CSE1204 - Information Systems 1

Process Modelling
Data Flow Diagrams

Process Modelling

Process modelling aims to graphically

represent the processes which capture,
manipulate, store
and distribute data.

• data flow diagrams

• function decomposition
• structured English
• decision tables and decision
trees

Data flow Diagrams

Data flow diagrams model the flow of data into, through,

and out of an information system:

• show the processes that change or transform

data
• show the movement of data between
processes
• represent a system as a network of
processes which transform data flowing
between them

1
 Example DFD

DFDs fit neatly onto a single page

• the user is not overwhelmed by a DFD WAREHOUSE
invalid orders ORDERS
CUST OMERS
shipping details

order details
books
1. 2.
orders c ustomer name,
RECEIVE SHIP
c ustomer address
ORDER BOOKS

CUST OMERS
billing information

c ustomer name,
c ustomer address
books
INVOICES

3. invoic es, statements

COLLECT
c ustomer name, PAYMENT S
invoic e details
CUST OMERS
payments,
enquiries

Data Flow Diagrams

Data flow diagrams (or DFDs):

• a well-known process modelling technique

• easily understood
• a good communication tool
• model both manual and automated processes

Components of a DFD
2
1. Process
calculate
price

2. Data flow loan application

3. Data store Products

4. External agent Suppliers

(Source/Sink)

2
 1. Process

• transforms incoming data flows into outgoing data

flows:
the work performed on data which changes it

• each process has a unique number and name

calculate
price

1. Process
• name each process using a verb and a noun phrase

eg. calculate price

check customer details
accept supplier delivery

• the name of a process should describe what the

process does

• avoid vague names where possible

e.g. "process data"

2. Data Flow
• a data flow represents data in motion, moving from one
place in the system to another

• describes a "packet" of data or data that moves

together:
a data flow may consist of many individual, related
pieces of data that move together to a common
destination

loan application

3
• name each data flow using a noun or noun phrase
eg. customer order
• the name of a data flow should describe the contents of the
data "packet"
• the name should include as much information as possible
about the data flow eg."customer payment" rather than just "payment"

valid customer
order
customer order 2
validate
customer
order invalid
customer order

3. Data Store

• a data store represents a collection of data flows "at

rest"
• each data store has a unique name
• the name should describe the contents of the
data store
• a data store may represent many different types of
physical locations of data
• a data store may be a temporary or a permanent
repository of data

The Data Store

different notations

4
 data flows to and from a data store can remain unlabelled
if all attributes in the store are moving,
i.e. if an entire data packet (or packets) is going into
or out of the data store

sales order check

sales
order Sales orders
3 weekly sales
produce totals
weekly
sales
totals

4. External Agent (Source/Sink)

• an external agent represents an entity with which the system
communicates and which is outside the scope of the system
eg. an outside organisation or individual,
another department or another system,
a person or group within the department supported by the system who
interacts with the system
• an external agent is a source if it is an origin of data coming
into the system
• an external agent is a sink if it is a destination of data leaving
the system

Suppliers

4. External Agent (Source/Sink)

• data flows connecting the external agents to the

processes within the system represent the interface
between the system and its environment

• external agents are outside the system and define its

boundaries

• an external agent may be both a source and a sink

what a sink does with data it receives from the system and how a
source produces data which it inputs to the system are outside the
boundary of the system and are not shown on the data flow
diagram

5
 Example DFD
Sales Order sales order purchase order
System Suppliers

goods returned

supplier delivery
goods unavailable Inventory
notice System

shipping slip checked supplier

invoice
Accounts
Department
Warehouse

Example Data Flow Diagram

sales order
Customers
1
check
sales
order

Suppliers

Sales orders
2
produce purchase order
purchase
order

Guidelines for Drawing DFDs

• each object on a data flow diagram must have a unique name

• each process must have at least one data flow coming in

(input) and at least one data flow going out (output)

• the inputs to a process are different from the outputs of that

process

• a process must be able to build its outputs using only the

information in its input data flows plus any constant
information

6
 Guidelines for Drawing DFDs

data flows are permitted:

• between processes
• from a data store to a process
• from a process to a data store
• from a source to a process
• from a process to a sink

Guidelines for Drawing DFDs

data flows are NOT permitted:
• between external agents
• between data stores
• from an external agent to a data store

Identify six errors in this abstract DFD

Data Store 2
EE1

Data store 1
1
process2

EE2
EE3
2
Process1

7
 Levelling Data Flow Diagrams
Any "real" system is too large to represent as a single data
flow diagram

• the solution is to decompose the system into a hierarchy

of levels of processing

• the process model of the system then consists of a set of

levelled data flow diagrams

• levelling of DFDs improves their readability and

usefulness as a communication tool

Levelling of DFDs

Levelling creates a hierarchical decomposition

of the processing within the system

Context diagram
• Level 0 diagram
Level 1 diagrams
Level 2 diagrams
.
.
.
Level n diagrams

Levelling of DFDs

Context diagram

Level 0 diagram 1 2 3 4

Level 1 1.1 1.2 1.3 2.1 2.2 3.1 3.2 4.1 4.2 4.3
diagrams

Level 2 1.1.1 1.1.2 3.2.1 3.2.2

diagrams

8
 Context Diagram

• the highest level data flow diagram is the context diagram

• the context diagram shows the interaction of the system with its environment
in terms of data flows
• the context diagram defines the boundary of the system (the scope of the
system)
• only the data flows which leave the system and the data flows which come from
outside the system are shown

Context Diagram
• the entire system is represented as a single process

• all external agents (sources and sinks) are shown

• no data stores are shown: they are inside the boundary of the system

EASY GO
HOTEL

Example Context Diagram

Sales Order sales order purchase order

System Suppliers

goods returned

supplier delivery
goods unavailable Inventory
notice System

shipping slip checked supplier

invoice Accounts
Department
Warehouse

9
 Level Zero Diagram
• the level zero data flow diagram is the diagram at the level
immediately below the context diagram

• it "expands" the single process on the context diagram to show the

major, high-level processes (or functions) within the system

1.0 4.0

2.0

3.0

Level Zero Diagram

• all external agents (sources and sinks) are included
because the level zero diagram, like the context
diagram, represents the entire system

• the number of each process ends in .0 which

corresponds to the level of the diagram:
e.g. 1.0, 2.0, 3.0 etc.
often just 1, 2, 3 etc. are used

Example Level Zero Diagram

Order Entry purchase order

sales order Suppliers
System 2.0
Monitor
Stock
levels
1.0 3.0
supplier
Fill delivery
out of stock Accept
Sales Deliveries
notice Orders Stock file

shipping slip
Accounts
checked supplier Department
Warehouse invoice

10
 Level 1 and lower level diagrams

each Level 1 diagram and diagrams at lower levels (e.g. Level 2,

Level 3) show only a part of the processing

Level 1 processes are numbered 1.1, 1.2, 1.3, and 2.1, 2.2, 2.3 etc
that part is shown in more detail than on the Level zero diagram

no external agents are shown on a Level 1 or lower diagrams, as

the entire system is not being represented

Level 1 diagrams

a set of data flow diagrams is created at Level 1

there is one Level 1 diagram for each of the processes at Level zero
each diagram decomposes a Level zero process into several processes

2.4
2.3
2.1

2.5

2.2

Levelling DFDs

1
2

Context diagram 3

Level zero diagram

2.1
3.1

2.2
3.2

Further diagrams or Diagram 3 (level 1)

Diagram 3 (level 1)
process descriptions

11
 Example Level zero diagram

SUPPLIERS
CUSTOMERS
p.o.
Customer details number
Product p.o. invoice
details product product
product qty product delivered
holding INVENTORY order
qty
qty
1
Record product product holding 2
and order qty delivery
Invoice Maintain
Product qty
Sales
Inventory
customer SALES ORDERS product
invoice order qty

Example Level 1 diagram

Customer 1.2 INVENTORY

details Check
1.1 Stock Available
product
Check holding
Customer
Credit product
qty INVOICES
product
customer credit available qty
Product status SALES ORDERS
details
1.4
1.3
Create customer
product Sales
Record invoice
qty Sales Invoice
Demand

Guidelines for Levelling DFDs

• numbering:
when a process is decomposed, its diagram is given the same
number as that process
• balancing of DFDs:
all data flows entering and leaving a process must appear on the
corresponding diagram which decomposes that process
• external agents:
are only included on the two diagrams which represent the entire
system, i.e. the context and level zero diagrams

12
 Guidelines for Levelling DFDs
balancing data flows:
if data flows are decomposed at lower levels then the
contents of the data flows across levels must be balanced

sales order customer

shipping customer
details order items
1
1.1
out of stock Fill 1.2
notice Sales Check
Orders shipping Check
details stock
shipping slip available

sales order = customer + customer order

shipping details items

Guidelines for Levelling DFDs

The access to data stores across levels of diagrams must be consistent:

•the direction of accesses must match and all accesses on higher level
diagrams must appear on corresponding lower level diagrams
•a data store is first shown on the highest level diagram where it is accessed
by more than one process
•it can then appear on all lower level diagrams where it is accessed

1
1.1 1.2
1.3

Guidelines for Levelling DFDs

partition processes to:

• form cohesive, related groups of activities

• minimise the data flows between them

all parts of the system need not be decomposed to the same

level

13
 Guidelines for Levelling DFDs
How many levels should be in a set of DFDs?

• each diagram usually has between 3 and 7 processes

• level the diagrams until bottom level or primitive processes are
reached

primitive processes cannot be further decomposed as a data flow diagram

An example - Context Diagram

Applicant Line
Manager
Address
Applicant name Position Spec

Position

Acknow- Employment Successful

ledgement System Applicant

Payroll
Decision Position System

An example - Level zero diagram

Applicant
Applicant Line
Position name Manager
Address Ackn’d Position Spec
1 Appn
Accept Decision
Acknow- Applic- 2
ledgement ation
Evaluate
Applicants
Applicant
Position

Unsuccessfuls file
Evaluation
Payroll
results Successful System
Applicant

14
 Level 1- Diagram 1

Address
Applicant name

Position Verified Acknow-

Application ledgement
1.1

Verify for 1.2

Complete-
ness Acknow-
ledge
Application Acknowledged
Application

Level 1 - Diagram 2
Ackn’d
Appn Position Spec

Qualified
2.1
applicant
unqualified
Screen applicant Decision
Applicants
2.2 2.3

Reject
Schedule
Unsuitabl
Evaluation
es
Decision
Successful
Applicant
Unsuccessfuls file Evaluation
results

References

WHITTEN, J.L., BENTLEY, L.D. and DITTMAN, K.C. (2001) 5th ed., Systems Analysis
and Design Methods, Irwin/McGraw-HilI, New York, NY. Chapters 8

HOFFER, J.A., GEORGE, J.F. and VALACICH (2005) 4th ed., Modern Systems Analysis
and Design, Benjamin/Cummings, Massachusetts.
Chapter 7

15