1
INTRODUCTION TO
SYSTEM ANALYSIS AND
DESIGN
Plan
2
What Is An Information System?
The Systems Development Life Cycle
Software Process Models
System Analyst
Analysis Methodologies
What Is An Information
3
System?
An information system is a collection of
interrelated components that collect,
store, process, and provide as output the
information needed to complete a
business task.
Examples of Information
4
Systems
Course registration system
Online order system
Online banking system
Automated Information
5
Systems
Most people in an organization are
involved in an information system (IS), IS
can be manual or automated
(computerized), We may need to
transform manual to computerized or to
update a previous automated system.
For an organization to create an
effective automated IS requires big
effort.
Failure of IT Projects
6
According to industry surveys,
only about 30 % of all application software development
projects are considered successful
Close to 20 % are failures, i.e. they are cancelled prior to
completion or completed but never used.
The remaining 50 % are challenged – not delivered on time
Example: general motors did not make
effective IS so they lost 40 billion dollars.
This shows the necessity for thorough
planning—especially when an organization
is trying to implement a new kind of system.
Systems analysis and design reduces the
chances for such failures
The Systems Development Life
Cycle
SDLC provides overall framework for managing systems
7 development process.
Design and operation, security
Planning and
An idea Implementation and support
Analysis
Time
The Systems Development
8
Life Cycle
Two main approaches to SDLC
Predictive approach – assumes project can be
planned out in advance
Adaptive approach – more flexible, assumes
project cannot be planned out in advance
All projects use some variation of SDLC
Choosing the Predictive vs.
Adaptive Approach to the
SDLC
9
Systems Life Cycle
six-phase problem-solving procedure for
examining and improving an IS
information system.
The six phases make up the systems life
cycle.
Systems life cycle
The phases are as follows:
Preliminary investigation: The information problems or
needs are identified.
Systems analysis: The present system is studied in depth.
New requirements are specified.
Systems design: A new or alternative information system is
designed.
Systems development: New hardware and software are
acquired, developed, and tested.
Systems implementation: The new information system is
installed and adapted to the new system, and people are
trained to use it.
Systems maintenance: In this phase, the system is
periodically evaluated and updated as needed.
Phase 1: Preliminary Investigation
Preliminary investigation is to
determine the need for a new
IS.
It is usually requested by a
manager who wants something
done that is not presently being
done.
In Phase 1 the systems
analyst is concerned with
three tasks:
Defining the Problem
Suggesting Alternative Systems
Preparing a Short Report
Phase 1: Preliminary Investigation
Defining the Problem: Determining what
information is needed, by whom, when, and
why, is accomplished by interviewing and
making observations.
For example, suppose AA account executives and
graphic designers currently just record the time
spent on different jobs on their calendars.
Phase 1: Preliminary Investigation
Suggesting Alternative Systems: This step is
simply to suggest some possible plans as
alternatives to the present arrangement. For
instance:
AA could hire more secretaries to collect the
information from everyone’s calendars.
Develop a new software that permits to automate
projects’ timing.
Or it could use the existing system of network-
linked microcomputers that stable workers and on-
call workers presently use.
Perhaps, you think, there is already some off-the-
shelf packaged software available that could be
used for a time-and-billing system.
Phase 1: Preliminary Investigation
Preparing a Short Report: The systems
analyst writes a short report summarizing
the results of the preliminary investigation
and suggesting alternative systems.
This document is presented to higher
management, along with a recommendation to
continue or discontinue the project.
For AA, your report might point out that billing is
frequently delayed. It could say that some tasks
may not get charged at all. Thus, as the analyst
has noted, you suggest the project might pay for
itself merely by eliminating lost or forgotten
charges.
Phase 2: Systems Analysis
Data is collected about the
present system.
This data is then analyzed,
and new requirements are
determined.
We are not concerned with a
new design here, only with
determining the requirements
for a new system.
This phase is concerned with
gathering and analyzing data.
Phase 2: Systems Analysis
Gathering Data: The systems analyst expands
on the data gathered during Phase1.
He adds details about how the current system works.
Data is obtained from observation and interviews.
It is also obtained from studying documents and
standard operating procedures.
One task is the organization chart, which shows
levels of management. The levels are top managers,
middle managers, and supervisors.
In addition, data may be obtained from
questionnaires given to people using the system.
Phase 2: Systems Analysis
Analyzing the Data: The idea is to learn
how information currently flows and to
point out incorrectness
Many different tools are available to assist
systems analysts in the analysis phase.
Some of the principal ones are: Checklists, Top-
down analysis methodology, Grid charts,
System flowcharts, Data flow diagrams.
Phase 2: Systems Analysis
Documenting the Systems Analysis Stage:
This stage is typically documented in a
report for higher management.
The systems analysis report describes the
current information system, the requirements
for a new system, and a possible development
schedule.
For example, at AA, the system flowcharts show the present
flow of information in a manual time-and-billing system.
Some boxes in the system flowchart might be replaced with
symbols showing where a computerized information system
could work better.
Management studies the report and decides
whether to continue with the project.
Phase 3: Systems Design
Systems design
consists of three
tasks:
Designing Alternative
Systems
Selecting the Best
System
Writing the Systems
Design Report
Phase 3: Systems Design
Designing Alternative Systems: In almost
all instances, more than one design can be
developed to meet the information needs.
Systems designers evaluate each
alternative system for feasibility:
Economic feasibility: Will the costs of the new
system be justified by the benefits it promises?
Technical feasibility: Are hardware, software,
and training available to make the system work?
Operational feasibility: Can the system actually
be made to operate in the organization, or will
people (employees, managers, clients) resist it?
Phase 3: Systems Design
Selecting the Best System: When choosing
the best design, managers must consider
these four questions:
Will the system fit in with the organization’s
overall information system?
Will the system be flexible enough so it can be
modified in the future?
Can it be made secure against unauthorized
use?
Are the benefits worth the costs?
Phase 3: Systems Design
Writing the Systems Design Report: The
systems design report is prepared for
higher management and describes the
alternative designs.
It presents the costs versus the benefits and
outlines the effect of alternative designs on the
organization.
It usually concludes by recommending one of
the alternatives.
Phase 4: Systems
Development
Systems Development
consists of three tasks:
Developing Software
Acquiring Hardware
Testing the New System
Phase 4: Systems
Development
Developing Software: Application software
for the new information system can be
obtained in two ways.
It can be purchased as off-the-shelf packaged
software and possibly modified,
or it can be custom designed. If any of the
software is being specially created, the steps of
programming should be followed.
Phase 4: Systems
Development
Acquiring Hardware: Some new systems
may not require new computer
equipment, but others will.
The kinds needed and the places they are to be
installed must be determined.
Switching or upgrading equipment can be a
tremendously expensive proposition.
Will a microcomputer system be sufficient as a
company grows?
Are networks expandable?
Will people have to undergo costly training?
Phase 4: Systems
Development
Testing the New System: After the
software and equipment have been
installed, the system should be tested.
Sample data is fed into the system. The
processed information is then evaluated to see
whether results are correct.
Testing may take several months if the new
system is complex.
Phase 5: Systems
Implementation
Another name for
Phase 5, systems
implementation, is
conversion.
It is the process of
changing (converting)
from the old system to
the new.
Phase 5: Systems
Implementation
Types of Conversion
Direct approach - the conversion is done simply
by abandoning the old and starting up the new.
This can be risky. If anything is still wrong with the new system, the
old system is no longer available to fall back on.
The direct approach is not recommended precisely because it is so
risky. Problems, big or small, will appear in a new system. In a large
system, a problem might just mean catastrophe.
Parallel approach - old and new systems are
operated side by side until the new one has
shown it is reliable.
This approach is low-risk. If the new system fails, the organization can
just switch to the old system to keep going.
However, keeping enough equipment and people active to manage
two systems at the same time can be very expensive. Thus, the
parallel approach is used only in cases in which the cost of failure or
of interrupted operation is great.
Phase 5: Systems
Implementation
Pilot approach - the new system is tried out in
only one part of the organization.
Once the system is working smoothly in that part, it is implemented
throughout the rest of the organization.
The pilot approach is certainly less expensive than the parallel
approach. It also is somewhat riskier.
However, the risks can be controlled because problems will be faced
in only certain areas of the organization. Difficulties will not affect the
entire organization.
Phased approach - the new system is
implemented gradually over a period of time.
The entire implementation process is broken down into parts or
phases. Implementation begins with the first phase and once it is
successfully implemented, the second phase begins. This process
continues until all phases are operating smoothly.
This is an expensive proposition, because the implementation is done
slowly. However, it is certainly one of the least risky approaches.
Phase 5: Systems
Implementation
In general, the pilot and phased approaches
are the most favored methods.
Pilot is preferred when there are many people in an organization
performing similar operations, for instance, all salesclerks in a
department store.
Phased is more appropriate for organizations in which people are
performing different operations.
The systems analyst succeeds in convincing
the top managers of AA to take a pilot
approach. The reason is that it is easy to select
one trial group.
Thus, the new time-and-billing system is tried first with a group of
people in a particular department.
Phase 5: Systems
Implementation
Training: Training people is important, of course.
Unfortunately, it is one of the most commonly
overlooked activities.
Some people may begin training early, even before the
equipment is delivered, so that they can adjust more
easily.
In some cases, a professional software trainer may be
brought in to show people how to operate the system.
However, some of the developed software may be
simple enough that the systems analyst can act as the
trainer.
Phase 6: Systems
Maintenance
After implementation
comes systems
maintenance, the last
step in the systems life
cycle.
This phase is a very
important and an ongoing
activity.
Most organizations spend
more time and money on
this phase than any other
Phase 6: Systems
Maintenance
Systems audit: In the systems audit, the
system’s performance is compared to
the original design specifications.
This is to determine if the new procedures
are actually increasing productivity. If they
are not, some further redesign may be
necessary.
Phase 6: Systems
Maintenance
Periodic evaluation: After the systems
audit, the new information system is
periodically evaluated and further
modified, if necessary.
All systems should be evaluated from time
to time to see if they are meeting the goals
and providing the service they are
supposed to.
Software Process Models
A variety of different life cycle models (Software
Process Models) are proposed:
Waterfall model: Activities are represented in separated
phases.
A document is delivered after each phase.
V model: Variant of waterfall model that makes the link
between the earlier and last phases.
Prototyping model: Evolving a prototype to have the
final system.
Spiral model: takes into account the risk aspect. After a
phase, we can go back to the previous one.
Waterfall Model
37
Waterfall model is an SDLC approach
that assumes the various phases of a
project can be completed sequentially –
one phase leads into the next phase.
Linear model (waterfall)
Requirement Analysis
System Analysis
Design
Coding and unit
• Meant for large, integrated systems
Testing
with diverse and dispersed
development Integration and
– Still popular (in modified forms) system testing
among big developers where
change is not acceptable
– Some call it a complete disaster Delivery
• Work divided into phases
– Each phase ends with a document as an output which describes the Maintenance
conclusions
– Each phase starts with a check of the previous step’s output – that
becomes the input for the next phase
V Model
Improvement of the waterfall model
Highlights the symmetry and the relations that there are
between the phases of the beginning of the life cycle and those
of end
Testing of the product is planned in parallel with a
corresponding phase of development
The phases of the beginning must be accompanied by a
planning of the phases of end
The activities of construction precede the activities of validation
and verification
But acceptance must be prepared in the beginning
V Model
+preparing the …
Maintenance
validation
Requirement Analysis Acceptance Testing
System Analysis validation System Testing
Global Design verification Integration Testing
Detailed Design verification Unit Testing
Coding
Prototyping Model
Consists in defining very summarily and quickly the needs
and to carry out a preliminary version of the system
Listen to customer
Effective if:
The customer does not know what he wants
We do not know what he wants Customer test
Build
Characteristics
The goal is to ensure the feasibility and to verify the requirements
The product is rebuilt by taking account of the feedback of the user
Several prototypes are developed (minimum of functionalities)
Only the prototype retained by the user will be evolved
Spiral Model
Generally conceived for the systems at risks
Combine prototyping and waterfall model to
develop and validate representative
prototype/components of the software
The accent is put on the evaluation of the risks
The model proceeds according to :
A succession of development cycles of 4 phases
The number of cycles is not known
42
Spiral Model
• At each increment, perform a
risk analysis
– What could go wrong?
– What can we do to solve it?
– Is it worth continuing?
43
What Is System Analysis
44
About?
Understanding the goals and strategies
of the business.
Defining the information requirements
that support those goals and strategies.
It is not about programming.
System Analysis Vs. System
45
Design
Systems analysis is a process of
understanding in detail what a system
should accomplish.
Systems design is a process of
specifying in detail how components of
an information system should be
physically implemented.
System Analyst
46
A business professional who uses
analysis and design techniques to solve
business problems using information
technology.
The Role of a System
47
Analyst
Investigate, analyze, design, develop,
installs, evaluate, and maintains a
company’s information systems.
A business problem solver
The solution to the “problem” is
generally a new information system
Help translate business requirements
into IT projects.
Typical Problems System
48
Analyst Typically Solve
Customers want to order product
anytime of the day.
Suppliers want to minimize inventory
holding costs.
Marketing wants to know customer
needs better.
Management wants to analyze financial
information more efficiently.
The Analyst as a Business
49
Problem Solver
Has computer technology knowledge and
programming expertise
Understands business problems
Uses logical methods for solving problems
Has fundamental curiosity
Wants to make things better
Is more of a business problem solver than
a technical programmer
What is your problem-solving
50
approach?
Assume that you are deciding whether to
buy a new laptop to replace the old one.
Analyst’s Approach to Problem
51
Solving
1. Research and understand the problem
2. Verify if the benefits of solving the problem
outweigh the costs
3. Define the requirements for solving the
problem
4. Develop a set of possible solutions
(alternatives)
5. Decide which solution is best and recommend
6. Define the details of the chosen solution
7. Implement the solution
8. Monitor to ensure desired results
Required Skills of the Systems
52
Analyst
1. Technical
2. Business
3. People
1-Technical Knowledge and
53
Skills
Computers / peripheral devices
(hardware)
Communication networks and connectivity
Database and database management
systems (DBMS)
Programming languages (for example,
VB.NET or Java)
Operating systems and utilities
1- Technical Knowledge and
54
Skills
Techniques:
Project planning, systems analysis
Systems design, database design,
network design
Construction, implementation, systems
support
2- Business Knowledge and
55
Skills
Business functions performed by
organization
Strategies, plans, traditions, and values of
the organization
Organizational structure
Organization management techniques
Functional work processes
3- People Knowledge and
56
Skills
Systems analysts need to understand how
people
Think
Learn
React to change
Communicate
Work (in a variety of jobs and levels)
3- People Knowledge and Skills
57
(continued)
Interpersonal and communication skills
are crucial to
Obtaining information
Motivating people
Getting cooperation
Understanding the complexity and workings of
an organization in order to provide necessary
support
Hard and Soft Skills
58
Hard skills are the skills in technical
areas such as database design,
programming, and telecommunication.
Soft skills are the skills in nontechnical
areas such as interviewing, team
management, and leadership.
Analysis Methodologies
59
Méthode d'Etude et de Réalisation
Informatique pour les Systèmes
d'Entreprise (MERISE)
Entity-Relationship Model
Unified Modeling Language (UML)
MERISE
60
Méthode d'Etude et de Réalisation
Informatique pour les Systèmes
d'Entreprise (MERISE) is a French
methodology developed and used by the
French government and thus by
universities.
MERISE considers three levels of
abstractions
Abstraction Data for both data and (Processing)
Treatment processing
level
from more abstract towards more concrete.
Conceptual Conceptual data model Conceptual treatment model
(CDM) (CTM)
Organizatio Logical data model Organizational treatment
nal (LDM) model (OTM)
ER DIAGRAM NOTATION
Symbol Meaning
ENTITY TYPE
WEAK ENTITY TYPE
RELATIONSHIP TYPE
IDENTIFYING RELATIONSHIP TYPE
ATTRIBUTE
KEY ATTRIBUTE
MULTIVALUED ATTRIBUTE
COMPOSITE ATTRIBUTE
DERIVED ATTRIBUTE
E1 R E2 TOTAL PARTICIPATION OF E2 IN R
E1 N
R E2
CARDINALITY RATIO 1:N FOR E1:E2 IN R
(min,max)
R E STRUCTURAL CONSTRAINT (min, max) ON
PARTICIPATION OF E IN R
UML
62
Unified Modeling Language (UML) is a collection of
diagrams that reflect the static structure and the dynamic
behavior of the modeled system:
Use case diagrams are designed for each user
Sequence diagram: shows the interaction between the user and the
interface
Component diagram that represents the decomposition of the entire
system into smaller components (subsystems)
Class diagram: corresponds to ERD or CDM
Object diagram: shows instances of classes
State charts
Deployment diagram
Flow charts
UML is a very prospering methodology that is being used
more and more every year (Rational Rose is a UML CAD
