day are estimated to be made in 2030. The following
figures show the intensity of the number of trips produced and attracted to each zone in the BAU
scenario.

The trip distribution for these trips is carried out using the gravity method for the calibrated parameters
estimated for the base year.

Figure 42 Trip production and attractions for BAU – 2030

Trip Productions - BAU 2030 Trip Attractions - BAU 2030

4.2.4 Mode shift for the BAU scenario

The existing investments in sustainable transport infrastructure like walking, cycling and public transport
is minimal, and in the BAU scenario this is assumed to continue. This causes people to shift towards the

12
Gadepalli, R., Jahed, M., Ramachandra Rao, K., and Tiwari, G. (2014). Multiple classification analysis for trip
production models using household data: Case study of Patna, India. Journal of Urban Planning and
Development, 140(1), 05013001. From http://ascelibrary.org/doi/abs/10.1061/%28ASCE%29UP.1943-
5444.0000168

98
modes for which the infrastructure is being provided, i.e. primarily cars and 2-wheelers and to some
extent auto-rickshaws. Exactly how many people shift varies from one zone to another based on the
zone’s values for various parameters that affect mode choice.

Mode shift analysis has been carried out for each TAZ separately based on the data collection and
projections. The assumptions taken for the shift are explained here using the aggregated city-level data
as an example. Several variables mentioned in the literature have been used to prepare a multinomial
logit model using the household interview data. The following variables have been explored for the
model:

 Household income
 Gender
 Vehicle ownership
 Trip length
 Travel time & cost
 Comfort of travel

Out of these variables, household income, vehicle ownership and trip length are observed to be the
significant variables in explaining mode choice behaviour. Vehicle ownership however is a function of
the income of the household, and hence household income and trip length are the two variables used to
estimate the mode shifts in various scenarios.

The following tables show the mode share variations for various trip lengths and income ranges in the
base year, derived from the household interview data. Incomes have been taken in absolute numbers
instead of the household asset-based low, middle and high income groups, to quantitatively measure
the proportion of people in each income group and their likely shift.

Table 40 Trip length vs mode share in Visakhapatnam: base year

Trip length range Car 2W Bus 3W Walk Cycle Others Total

<1 km 1% 7% 2% 3% 84% 2% 0% 100%
1-3 km 2% 21% 18% 18% 33% 7% 1% 100%
3-5 km 2% 25% 38% 19% 9% 5% 1% 100%
5-10 km 3% 29% 46% 16% 0% 4% 1% 100%
>10 km 4% 21% 61% 10% 0% 1% 2% 100%
Total Trips 2% 15% 18% 9% 52% 3% 1% 100%

99
 Table 41 Household income vs mode share in Visakhapatnam: base year

Income ranges (Rs.) Car 2W Bus 3W Walk Cycle Others Total

< 5000 1% 5% 20% 8% 61% 5% 1% 100%
5000-10000 1% 13% 17% 9% 55% 3% 1% 100%
10000-20000 2% 23% 18% 11% 43% 2% 1% 100%
20000-50000 7% 29% 21% 9% 31% 2% 0% 100%
> 50000 12% 19% 15% 11% 39% 2% 1% 100%
Total trips 2% 15% 18% 9% 52% 3% 1% 100%

For the horizon year, the household income is projected to grow at an annual growth rate of 8 per cent
for all income groups, and the new income ranges for the city are shown in the following table.

Table 42 BAU scenario income projections

Income range (Rs.) <5000 5000-10000 10000-20000 20000-50000 >50000 Grand total
Base year 26.7% 42.0% 20.8% 8.9% 1.6% 100%
2030 BAU scenario 0.5% 1.5% 11.5% 55.2% 31.2% 100%

Also, in a BAU scenario, the residential areas are spread across the city while the commercial areas are
concentrated in the core city area. This results in increased trip lengths for the people residing in the
outgrowths of the city. The aggregate trip length distribution in the city is derived from the origin-
destination matrix available in the travel demand model, which is shown in the table below.

Table 43 BAU scenario trip length range

Trip length <1 km 1-3 km 3-5 km 5-10 km >10 km Grand total

Base year 54% 17% 10% 10% 9% 100%
2030 BAU scenario 20% 25% 9% 17% 28% 100%

The BAU mode shares are estimated for these projected incomes and trip lengths based on the mode
shares shown in Table 40 and Table 41. The aggregate mode shares for the horizon year BAU scenario
model are shown in the following table.

100
 Table 44 BAU scenario – 2030 aggregate mode shares

Mode Proportion of trips

Car 8%

2W 25%

Bus 10%

Auto 19%

Walk 36%

Cycle 2%

Total 100%

These mode shares are applied to the OD matrices derived from trip distribution.

4.2.5 Trip assignment

The resultant OD matrices are assigned on to the 2030 road network to estimate the likely traffic load
on various corridors. The resultant network flow diagram is shown in the following figure. The results
from this travel demand model are used as one of the inputs for SIM-air to estimate the emissions. The
summary statistics of the model are shown in section 4.2.6.

101
 Figure 43 BAU Scenario – 2030 network loads

It is observed that the majority of links in the core city area are very heavily loaded, and will have a
volume to capacity ( V/C )ratio of more than 1.5 even for the master plan ROWs. The master plan ROWs
themselves are quite ambitious since creating them in the core area would be next to impossible.
However, the links in the three outgrowths, i.e. Madhurawada, Pendurti and Gajuwaka, will continue to
be underutilised even in 2030. Supporting such a scenario would require developing a few expressways
in the cities and many flyovers to relieve the resultant congestion. This would need heavy capital
investment from the GVMC and also would induce even more people to shift to cars and 2-wheelers.

4.2.6 Emission modelling

SIM-air is applied for the horizon year using the same methodology as followed for the base year, and
the likely emissions in the BAU scenario are estimated. The following two changes have been made to
the base year SIM-air model to make it reflective of the BAU scenario:

 The total number vehicles in the city has been increased based on the increased number of trips
and the estimated share of each mode among these trips
 The vehicle kilometres travelled for each vehicle have also been increased for the increased trip
lengths predicted for each mode.

102
The fuel type mix for various modes has been kept the same as the base year. The following table shows
the input values given for the BAU scenario:

Table 45 SIM-air inputs for BAU scenario: 2030

S. No. Vehicle type No. of vehicles VKT/year % fuel type
km Petrol Diesel Gas
1 2Ws 1,248,716 10,898 100% 0% 0%
2 Cars 528,989 17,039 46% 47% 7%
3 MUVs 14,366 12,724 0% 100% 0%
4 Taxis 37,556 17,039 46% 47% 7%
5 3Ws 108,191 24,253 0% 100% 0%
6 Buses 1,341 99,171 0% 100% 0%
7 HDVs 36,315 27,008 0% 100% 0%
8 LDVs 8,007 39,767 0% 100% 0%
9 Total 1,983,481

The vehicle kilometres grid has been prepared even for the 2030 BAU traffic loads, and is incorporated
in the model. The resulting emissions values are summarised in the following table.

Table 46 2030 BAU scenario emissions

Type of Pollutant Unit Total Emissions
PM2.5 tonnes 4,443
PM10 tonnes 4,934
SO2 tonnes 307
NOx tonnes 37,539
CO tonnes 79,172
VOC tonnes 49,132
CO2 mil tonnes 4.3
CO2 tonnes per capita 1.07
It is observed that the per capita CO2 emissions would increase by 2.5 times from 0.4 tonnes CO2 in the
base year to 1.1 tonnes CO2 in the horizon year. This is a quantum jump and is completely against the
low-carbon development targets that the country has set for itself. The BAU scenario shows that the
emission levels are likely to increase substantially from the base scenario. Therefore, a backcasting
approach has been used where a number of attritions were conducted to achieve the base scenario
modal share composition based on the two variables income and trip length, which were used to
estimate the mode shifts in various scenarios. Thereafter, vehicular trips by each mode were obtained,
and this city traffic split by each mode was such that the CO2 emissions did not exceed the above

103
emission-level vision. Additional measures were identified for aligning the LCMP with the national
transport action plan.

4.2.7 Traffic fatality modelling

Traffic fatality projections have been carried out for the horizon year based on the methodology13
explained below:

 The traffic fatality data collected from the first information reports (FIR) from the Traffic Police are
used as the input
 A victim vs impacting vehicle matrix is prepared for the fatalities
 The annual average distance travelled by each mode is derived from the travel demand model
 An injury risk matrix, i.e. fatalities/((distance travelled by victim mode) x (distance travelled by
impacting mode)), is created for each cell of the victim vs impacting vehicle matrix and distance
travelled by each mode
 The distance travelled by each mode in the horizon year is derived from the travel demand model
for the BAU scenario
 Fatalities for the horizon year are estimated by multiplying the cells of the injury risk matrix by the
distance travelled by the victim and impacting vehicles corresponding to that cell of the matrix.

Based on this methodology, the mode-wise fatalities are estimated for the BAU scenario and the results
are shown in the table below. It is observed that the total fatalities would go up from 425/year in the
base year to 1606 in 2030. This is due to the high risk exposure for pedestrians and 2-wheelers observed
in the city, and the higher number of trips made in 2030. This is clearly not acceptable, and adequate
measures to reduce these fatalities need to be taken in the city.

13
Woodcock, J., Edwards, P., Tonne, C., Armstrong, B. G., Ashiru, O., Banister, D., et al. (2009). Public health
benefits of strategies to reduce greenhouse gas emissions: Urban land transport (Health and Climate Change series
No. 2), Lancet, 374, 1930-43. Published online November 25, 2009, DOI: 10.1016/S0140‐6736(09)61714‐1.

104
 Table 47 2030 BAU scenario traffic fatalities

Modes/fatalities* Truck Bus Car MTW Cycle Total

Truck 5 1 18 17 0 41

Bus 1 0 0 3 0 4

Car 29 0 42 9 0 80

MTW 181 47 471 109 0 808

Cycle 13 4 35 13 0 65

Pedestrians 58 30 341 178 1 608

Total 287 82 907 329 1 1606

*Based on fatalities caused per 100,000 users of the mode derived for the base year.

4.3 BAU Scenario Summary

The following are the key observations for the BAU scenario analysis for 2030:

 The household incomes will increase, thereby increasing vehicle ownership and changing trip-
making patterns. The proposed concentration of commercial development in the core city will
increase trip lengths and hence induce mode shifts
 The mode share of walking will reduce from 52 per cent currently to 36 per cent, and cars and two-
wheelers would increase from 17 per cent to 33 per cent. The number of bus trips would reduce by
half and auto-rickshaw trips would double
 The above mode shifts will cause the addition of 15 lakh vehicles on to the road, which will result in
severe congestion on many corridors. Even the ROWs envisaged by the master plan would not be
able to cater to this demand
 The per capita CO2 emissions would increase by 40 per cent, which is against the targets of the
national climate change policies
 The traffic fatalities will increase up to 1,606 per year.

105
Considering these drawbacks, it can be concluded that the BAU scenario is not even an option for the
city, and various sustainable and low-carbon measures need to be put in place to avoid it. The following
chapter discusses the options that the city can adopt to grow along a more sustainable and low-carbon
path.

106
5 Low-Carbon Development Scenarios
It is evident from the earlier chapter that BAU is not an option in line with national and international
goals for sustainable transport and climate change. Therefore to avoid such a scenario, various other
low-carbon and more sustainable scenarios of development have been studied. These are presented as
low-carbon scenarios in the current chapter. The interventions can relate to land use, public transport,
non-motorised transport, or technology interventions like fuel/vehicle technology, etc. Each of these
scenarios and the likely outcomes on the travel pattern in the city, emissions and traffic fatalities are
explained in this chapter.

The global emission pathways designed to stay within the 2°C rise in global temperature require an
overall reduction in emissions from the 2010 level, as shown in the figure below. India subscribes to the
2°C vision, and therefore the Low-Carbon Development (LCD) scenario assumes the cities will make
efforts to limit CO2 emissions so that the targeted emissions for the horizon year of 2030 are equivalent
to existing levels, or slightly higher. The LCD scenarios will together aim to meet this target at least for
the transport sector.

Figure 44 Emission pathways to reach global temperature targets

107
5.1 Land Use (LU) Scenario

The BAU land use pattern, as envisaged by the Visakhapatnam Urban Development Authority (VUDA)
master plan, concentrates the commercial activity in the city in the core area, while residential plots are
developed all across the city. This would result in a further growth of population density in the core city
area, since people would prefer to stay closer to their jobs. However, it is not a desirable situation for
the city as that would mean that the population density in the core-city will be more than 44,000 people
per square kilometre in 27 of the 97 TAZs (Census, 2011). This in turn will lead to more than 70,000 trips
made per day from each of these TAZs, which will stretch the GVMC’s resources to provide
infrastructure and access to the required amenities for so many people. However, at the same time
there would be up to 26 zones in the outgrowths with population densities of less than 9,000
people/km2. Also, the trip lengths would increase in the city, since people in the outgrowths would still
need to travel to the core city for their work, shopping and other activities. Longer trips would
encourage the use of motorised modes of transport, and hence increase the carbon footprint of the city.

In the land use scenario, an alternative pattern of city growth is envisaged where the commercial
activity is spread across the entire city, i.e. even the outgrowths. Since the core city already has a high
density of population, this would also ensure that people move to the outgrowths. Such a scenario is
modelled and the results are explained in this section of the report.

5.1.1 Land use – travel demand model

The outgrowths are assumed to have a higher population growth rate compared to the core city area,
since it already has a high density. Even among the outgrowths, the zones adjacent to the highway and
major arterial roads are given a higher growth rate since they have better access to amenities and would
be more attractive for people to live in. The following figure shows the growth factors adopted to
distribute the projected city population for 2030 among various TAZs. The 2011 population of each TAZ
is taken as the base for the growth factor.

108
 Figure 45 TAZ-wise growth factors for land use scenario population growth

For this population growth and the distribution of commercial activity explained above, there would be
a better spread of productions and attractions across all TAZs, and the number of TAZs with high
intensity productions will also reduce by half. The following figure shows the productions and
attractions for the LU scenario.

109
 Figure 46 Land use scenario trip productions and attractions

LU-Productions 2030 LU-Attractions 2030

Due to the mixed land use pattern of growth proposed for the city in the land use scenario, there would
be shorter trip lengths even in the newly developed outgrowth areas in the city. The following table
shows the trip length distribution in the land use scenario derived from the travel demand model.

Table 48 Land use scenario trip length distribution

Trip length <1 km 1-3 km 3-5 km 5-10 km >10 km Grand total

54% 17% 10% 10% 9% 100%

2030 BAU scenario 20% 25% 9% 17% 28% 100%

2030 Land use scenario 37% 22% 13% 12% 17% 100%

The income ranges would be the same as in the BAU scenario. The mode share patterns explained in
Table 40 and Table 41 are applied for the LU scenario trip length distribution for each TAZ. The resultant
mode shares are applied to the aggregate person trip-based OD matrix from the BAU scenario, and the
mode-wise OD matrix for the LU scenario is derived for 2030. The aggregate citywide mode shares for
the LU scenario are shown in the following table.

110
 Table 49 Land use scenario mode shares
Mode Proportion of Trips

Car 4%
2-wheeler 21%
Bus 11%
Auto-rickshaw 22%
Walk 38%
Cycle 4%
Others 1%
Total 100%

The mode-wise person trips are converted to vehicle trips using their occupancy values, and the OD
matrix is assigned on to the network. The following figure shows the likely traffic loads in the LU
scenario.

Figure 47 Land use scenario – 2030 network loads

111
It is observed that the traffic loads in the core area have significantly reduced from that of the BAU
scenario, and can be handled by the ROWs proposed in the master plan.

5.1.2 Land use scenario – emissions and fatality projections

The parameters used in the BAU scenario are taken as the base even for LU emissions modelling. Some
of the inputs were changed to reflect the travel demand modelling results for the LU scenario, such as:

 The likely number of vehicles in the city, based on the mode shares estimated in the LU scenario
 The trip assignment results from LU scenario, which were updated to reflect the vehicle
kilometres travelled in each cell so that the aggregate emissions in the city were distributed
across the city accordingly.

The fuel type mix for various modes has been kept the same as the base year. The following table shows
the input values given for the LU scenario:

Table 50 SIM-air inputs for the land use scenario

S. No. No. of vehicles VKT/year % fuel type
Vehicle type
km Petrol Diesel Gas
1 2Ws 1,029,119 10,898 100% 0% 0%
2 Cars 267,091 17,039 46% 47% 7%
3 MUVs 38,676 12,724 0% 100% 0%
4 Taxis 13,261 17,039 46% 47% 7%
5 3Ws 122,869 24,253 0% 100% 0%
6 Buses 1,457 99,171 0% 100% 0%
7 HDVs 31,875 27,008 0% 100% 0%
8 LDVs 7,028 39,767 0% 100% 0%
9 Total 1,511,377

112
The resultant emissions for this scenario are shown in the following table.

Table 51 2030 land use scenario emissions

Type of pollutant Unit Total emissions
PM2.5 tonnes 1435.3
PM10 tonnes 1594.5
SO2 tonnes 209.6
NOx tonnes 65552.9
CO tonnes 33768.4
VOC tonnes 40054.0
CO2 mil tonnes 2.8
CO2 tonnes per capita 0.68
It can be observed that the CO2 emissions per capita has reduced from 1.1 tonnes in the BAU scenario to
0.68 tonnes in the LU scenario, i.e. a reduction of 40 per cent.

The traffic fatalities have also been projected based on the methodology explained in section 4.2.6 using
the vehicle kilometre numbers for the land use scenario. It is observed that the number of fatalities in
such a scenario reduces to 1,050 against the 1600 fatalities estimated in the BAU scenario. This is due to
the fewer vehicle kilometres travelled due to the shorter trip lengths in the land use scenario. The
following table gives the mode-wise split of these 1,050 fatalities.

Table 52 2030 land use scenario traffic fatality projections

Victim mode Fatalities

Truck 29

Bus 5

Car 29

MTW 461

Cycle 86

Pedestrians 435

Total 1045

113
5.2 Public Transport (PT) Scenario

Even though the LU scenario has lower emissions than the BAU scenario, it still has 1.5 million vehicles in
the city compared to the 0.5 million vehicles in the base year. This is due to the reduction in bus trips
and the huge increase in auto-rickshaws, cars and two-wheelers. As shown in Table 35, the daily average
occupancy of buses in Vizag is 30, while that of cars is 2.5, 2-wheelers is 1.5, and auto-rickshaws is 4.9.
This means that the reduction of one bus from the network can add 12 cars, 20 2-wheelers or 7 autos.
This shows that the best way to reduce the vehicle trips likely to be observed in the LU scenario is to
shift them from private modes of transport to city buses as far as possible, and to auto-rickshaws if
bussing is not feasible for the trip. Also, the emissions per passenger kilometre are lowest in buses,
followed by auto-rickshaws. Therefore, the carbon footprint would also reduce in addition to congestion
on the roads. The base year traffic fatality data show that travelling by bus or auto-rickshaw is much
safer than a 2-wheeler. In summary, shifting trips from cars and 2-wheelers to buses and auto-rickshaws
will be beneficial in terms of congestion, emissions and also traffic safety.

The public transport improvement scenario presents the analysis carried out to estimate the likely shift
of car and 2-wheeler trips to public transport, i.e. bus and auto-rickshaw. The LU scenario growth
patterns are taken as the base, and an iterative process of PT development is considered for this
analysis. This is because in the current scenario, the 670 buses available in the city carry 500,000 trips
per day, and they are already stretched to their capacity. Hence it cannot be expected of more people to
shift to this mode in the future. In the public transport scenario, the availability and quality of public
transport is increased incrementally and the likely shift it induces is observed in the following steps:

 The public transport scenario is analysed as an add-on to the land use scenario, i.e. mode shares
from the land use scenario are taken as the base for the NMT scenario
 The household income growths in the BAU still hold good, and the shift it would cause towards
cars and 2-wheelers is estimated
 In the base year scenario shown in Table 41, it can be observed that as household incomes
increase, public transport use reduces. This is due to the fact that the existing infrastructure
encourages private modes of transport through road improvements and free parking. The
existing system also actively discourages the use of public transport through inadequate routes,
low levels of comfort, an unreliable service and higher taxes
 In the public transport scenario, this trend is reversed through the provision of a good quality
public transportation system throughout the city. Auto-rickshaws are currently complementing

114
 the city bus system, and hence are considered to be a part of the overall public transport system
of the city, even in the public transport improvement scenario
 In this scenario, the supply is increased such that initially more buses and auto-rickshaws are
provided, covering a larger road network. Also, measures like BRT systems are introduced along
high demand corridors, and the supply of road space is limited for motorised vehicles
 Data on the willingness to shift to public transport in such a scenario is collected through the
stated preference surveys. However, due to the lack of a history and perception of good quality
public transport, and the policies discouraging private modes and increasing parking prices, the
willingness to shift numbers are similar to the BAU scenario. The resultant number of motorised
vehicles even in this scenario is more than the preferred amount, and the resultant reduction in
carbon footprint is negligible
 Therefore a strategic mode choice model is adopted to determine the likely mode share in the
public transport improvement scenario. Car and two-wheeler trips are shifted to public
transport modes like bus and auto-rickshaw
o The two significant variables for mode choice, i.e. income and trip length, are
considered for the analysis
o Car and 2-wheeler use are an indication of income, since in the BAU scenario higher
income literally translates into higher car and 2-wheeler use
o Of car and 2-wheeler users, the shift to bus and auto-rickshaw is estimated separately
for separate trip length ranges.
 The following are the mode shift assumptions derived from the stated preference survey:
o 25 per cent of car trips between 5 and 10km shift to auto-rickshaws
o 50 per cent of car trips longer than 10km shift to buses
o 50 per cent of 2-wheeler trips less than 3km shift to auto-rickshaws
o 25 per cent of 2-wheeler trips between 3 and 5km shift to bus, and 25 per cent shift to
auto-rickshaws
o 40 per cent of 2-wheeler trips longer than 5 and 10km shift to bus, and 10 per cent to
auto-rickshaws
o 50 per cent of 2-wheeler trips longer than 10km shift to buses.
 These shifts are applied to the trips in each TAZ and the resultant zone-wise mode shares are
derived.

The following table shows the aggregate mode shares of the city in 2030 derived after the PT scenario.

115
 Table 53 2030 mode shares – PT scenario
Mode Proportion of trips

Car 3%
2W 12%
Bus 29%
Auto 13%
Walk 38%
Cycle 4%
Others 1%
Total 100%

The OD matrices from the LU scenario are modified for these mode shares and are assigned to the
modified PT scenario road network. The following figure shows the link-wise bus loads on the network
for this OD matrix. The corridors with more than 5000 peak hour per direction trips (PHPDTs), and 15000
PHPDTs are highlighted since these are the potential links for the development of mass transit corridors
in the future. The infrastructure recommendations for this scenario are discussed in the next chapter.

Figure 48 PT scenario corridor loads

116
5.2.1 PT scenario – emissions and fatality projections

The following SIM-air inputs in the LU scenario were changed to reflect the travel demand modelling
results for the PT scenario:

 The likely number of vehicles in the city based on the mode shares estimated in the PT scenario
 The trip assignment results from the PT scenario are updated to reflect the vehicle kilometres
travelled in each cell so that the aggregate emissions in the city are distributed across the city
accordingly.

The fuel type mix for various modes has been kept the same as the base year. The following table shows
the input values given for the PT scenario:

Table 54 SIM-air inputs for the PT scenario

S. No. No. of vehicles VKT/year % fuel type
Vehicle type
km Petrol Diesel Gas
1 2Ws 588,771 10,898 100% 0% 0%
2 Cars 209,820 17,039 46% 47% 7%
3 MUVs 34,533 12,724 0% 100% 0%
4 Taxis 11,840 17,039 46% 47% 7%
5 3Ws 74,120 24,253 0% 100% 0%
6 Buses 3,904 99,171 0% 100% 0%
7 HDVs 28,460 27,008 0% 100% 0%
8 LDVs 6,275 39,767 0% 100% 0%
9 Total 957,723
The resultant emissions for this scenario are shown in the following table.

Table 55 2030 PT scenario emissions

Type of pollutant Unit Total emissions
PM2.5 tonnes 1,069.4
PM10 tonnes 1,188.0
SO2 tonnes 191.2
NOx tonnes 49,084.9
CO tonnes 26,085.7
VOC tonnes 24,750.2
CO2 mil tonnes 2.4
CO2 tonnes per capita 0.59

It can be observed that the per capita CO2 emissions have reduced from 1.1 tonnes in the BAU scenario
to 0.59 tonnes after the LU and PT scenarios, i.e. a reduction of 45 per cent.

117
The traffic fatalities in this scenario reduce significantly from the 1,050 predicted in the LU scenario to
around 800, even without any additional safety measures. This is due to the high proportion of PT trips
in this scenario, which are the safest among users of all modes in the city. The following table gives the
mode-wise estimate of traffic fatalities.

Table 56 2030 traffic fatalities – PT scenario

Victim Mode Fatalities

Truck 23

Bus 12

Car 20

2-wheeler 262

Cycle 84

Pedestrians 387

Total 788

5.3 Non-Motorised Transport (NMT) Scenario

It can be observed from the mode shares for the base year, and the horizon year mode shares for the
BAU, LU and PT scenarios, that the proportion of walking trips has reduced significantly in all the
scenarios. The cycle mode share has also reduced from 3 per cent to 2 per cent, and only the captive
cycle users are likely to use this mode, as they travel longer than walking trips and cannot afford any
other mode of transport. This is a worrying trend, since walking and cycling have no carbon footprint
and are the most significant modes in making the future mobility of the city low-carbon. Hence, in this
scenario various measures to encourage walking and cycling are assumed to be put in place, and the
resultant mode shift in such a scenario is explored.

Currently, the city only has 78km of footpaths out of the total arterial road length of 430km, and even
these footpaths are not continuous, universally accessible or properly lit at night. The funds allocated to
walking and cycling are less than 1/100th of the total funds allocated for transport. Consequently
pedestrians share ROW with vehicular traffic, resulting in up to 200 fatalities per year. Hence to reverse
this trend, safer walking and cycling infrastructure is recommended to be provided in this scenario. This
includes facilities like good quality footpaths and cycle tracks, street amenities and crossing facilities.

118
These facilities are expected to be provided on top of the LU and PT scenario recommendations, and the
likely mode shift in such a scenario is estimated. A bicycling compatibility index (BCI) is developed for
various links to understand the relative attractiveness of various corridors. This in turn will lead to a
mode shift towards cycling. The following are the assumptions for a mode shift in this scenario:

 The NMT scenario is analysed as an add-on to the public transport scenario, i.e. the mode shares
from the public transport scenario are taken as the base for the NMT scenario
 The household income growths in the BAU scenario still hold good, and the shift it would cause
towards cars and 2-wheelers is estimated
 In the base year scenario shown in Table 40, it can be observed that as trip lengths increase,
non-motorised transport, i.e. walking and cycling, reduces drastically. This is because the
existing infrastructure is hostile towards pedestrians and cyclists, and it is very uncomfortable to
make long trips for these users
 Hence it is observed that as the trip length increases, walking and cycling trips gradually shift
towards high-carbon motorised private modes like cars and 2-wheelers, and public transport
modes like buses and auto-rickshaws. It is to be noted that in this case, even the shift to public
transport, despite being low-carbon, is not desirable since walking and cycling are zero carbon
modes
 In the NMT improvement scenario, better infrastructure for walking and cycling is assumed to
be implemented in the city. This includes continuous footpaths and cycle paths, good crossing
facilities, parking for cyclists, signal prioritisation, strong parking policies and enforcement to
remove the hindrance for NMT users
 Data on the willingness to shift to walking and cycling in such a scenario is collected through the
stated preference surveys. However, due to the lack of a history and perception of good quality
public transport, and the policies discouraging private modes and increasing parking prices, the
willingness to shift numbers are similar to the BAU scenario. The resultant number of motorised
vehicles even in this scenario is more than the preferred amount, and the resultant carbon
footprint reduction is negligible
 Therefore a strategic mode choice model is adopted to determine the likely mode share in the
NMT improvement scenario
 Motorised trips are shifted to walking and cycling as a function of their trip length:
o The shorter trips are shifted to walking and cycling

119
 o Medium-length trips are shifted predominantly to cycling
o The longer trips in cars and two-wheelers are shifted to public transport modes like
buses and auto-rickshaws. This is because improving walking and cycling infrastructure
provides better access to public transport, making them more usable for users.
 The following assumptions are the detailed assumptions derived from the stated preference
survey:
o Of 2-wheeler trips shorter than 3km, 25 per cent shift to walking, and 25 per cent shift
to cycling
o 50 per cent of 2-wheeler trips between 3 to 5km long shift to cycling
o 25 per cent of bus trips between 3 to 5km long shift to cycling
o 25 per cent of auto-rickshaw trips between 3 to 5km long shift to cycling
o 50 per cent of 2-wheeler trips over 5km shift to public transport, i.e. bus and auto–
rickshaws, due to the better access provided by walking and cycling infrastructure
o 50 per cent of car trips over 10km shift to buses due to better access facilities.

The project recommendations for this scenario are explained in the next chapter. Based on these
assumptions, the TAZ-wise mode shares in this scenario are estimated. The following table shows the
city-level aggregate mode share derived for this scenario.

Table 57 2030 mode shares – NMT scenario

Mode Proportion of trips

Car 3%
2W 10%
Bus 25%
Auto 8%
Walk 45%
Cycle 7%
Others 1%
Total 100%

5.3.1 NMT scenario – emissions and traffic fatality projections

The following SIM-air inputs in the NMT scenario were changed to reflect the travel demand modelling
results for the NMT scenario:

120
  The likely number of vehicles in the city based on the mode shares estimated in the NMT
scenario
 The trip assignment results from the NMT scenario are updated to reflect the vehicle kilometres
travelled in each cell so that the aggregate emissions in the city are distributed across the city
accordingly.

The fuel type mix for various modes has been kept the same as the base year. The following table shows
the input values given for the NMT scenario:

Table 58 SIM-air inputs for the NMT scenario

S. No. No. of vehicles VKT/year % fuel type
Vehicle type km Petrol Diesel Gas
1 2Ws 514,559 10,898 100% 0% 0%
2 Cars 209,820 17,039 46% 47% 7%
3 MUVs 35,914 12,724 0% 100% 0%
4 Taxis 12,314 17,039 46% 47% 7%
5 3Ws 47,861 24,253 0% 100% 0%
6 Buses 3,374 99,171 0% 100% 0%
7 HDVs 29,598 27,008 0% 100% 0%
8 LDVs 6,526 39,767 0% 100% 0%
9 Total 859,966

The resultant emissions for this scenario are shown in the following table.

Table 59 2030 NMT scenario emissions

Type of pollutant Unit Total emissions
PM2.5 tonnes 978.4
PM10 tonnes 1087.0
SO2 tonnes 182.1
NOx tonnes 39,304.9
CO tonnes 24,671.3
VOC tonnes 22,051.6
CO2 mil tonnes 2.3
CO2 tonnes per capita 0.56

It can be observed that the CO2 emissions per capita have reduced from 1.1 tonnes in the BAU scenario
to 0.56 tonnes after implementing the LU, PT and NMT scenarios, i.e. a reduction of almost 50 per cent.
However, it should be noted that it is still higher than the existing per capita emissions of 0.4 million
tonnes in the base year. This is due to the fact that the mode share of NMT trips in this scenario is still

121
less than the base year due to the larger population expanding the urban sprawl and thereby increasing
trip lengths, shifting people away from walking. Also, the motorised modes have longer trip lengths,
increasing the vehicle kilometres travelled and hence the aggregate per capita emissions in the city.

Traffic fatalities have been estimated using the same methodology (section 4.2.6) as for the BAU, LU and
PT scenarios. The majority of the fatalities are currently being caused for NMT users and speeding 2-
wheelers. Since the NMT scenario assumes improved walking and cycling infrastructure and measures
like traffic-calming, it is assumed that the risk of fatalities reduces significantly. The actual amount of
reduction depends on the degree of infrastructure provided. Since these measures affect pedestrians,
cyclists and 2-wheelers the most, their fatality risk in this scenario is reduced by 75 per cent, while the
fatality risk for the other modes is kept the same as current. Even though the target should be to have
zero fatalities, this would represent the best case scenario in the current system of practice. The
resultant projection of fatalities in this scenario is shown in the table below. It is to be noted that the
annual number of fatalities in this scenario goes down to 250 in 2030, i.e. 8.6 fatalities per 100,000
people, against the 2012 number of 425, i.e. 24 fatalities per 100,000 people.

Table 60 2030 traffic fatalities – NMT scenario

Victim mode Fatalities

Truck 22

Bus 12

Car 20

2-Wheeler 55

Cycle 37

Pedestrians 107

Total 253

5.4 Technology Improvement Scenario

The technology improvement scenario is considered in addition to the land-use, public transport and
non-motorised transport scenarios explained above. This scenario explains the likely impact of
improving future vehicle and fuel technologies on the emissions caused in the city. This includes

122
improving the energy efficiency of existing vehicle and fuel types, and also the introduction of new fuel
types as listed below:

 The existing fuels, i.e. petrol and diesel, are assumed to be upgraded to Bharat Stage IV by 2015
and Bharat Stage V by 2020. The resultant reduction in emission factors of vehicles are
incorporated in the SIM-air model for the 2030 scenario with land use, public transport and non-
motorised transport initiatives
 The introduction of bio-fuels as an alternative to fossil fuels like petrol and diesel is explored in
this scenario. As per the projections of the International Energy Agency (IEA), the share of bio-
fuel in transport is taken as 9.5 per cent.

The resultant reduction in emissions in such a scenario is modelled in SIM-air, the inputs and results of
which are shown in the following table. The emissions in this scenario are likely to reduce to 0.40 (equal
to the base scenario) from 0.56 million tonnes in the NMT improvement scenario, i.e. a 28.5 per cent
reduction. The significance of implementing all the scenarios is evident when compared with the BAU
emission of 1.1 million tonnes per capita. By implementing all the scenarios, the total reduction in per
capita CO2 emissions is 63.63 per cent.

Table 61 SIM-air inputs for the NMT scenario

S. No. No. of vehicles VKT/year % fuel type
Vehicle type km Petrol Diesel Gas Bio fuel
1 2Ws 514,559 10,898 91% 0% 0% 10%
2 Cars 209,820 17,039 42% 43% 6% 10%
3 MUVs 35,914 12,724 0% 91% 0% 10%
4 Taxis 12,314 17,039 42% 43% 6% 10%
5 3Ws 47,861 24,253 0% 91% 0% 10%
6 Buses 3,374 99,171 0% 91% 0% 10%
7 HDVs 29,598 27,008 0% 91% 0% 10%
8 LDVs 6,526 39,767 0% 91% 0% 10%
9 Total 859,966

123
 Table 62 2030 NMT scenario emissions
Type of pollutant Unit Total emissions
PM2.5 tonnes 469.4
PM10 tonnes 521.5
SO2 tonnes 117.3
NOx tonnes 13,084.5
CO tonnes 11,866.9
VOC tonnes 19,956.7
CO2 mil tonnes 1.5
CO2 tonnes per capita 0.4

5.5 Summary of Scenario Analysis

Table 63 shows the summary of outcomes from the scenario analysis, and gives a comparative
performance of all the scenarios. It is clear from the table that a combination of all three interventions,
i.e. land use, public transport and non-motorised transport, are needed to be implemented together to
have the best case scenario for the city in terms of congestion, emissions and traffic fatalities. Moreover,
if there is improvement in the vehicle and fuel technology used, CO2 emissions can be further reduced
and brought equal to emissions in the base scenario.

124
 Table 63 Summary of scenario analysis
Scenario Base Year BAU BAU + LU BAU + LU + BAU + LU + BAU + LU + PT +
PT PT + NMT NMT +
Technology
Horizon Year 2011 2030 2030 2030 2030 2030
Population 1,730,320 2,946,000 2,946,000 2,946,000 2,946,000 2,946,000
Total trips 2,438,130 4,831,440 4,831,440 4,831,440 4,831,440 4,831,440
Car 2% 8% 4% 3% 3% 3%
2W 15% 25% 21% 12% 10% 10%
Mode share

Bus 19% 10% 11% 29% 25% 25%

Auto 9% 19% 22% 13% 8% 8%
Walk 52% 36% 38% 38% 45% 45%
Cycle 3% 2% 4% 4% 7% 7%
Car 181,982 1,442,475 728,319 572,149 572,149 572,149
Veh-km travelled

2W 1,414,115 4,670,392 3,849,064 2,202,096 1,924,532 1,924,532

Bus 166,768 173,932 188,943 506,251 437,523 437,523
Auto 250,779 1,049,113 1,191,440 718,725 464,099 464,099
Walk 887,479 1,217,523 1,291,196 1,291,196 1,520,230 1,520,230
Cycle 234,060 309,212 580,795 580,795 1,087,356 1,087,356
Total vehicles 504,418 1,983,481 1,511,377 957,723 859,966 859,966
on the road
CO2 emissions 0.94 4.34 2.76 2.39 2.28 1.5
(mill-tonnes)
Per capita CO2 0.40 1.07 0.68 0.59 0.56 0.40
emissions
(tonnes)
Traffic 425 1,600 1,050 800 250 250
fatalities/year

125
6 Recommendations
The various interventions in the city that need to be made to achieve the target mode shares shown in
the scenario analyses are explained in the current chapter. In the long term, the entire road network of
the city should be universally accessible by all modes as given in the code of practice for urban roads14.
However, these are segregated into long-term, medium-term and short-term recommendations to take
them up in a phase-wise manner. The road network development of the city forms the long-term
intervention, the mass transit corridors and their feeder routes form the medium-term intervention in
the city. Also, a few priority walking and cycling projects are identified that need to be implemented in
the short term. These recommendations are explained in detail in this chapter.

6.1 Long-Term Intervention: Road Network Development

Road network development involves two major components: regional connectivity and the hierarchy of
roads within the city. To cater to these two objectives, it is recommended that the road network be
developed in a ring-radial manner for the arterial roads. This involves making the following changes to
the road network:

 Diverting the existing NH-5 that passes through the city via SH-38, and thereby making it the
outer-ring road for the city
 Converting the existing NH-5 corridor into an urban road by providing various required
amenities like footpaths, bus stops, regular at-grade crossing facilities, etc.
 The NH-5 would then provide axes of growth for the city in three directions, one of which should
be extended to SH-38 to meet the proposed outer-ring road
 The NH-5 between NAD X-Road and Gajuwaka, together with the existing BRT corridor, would
act as the inner-ring road for the city.

The following figure shows the proposed ring-radial pattern of growth and the key corridors. Regional
connectivity is also being provided by public transport modes like railways and regional buses (public
and private). Intermodal integration hubs need to be developed for passengers at these regional mode
terminals so that they have a smooth shift to their preferred mode in reaching their destinations in the
city.

14
Codes of practice for urban roads. (2012). Institute of Urban Transport, Ministry of Urban Development. From
http://iutindia.org/downloads/Documents.aspx

126
 Figure 49 Proposed ring-radial pattern of growth for the road network

Madhurawada

Pendurti

Gajuwaka

6.2 Long-Term Intervention: Land Use Interventions

The population growth rate would be reduced in the core city area, and subsequently increased in the
three outgrowths of the city, i.e. Madhurawada, Pendurti and Gajuwaka (shown in the figure above).
This would encourage more commercial development in these areas than proposed in the master plan,
and encourage mixed land use provisions. This would ensure that the future holds a higher population
growth rate in these areas, leading to an even distribution of population across the city.

6.3 Long-Term Public Transport Interventions

As explained in chapter 4, public transport is currently comprised of two components: city buses and
auto-rickshaws. The measures required for these modes are explained in this section.

127
6.3.1 City bus system

The existing bus fleet is inadequate to cater to the raise in demand that is likely to occur in the horizon
years. Hence the existing fleet needs to be augmented with more buses. Currently, a bus carries 750
passengers per day on average. Assuming a better level of service and comfort, by the horizon year each
bus is assumed to carry 600 passenger trips per day. The additional fleet required in the city based on
this calculation is observed to be 1100 buses by 2030.

The fleet augmentation should be coupled with the following measures for it to be most useful:

 Better routing and scheduling of vehicles for them to be more efficient. The routes need to be
updated regularly and dynamic scheduling systems should be used to optimise the available
fleet
 Buses need to be universally accessible, i.e. children, women, elderly people and wheelchair
users should also be able to use them. As many low-floor buses as possible should be added to
the fleet to improve accessibility
 Supporting infrastructure like bus stops and access facilities to the bus stop need to be provided
 Adequate bus depots and terminals need to be provided to cater to the increased fleet required
for the horizon year.

The locations for these features are identified in addition to the inter-modal integration hubs mentioned
earlier, and are shown in the following figure. VUDA needs to allocate the land for upcoming depots to
the APSRTC for it to be functional.

128
Figure 50 Location of supporting infrastructure for buses

129
6.4 Medium-Term Intervention: Mass Transit Corridors

The PT intervention scenario envisages the development of various mass transit corridors in the city.
Figure 48 shows the estimated corridor loads for 2030, based on which the type of mass transit corridor
required for the city shall be identified. The planning commission guidelines given in the 12 th Five year
plan have been considered as a base for determining the type of corridor required. The following table
gives these guidelines:

Table 64 Selection criteria of mass rapid transit modes 15

Mode choices PHPDT in 2021 Population as per Average trip length for
2011 census motorised trips in km
(million)
Metro rail >=15000 for at least 5km >=2 >7-8
continuous length
LRT primarily at grade =<10,000 >1 > 7-8

Monorail =<10,000 >2 About 5-6

Bus Rapid Transit >=4,000 and >1 >5

System
up to 20,000
Organised city bus >1 lakh hilly towns >2 to 3
service (50,000)

The current population of GVMC is 1.7 million, the average trip length is 4.1, and PHPDTs more than
15,000 exist only in two stretches, both of which being less than 5km in length, even in 2030. The
guidelines suggest that these demands should be achieved in 2020 for the city to plan for a metro.
Hence metro is not a feasible option for the city. However, the city qualifies in two of the three criteria
for a BRT system, i.e. demand and population. Even the trip length is not too far from the required trip
length. Also, the city has a successful BRT system in operation, and should expand on this to improve its
network connectivity.

Considering these points, a total of 68km has been identified for the development of BRT corridors
across the city. Their order of priority is as follows:

1. NH-5 from Tagarapuvalasa to Maddilapalem

15
12th Five year plan, Proceedings of the working group on Urban Transport

130
 2. NH-5 from Maddilapalem to NAD Junction
3. Dwaraka Nagar to Steel Plant Via Scindia
4. NH-5 from NAD Junction to Gajuwaka
5. Asilmetta to Yendada via Beach Road
6. Gosala to Venkojipalem via NSTL

These corridors are shown in the city road network in the following figure. It can be observed that
together, these corridors provide mass transit connectivity even for the outgrowths of the city.

Figure 51 Existing and proposed BRT corridors in the city

The various salient features of these corridors are shown in the following table.

131
 Table 65 Salient features of selected BRT corridors

Corridor Length Likely 2030 Road Existing Master plan Cost estimate
no. (in km) ridership ownership ROW‡ (m) ROW (in m) (in Rs. Crores)
(in PHPDT*)

1 14 13,000 NHAI** 60 80 210

2 10 10,000 NHAI 60 80 150

3 19.2 9,000 Defence 14 40 290

4 8.5 10,000 NHAI 60 80 130

5 7 6,000 GVMC 18 45 105

6 9 3,500 GVMC 8 20 135

Total 67.7 1,020

*PHPDT – peak hour per direction trips

**NHAI – National Highway Authority of India

‡- ROW – Right of Way

It can be observed that the four corridors that yield the highest ridership and also have adequate ROW
for the future are either with the NHAI or with defence establishments like the Navy. However, based on
the recent directive from the Ministry of Urban Development (MoUD) to other ministries like the
Ministry of Road Transport and Highways (MORTH), the highways passing through cities need to be
designed as urban roads. Hence it is recommended that the identified corridors are either handed over
to the GVMC or retained with the current road-owning agency, but are designed as BRT systems. In
either case, the need for developing a BRT system along these corridors is of utmost importance for
future sustainable transport in the city.

The other corridors identified above currently have lower ridership, but can be taken up as a BRT line
keeping the future traffic needs of the city in mind. The figure above shows the corridors identified on
the road network map along with the existing BRT corridors. The network-level mass transit connectivity
that is achieved by implementing the recommended corridors can be observed from the figure.

132
6.5 Medium-Term: Feeder Route Network Development

A 30km feeder network to the BRT corridors identified above has also been identified and is shown in
the figure below. Routes are selected in such a way that they integrate the various mass transit corridors
and also provide access to these corridors for the entire city. These routes need to be developed in such
a way that the needs of the short distance trips made along these corridors are served. These routes
need to be designed in such a way that they provide the best possible access by walking, cycling, auto-
rickshaw and city bus.

The following figure shows the entire city network with the existing BRT system, proposed BRT and also
the feeder route network identified.

Figure 52 Feeder route network in addition to the existing and proposed BRT network

133
6.5.1 Auto-rickshaw interventions

Since auto-rickshaw operations are carried out by individuals, the recommendations are made in terms
of the policies and infrastructure needed to support them. The following are the major
recommendations to be taken to improve their operations:

 Appropriate parking/stands at all locations, such as:

o Commercial areas like shopping malls and retail areas

o Public spaces like parks, beaches and tourist locations

o Residential colonies: each colony to have an authorised auto-rickshaw stand.

 Limit to be raised on the total no. of autos in the city

o Currently more than 30,000 autos are registered with the Road Transport Authority
(RTA), but 8,000 is the limit that was existing for the erstwhile Visakhapatnam Municipal
Corporation (VMC). The limit has not been updated even after the formation of the
GVMC, and hence needs to be revisited. Considering the benefits of auto-rickshaws in
mobility, emissions and safety, it is recommended that the limit on their number be
lifted in the city.

 Stopping and boarding facilities in sync with bus-stops so that the auto-rickshaw services are
integrated with city buses and act as a feeder mode to them

 Shared autos to be used as both a primary public transport system in the outgrowth areas
where the demand doesn’t justify as bus system, and as a feeder service to city buses in other
areas based on corridor demand.

6.6 Short-Term NMT Interventions

Non-motorised transport comprises of walking and cycling infrastructure. While the mass transit and
feeder network identified above forms the medium-term intervention needed for the city, a few
immediate measures are needed for walking and cycling to improve the existing level of service and
safety situation. These short-term recommendations for these modes are explained separately in this
section.

134
6.6.1 Walking infrastructure

Currently, only 78km out of the 430km of arterial and sub-arterial roads in the city have some kind of
footpaths present. This should be upgraded to the entire 430km stretch over the course of time, and in
the long run all roads need to be pedestrian friendly, including the collector and access roads. A few
priority corridors to be taken up immediately are identified and shown in the following figure. Some of
the corridors are planned to be taken up under Public Private Partnership (PPP) mode to overcome the
current lack of funds to spend on walking infrastructure. All the footpaths to be developed need to
follow the ‘urban road design guidelines’ released by the Ministry of Urban Development.

Figure 53 Priority footpaths in the city

6.6.2 Cycling infrastructure and public bicycle sharing (PBS)

One of the major reasons for low cycling use in the existing scenario is the lack of safety on the roads for
the cyclists, and also the lack of parking for cycles in the city. On the arterial roads with ROW over 30m,
segregated cycle tracks are needed to separate the cyclists from high speed motorised traffic. On roads
with narrow ROW, traffic-calming measures need to be taken up so that the traffic speed is not too high
compared to the speed of the cyclists, thereby increasing their safety. All roads with a proposed ROW of
30m or higher, as per the VUDA master plan, are identified for the provision of segregated cycle tracks.
These links are shown in the following figure. The remaining roads should also be made cycle-friendly by
applying measures like traffic-calming, good crossing facilities and other amenities like street lighting.

135
 Figure 54 Links with ROW>30m for segregated cycle tracks

The city is interested in developing a public bicycle sharing (PBS) scheme to promote cycling further in
the city. Two networks have been identified that can act as a pilot for the implementation of PBS in the
city, and are shown in the following figure.

Figure 55 PBS networks identified in GVMC

PBS in the core city area PBS in Gajuwaka Area

136
6.7 Policy Recommendations

Various urban transport-related policies adopted by the Ministry of Urban Development (MoUD) at the
national level are recommended to be implemented in Visakhapatnam as well. The detailed
specifications of their implementation are given in the respective policy documents16:

 Forming the Unified Metropolitan Transportation Authority (UMTA) for Visakhapatnam, which
will be the final decision-making authority for all urban transport issues in the city. The UMTA
will comprise all relevant stakeholders, and shall be housed within the Visakhapatnam Urban
Development Authority (VUDA) on the lines of UMTA formed for Hyderabad16
 Setting up an Urban Transport Fund (UTF) for the city, which will form the central pool of funds
to invest in sustainable transport initiatives. The fund collects revenues from various sources like
parking, land value capture, green cesses on fuels, etc., that can be used to fund public transport
and non-motorised transport initiatives
 Adopting a parking policy for the city that recognises land as a public good and charges the area
provided for parking in such a way that the true value of the land is reflected in the parking
price. This includes creating separate parking price slabs for commercial and residential area,
the core city and outskirts, etc.
 Five per cent of the annual budget spent on transport needs to be allocated to non-motorised
transport (NMT), which includes projects on pedestrian walkways, crossing facilities, cycle
tracks, cycle parking, bus stops and other street infrastructure like street lighting, landscaping,
etc.
 Removing the existing limits on permits issued for auto-rickshaws in the city.

Other policy recommendations and reforms given by various national policy documents like the National
Urban Transport Policy (NUTP), the 12th Five year Plan – Urban Transport recommendations, the
National Transport Development Policy Committee (NTDPC), the High Powered Expert Committee
(HPEC) report, the National Mission for Sustainable Habitat (NMSH), and the National Action Plan for
Climate Change (NAPCC) need to be adopted by the city.

16
Ministry of Urban Development. (2014). Urban Transport. From http://moud.gov.in/urbantransport

137
6.8 Implementation Plan

Two of the above projects shall be taken up for the preparation of project proposals as a part of the
current study. The following are the two projects suggested by the stakeholders as the most important
for the LCMP to develop further for funding:

 Development of 32.5km of NH-5 into a BRT corridor

 Detailing the PBS projects and applying for funding.

138
 

6.9 References

 12th Five year plan, Proceedings of the working group on Urban Transport
 Codes of practice for urban roads. (2012). Institute of Urban Transport, Ministry of Urban
Development. From http://iutindia.org/downloads/Documents.aspx
 De Dios Ortuzar, J., & Willumsen, L. G. (2011). Modelling transport, 4th ed, Chichester: Wiley.
 Dhar, S., Pathak, M., & Shukla, P. R. (2013). Promoting low carbon transport in India: A
guidebook for city planners and practitioners. UNEP Risø Centre on Energy, Climate and
Sustainable Development.
 Gadepalli, R., Jahed, M., Ramachandra Rao, K., and Tiwari, G. (2014). Multiple classification
analysis for trip production models using household data: Case study of Patna, India. Journal of
Urban Planning and Development, 140(1), 05013001. From
http://ascelibrary.org/doi/abs/10.1061/%28ASCE%29UP.1943-5444.0000168
 Guevara, C. A., & Thomas, A. (2007). Multiple classification analysis in trip production models.
Transport Policy, 14, 514-522.
 IIT (Delhi), IIM (Ahmedabad), & CEPT University. (2012). Low-carbon comprehensive mobility
plan methodology.
 Ministry of Urban Development. (2014). Urban Transport. From
http://moud.gov.in/urbantransport
 Mohan, D. (2008, January). Mythologies, metros & future urban transport. Transportation
Research and Injury Prevention Programme, Indian Institute of Technology, IIT, Delhi. From
http://tripp.iitd.ernet.in/delhibrts/metro/Metro/Metro%20Mythology08.pdf
 United Nations, Department of Economic and Social Affairs, Population Division. (2011). On-line
data: Urban agglomerations. From http://esa.un.org/unup/unup/index_panel2.html
 Urban Emissions Info. (2014). SIM-air. From http://urbanemissions.info/model-tools/sim-
air.html
 VUDA. (2007, April). Revised master plan for Visakhapatnam metropolitan region – 2021.
 Woodcock, J., Edwards, P., Tonne, C., Armstrong, B. G., Ashiru, O., Banister, D., et al. (2009).
Public health benefits of strategies to reduce greenhouse gas emissions: Urban land transport
(Health and Climate Change series No. 2), Lancet, 374, 1930-43. Published online November 25,
2009, DOI: 10.1016/S0140‐6736(09)61714‐1

139
Annexure-I: Survey Formats

Household Survey Part I: (Revealed Preference Survey)

1. Reference
Date: Surveyor name:

Area: Ward No: Address/ Door No.:

Contact number of respondent (Landline and mobile): Email id:

Settlement Code: _____ (1. Apartment 2. RCC Ordinary Building 3. Masonry Terraced Building 4. Masonry Tiled Building 5. Ordinary Tiled Building 6. Hut)

2. Household Information (Socio-economic)

S. No. (Tick Sex
Name Relation with head Age Education Main Activity Subsidiary activity
Respondent) (M/F)
1 2 3 4 5 6 7
1 Head 1

140
 3. Household Assets
Household Assets owned
Y/N (Yes / No) Number
Mobile phone
Fridge
LPG Stove/Cylinder
Cooler
A.C.
T.V.
Desktop/Laptop Computer

4. Housing and Living Conditions

1 What is the type of the house? Kutcha Semi-Pucca Pucca
2. What is your total Household’s Income?
3 What is the tenure arrangement of the house you live in? Rented Shared
4 If it is Rented, What is the rent you pay for it? Rs. __________________ /Month
5 Tick and write the appropriate spaces in the house Rooms _________ (no.) Separate Kitchen Y/N Floors_____ (no.)
6 What is the area of the house? Area_______________ ______________(unit)
7 What is your property tax? (Tick and Write the appropriate value) Rs.____________ No
8 How long have you been staying in this house? ___________ Years
9 Where were you living before?
Household Tap (Y/N) Piped Sewerage (Y/N) H/h Toilets (Y/N) No.

5. Vehicle Ownership in the household 6. Accessibility to important destination

Distance Walking
Present Before 2 year Utility (km) minutes
Type Age/ Year Fuel Mileage Type Fuel Mileage Grocery Store
1 Milk booth
2 Vegetables
3 Dhobi
4 Doctor
5 e-Seva

141
 Type: Car, Two-Wheeler, Bicycle, Auto-rickshaw, Cycle Rickshaw Others (specify)
7. Travel Diary of each Individual (Separate Access and In vehicle trips for Bus and Auto)
HH member no Day of Trip: Mon/Tue/Wed/Thur/Frid
17 18
Seg Purpose Mode Start Location Waiting Time Start time End Location Travel time (min) Distance (km) Fare/Parking cost Trip Frequency
1
2
3
4
5
6

HH member no Day of Trip: Mon/Tue/Wed/Thur/Frid

Seg Purpose Mode Start Location Waiting Time Start time End Location Travel time (min) Distance (km) Fare/Parking cost Trip Frequency
1
2
3
4
5
6

HH member no Day of Trip: Mon/Tue/Wed/Thur/Frid

Seg Purpose Mode Start Location Waiting Time Start time End Location Travel time (min) Distance (km) Fare/Parking cost Trip Frequency
1
2
3
4

17
Trip Purpose: 1-Home, 2-Work, 3-Education, 4-Access to public transport, 5-Access to auto-rickshaw, 6-Shopping, 7-Recreation, 8-Social trip, 9-
Temple/Church/Mosque, 10-Personal business, 11-Other.
18
Mode: 1-Car, 2-2-wheeler, 3-Bus, 4-Auto-direct, 5-Auto-shared, 6-Walk, 7-Bicycle, 8-Cycle-rickshaw, 9-Company bus (or any other private arrangements), 10-
Taxi, 11-Any other (please specify).

142
 5
6

HH member no Day of Trip: Mon/Tue/Wed/Thur/Frid

Seg Purpose Mode Start Location Waiting Time Start time End Location Travel time (min) Distance (km) Fare/Parking cost Trip Frequency
1
2
3
4
5
6

8. Choices and opinions

No. Mode Nearest stop Time Avg. How often do you use it in a Is it reliable service? Is it safe? Is it too
(distance) taken Waiting week? (no. of times per week) expensive?
to time
reach
1 Public Bus Good Ok Bad Good Ok Bad Good Ok Bad
2 Shared Auto Good Ok Bad Good Ok Bad Good Ok Bad
3 Bicycle Good Ok Bad Good Ok Bad Good Ok Bad
4 Do you think it is safe and convenient to walk on roads of Vizag? Good Ok Bad
5 Are you satisfied overall, with the way you travel in the city? Yes No

9. What do you think needs to be improved in the city?

______________________________________________________________________________________________________________________________
______________________________________________________________________________________________________________________________
______________________________________________________________________________________________________________________________
______________________________________________________________________________________________________________________________

143
 Household Survey Part II: (Stated Preference Survey)
Scenario 1 Choices State the second preferred mode if 1st mode is not available
Member Work trip Daily needs School/College Member Walk Bicycle Bus MTW Car Auto Reason
1 1
2 2
3 3
4 4
5 5
6 6
7 Reasons for Not using the 2nd Mode
Inaccessible Not socially
Scenario 2 Choices i iv acceptable
Mem no Work trip Daily needs School/College ii Unaffordable v Unavailable
1 iii Fear of accident vi Others (specify)
2
3
4
5
6
7
Scenario 3 Choices
Mem no Work trip Daily needs School/College
1
2
3
4
5
6
7

144
 SCENARIO 1
Two
Attribute Car Bus Auto/Taxi Bicycle Walk
Wheeler
15% Less
More due to More due to More due to Comparable to car
Travel time (Independent 15% less (Footpath)
congestion congestion congestion (Indep lane)
lane)

More due to More due to More due to

Travel Cost increased increased Same increased - -
travel time travel time travel time

Frequency
- - - - - -
(Transit)

No gradient, better No gradient, indep

Same as Same as Same as Same as
Comfort surface, access footpath, better
today today today today
control, more width surface, more width

Better (Indep lane,

Same as Same as Same as Same as Better (Indep lane,
Safety Traffic speed
today today today today Traffic speed control)
control)

145
 SCENARIO 2
Two
Attribute Car Bus Auto/Taxi Bicycle Walk
Wheeler

More due to More due to 15% Less More due to 25 % less (Indep 15% less
Travel time
congestion congestion (Independent lane) congestion lane) (Footpath)

More due to
More due to More due to increased
Travel Cost increased increased 25 % Higher fare travel time, - -
travel time travel time increased
fare

Frequency
- - 20 % More - - -
(Transit)

More due to level

Same as boarding, leg No gradient, better No gradient, indep
Same as Same as
Comfort today for room, Standing surface, access footpath, better
today today
vehicle space, Air control, more width surface, more width
Conditioning

Better (Indep lane, Better (Indep lane,

Same as Same as Lesser Risk , Same as
Safety Traffic speed Traffic speed
today today lighting of stops today
control) control)

146
 SCENARIO 3
Two
Attribute Car Transit Auto/Taxi Bicycle Walk
Wheeler
15% Less
More due to More due to More due to 25 % less 15% less
Travel time (Independent
congestion congestion congestion (Indep lane) (Footpath)
lane)

More due to More due to

increased travel More due to increased
Travel Cost time, increased increased 25 % Higher fare travel time, - -
fuel cost, parking travel time increased
cost fare
Frequency
- - 20 % More - - -
(Transit)

More due to
No gradient, No gradient,
Same as today for level boarding
Same as Same as better surface, indep footpath,
Comfort vehicle, farther leg room,
today today access control, better surface,
parking places Standing space,
more width more width
Air Conditioning

Better (Indep Better (Indep

Same as Lesser Risk, Same as
Safety Same as today lane,Traffic lane,Traffic
today lighting of stops today
speed control) speed control)

147
 Codes for Part-I: Revealed Preference Survey
Code (Relation with Head of Education (5) Activities (6-7)

the Household) (2)

1. Self 1. Illiterate 1. Salaried employment (regular waged) 7. Household work

2. Wife / Husband 2. Literate 2. Daily Wages employment 8. Attending School

3. Son / Daughter 3. Primary education (Up to 8th) (casual labour) 9. Attending College

4. Grandson / Grand Daughter 4. Matriculation/up to 12th 3. Self Employed 10. Pensioners/remittance recipient

5. Mother / Father 5. Graduate (work in h/h enterprise) 11. Unemployed – due to disability

6. Mother in law / Father in law 6. Certificate Course 4. Domestic Worker at Fixed Rate 12. Unemployed – seeking work

7. Daughter in law 7. Others (Specify) 5. Honorary Worker 13. Others – specify

8. Uncle/ Aunt 6. Home-based paid work

9. Others

10. Brother / Sister / Nephew / Niece

148
 Traffic Volume Count Survey Format

Low-Carbon Mobility Plan for Visakhapatnam

Name of
Date
Intersection/Junction:______________ Day:
:
__________
Location_ID:_____________ Direction/ Movement
________ From:___________________________

Non Motorised
Motorised Traffic
Traffic
Passenger Vehicles Goods Vehicles

Bus Oth
ers
Time Ani
Car/Jee Trac Cycle (pl.
Period 7 seater Cy mal
Auto p/Van Goods 3- tor Ricks spec
Two (3w) Ta 2-Axle M cle Dra
rickshaw (Other INTER Autoricksha LCV Axle / haw ify)
wheelers / Maxi xi Mini CITY Pvt Truck AV wn
(3w) than CITY w (3w) Truck Trol
Cabs Bus BUS Bus
Taxi) Bus ly

:00

:15
:15

:30
:30

:45
:45

:00

Total

149
 Origin-Destination (OD) Survey Format
Origin-Destination Survey

Date: Location: Name Of The Enumerator:

Direction: Junction Whether:

Name:

S No Type of Registration Origin District Destination District/Place Purpose of Trip Occupancy Type of Route Remarks
Vehicle No Trip (ref Frequency Commodity Course(Via)
/Place
Code) & Tonnage
(if goods)

150
 Road Inventory Survey

Survey format 1. Inventory for pedestrian facility

Along Road (Survey format 1a)
Name Width of Segregation tools to separate From To Length Encroachment/ other barriers Pavement condition Lighting Barrier-free
of road footpath (m) footpath from MV lane (kerbs/ (km) (Parking/ vendors/trees/ light (Good/ Average/ (Access at en
L R green belt/ fences/ others) poles/ other services) Bad) Left Right guiding tiles
audible/non

At Intersection (Survey format 1b)

Name of intersection Type of intersection At Type of crossing Level/ raised/ Signalised Pedestrian Traffic- Crossing Barrier-free
grade/ flyover/ clover foot over bridge/ subway Y/N accentuated calming distance access
leaf/ roundabout/ others signal (y/n) tools (meters) guiding tiles/
Rumble audible
strips/ pedestrian
speed crossing /
breakers none

151
 Survey format 2. Inventory for NMV (bicycles and cycle rickshaws) facilities
Along Road (Survey format 2a)
Name Width Segregation tools to separate From To Length Encroachment Pavement condition Lighting
of of NMV lane from other modes (km) (Parking/ vendors) (Good/ Average/ Bad) L (y/n) R (y/n)
road NMV (Painted marking/ kerbed/
lane none)
(m)
L R

At Intersection (Survey format 2b)

Name of Type of intersection (At Type of crossing (Level/ Signalised (y/n) NMV Traffic-calming Crossing Other
intersection grade/ flyover/ clover raised/ grade separated) accentuated tools (Rumble distance facilities
leaf/ roundabout, etc) signal (y/n) strips/ speed (NMV box
breakers) etc.)

152
 Parking Area (Survey format 2c)
Name of Location Nearest Pt stop Distance Number of Parking charges
Parking lot to PT stop Parking Bicycle Cycle Rickshaw

153
 Survey format 3. Inventory for public transport (bus) – survey form 1-3B in ADB toolkit
Along Road (Survey format 4a)
Name Applicable in case of BRT corridor No. of Average
of Width of From To Length Bus lane location wrt Type of bus Bus lane Segregation tools routes speed
road Bus lane (km) road section (Median/ infrastructure (Kerbs/lane marking/ catered (kph)
(m) left side) (open/close) fences)
L R

Bus stop details (Survey format 4b)

Name Location Bus stop Applicable for BRT Near/ far Traffic- Safe crossing Access Barrier- Passenger
of Bus (coordinates) capacity Location Type of bus junction calming tools facility (Zebra distance to free amenities
stop wrt road stop (Rumble crossing/ bus stop access (Sitting area/
section (Staggered/ strips/ speed signalised from either (y/n) toilets/
(Median/ island) breakers) crossing/ FOB/ side (m) hawkers)
left side) subways)
X Y

154
 Survey format 4. Inventory for private motorised vehicles
Along Road (Survey format 5a)
Name of Width From To Length No. of Lighting Average speed (kph) Road encroachment On road Parking restriction Vehicl
road (m) (km) lane (Parking/vendor/other) (Restricted / Unrestriced) PMV
L R MTW Cars IPT

At intersection (Survey format 5b)

Name of Type of intersection (At grade/ Signalised Turning time from Type of traffic operation Signal Intersection
intersection flyover/ cloverleaf, etc) (y/n) each direction (Automated/ Human) Phasing design

155
 Parking areas (Survey format 5c)
Name of Parking Location Type of Parking (On/off road) Nearest Pt stop Distance to PT stop Number of Parking
Parking charge (km) Parking charges (Rs)
lot (Paid/ MTW Car MTW Car
unpaid)

156
 Auto-Rickshaw Operator Survey
Date: Name Of The Enumerator: Weather:
Location of Survey: Direction: Average Frequency along road
1 2 3
Registration No.
Make of Vehicle Mahindra/ Bajaj/ Ape etc.
Name of Driver
Age of Driver
Ownership 1-Owned, 2-Rented
Age of Vehicle
No. of Autos Owned
Residence location
Resident/ migrated
Registered Auto Stand
Type of Auto 1-4-seater, 2-7-seater
Usage type of Autos 1-Shared, 2- Direct, 3-Both
Description of Route Start
Via
End
No. of hours of operation per
day
Type of Fuel 1-Petrol, 2-Diesel, 3-CNG
Vehicular km/day
Occupancy Peak hour
Off-Peak
Average Income/day
Average Fare/ passenger

157
 Basis for fare fixation 1.Bus fare/ 2. distance travelled by
passenger/ 3.Operating cost/ 4. Any
other
Source of Finance/ Loan
Amount of loan on the Vehicle
Average Expenditure/day in
Operations Rent for Auto
Loan repayment amount
Fuel Expenditure
Maintenance costs
Fines & Bribes per day/month
Road Tax
Pollution Check
Parking Charge at night-stands
Property Tax for stand
Other daily expenditure
Household Income/ Month
Household Savings/ Month
Alternative Modes Bus/ 2-Wheeler/ Car/ Share auto/
Auto
Amenities required More Parking across the city
Night Stands for parking
Bus Integration
Problematic Locations
Public toilets+Water
Any other

158
 Petrol Pump Survey Format
1. Type of vehicle (Tick one) Car SUV 3-wheeler 2-wheeler Bus Truck Other (Specify)

2. Type of fuel (Tick one) Petrol Diesel CNG LPG Other (Specify)

3. Make 4. Model 5. Year of Mfg

6. Mileage Km/Iitre 7. Odometer Reading kilometers

8. Pollution Parameters Carbon Monoxide_____________________ Hydrocarbon ______________________________

(Attach PUC Copy)

Nitrogen Oxides _______________________ Particulate matter ___________________________

159
View publication stats