Data Flow Diagramming

WMSU
Data Flow Diagramming
• Some AIS professionals prefer one documentation method over another, but many use both
flowcharts and data flow diagrams depending upon the situation and tasks at hand.

• According to the research (Bradford, Richtermeyer, and Roberts (2007)), over 20 percent
reported using data flow diagrams to document information systems.

– used by managers for describing business processes, evaluate the current system, design
or change a system, and assess the internal control environment.
Data Flow Diagramming
What is data flow diagram?

A structured analysis technique that employs a set of visual representations of the data that

moves through the organization, the paths through which the data moves, and the processes

that produce, use, and transform data.

Data Flow Diagramming
Why use data flow diagram?

• Represent information flowing between objects in the data-flow diagram.

• Model the flow of information into the system, out of the system, and between elements within

the system.

• Usually illustrate information flows between two external entities (outside of the system

boundaries)
DFD Symbols and Design Consideration
• A process is any set of procedures an
organization uses to gather data, change
the data into information, or report the
information to system users.
• Characteristics: a number and a name

• Process names should be verb phrase.

“action word”. E.g. calculate, record, check,
approve.
DFD Symbols and Design Consideration
• An external entity is any person or
organization outside the boundary of an
information system.

• For example, if you were constructing a data

flow diagram for an organization’s cash
management system, external entities might
include customers, suppliers, or the bank.

– whether or not a person or an

organization is an external entity depends
upon your definition of the system, not the
organization.
DFD Symbols and Design Consideration
• A data store is a place for collecting data;
you might think of it as a “file,” whether
paper-based or electronic.
• Labeled with noun phrases such as
customer data, vendor data, or inventory
data.
• Data stores can be linked to processes or
external entities in a data flow diagram; they
cannot be linked to one another.
DFD Symbols and Design Consideration
• A data flow is represented by a directional
line in a data flow diagram. Data flows
should have only one arrow on one end to
conform to DFD design conventions.
• Labeled with noun phrases: desired
information, accounts payable data,
customer order data, and the like. Do not
label with verb phrase.

• Referring to the data content.

DFD Symbols and Design Consideration
Hoffner, George, and Valacich (1996) suggested the following rules/conventions associated
with good data flow diagrams:

1. All processes should have unique names. If two data flow lines (or data stores) have the
same label, they should both refer to the exact same data flow (or data store).

2. The inputs to a process should differ from the outputs to a process.

3. Any single DFD should not have more than about seven processes.

4. No process can have only outputs. If an object has only outputs, then it must be an external
entity

5. No process can have only inputs. If an object has only inputs, it must be a data store.
DFD Symbols and Design Consideration
6. A process has a verb phrase label.
7. Data cannot be moved directly from one data store to another data store. Data must be
moved by a process.

8. Data cannot move directly from an external entity to a data store. Data must be moved by a
process that receives data from the entity and places the data into the data store.

9. Data cannot move directly to an external entity from a data store. Data must be moved by a
process.

10. A data store has a noun phrase label.

11. Data of any concern to the system cannot move directly from one external entity to another
external entity. They must be moved by a process.
DFD Symbols and Design Consideration
12. An external entity has a noun phrase label.
13. A data flow has only one direction between symbols.

14. A data flow cannot go directly back to the same process it leaves.

15. A data flow can go directly into a data store. When it does, it signifies an update (delete,
add, or change). Likewise, a data flow can come directly from a data store; in that context, it
refers to a retrieval or use of the data in the store.

16. A data flow has a noun phrase label

Different Levels of DFD
• Context Diagram show how the process, represented as a single circle, relates to the
external entities, represented with rectangles.
Different Levels of DFD
• Level One diagram present a more detailed view of the system (process) than context
diagrams, by showing the main sub-processes and stores of data that make up the system
(process) as a whole.

Customer Warehouse

Status 1.0
Message Check Shipping
Status order
2.0 3.0
Issue Generate
Order
Status Shipping
data
Message Order
Pending Orders
Different Levels of DFD
• N-level diagram - a major advantage of the data-flow modelling technique is that, through a
technique called “levelling”, the detailed complexity of real world systems can be managed
and modeled in a hierarchy of abstractions. Certain elements of any dataflow diagram can be
decomposed (“exploded”) into a more detailed model a level lower in the hierarchy.
Different Levels of DFD
Level Name Number of DFDs in a Leveled Set Numbering Format
Context One 0
One One 1.0, 2.0, …
Two More than one Process 1.0 is subdivided into 1.1, 1.2, …, 1.n
Process 2.0 is subdivided into 2.1, 2.2, …, 2.n
Three More than one Process 1.1 is subdivided into 1.1.1, 1.1.2, …,
1.1.n
Process 1.2 is subdivided into 1.2.1, 1.2.2, …,
1.2.n
Different Levels of DFD
Here are some questions my students often raise when learning how to draw data flow
diagrams:
1. How many processes should a single DFD have? As you might suspect, this question is
open to interpretation. However, most systems analysts would limit the number of
processes to seven, as suggested by the third of Hoffner’s principles.
2. How many levels should be in a leveled set of DFDs? Like the previous question, this one is
nondeterministic.
3. When do you stop breaking down a process? You stop breaking down (decomposing/
exploding) a process when it is primitive. In other words, when the process label is fairly
self-explanatory.
4. Does each process have to be decomposed to the same level? No.
DFD Example
Netflix has grown into one of the most used video streaming services in the world. The company
was one of the first to see the potential of video streaming technology and began to transition to
a subscription video-on-demand model in 2007. - Netflix – statistics & facts | Statista

Accordingly it “is the world’s leading Internet television network with over 44 million members in
41 countries.” New clients create an individual account, then add movies and television shows
(programming) to their queue (list) for later viewing.
DFD Example
• Let’s create a DFD of the Account and List Creation Process
• Before creating the DFD:

– boundaries should be established

– understanding the boundaries means you have a clear picture of the external entities
involved in the process
DFD Example
DFD Example
End of Presentation