Giới thiệu phân tích và thiết kế hướng đối tượng

Software systems for enterprises need to be designed after
an in-depth analysis. As software systems become more
and more complex, it becomes increasingly difficult to
correctly analyze the requirements and convert them into
an appropriate design. Therefore, tools are required to deal
with this complexity. Object-Oriented Analysis and
Design (OOAD) provide you the necessary tools to
analyze and design complex software systems.
This chapter introduces the various phases of the software
development life cycle (SDLC) and discusses the various
software development approaches. Then, it discusses the
object-oriented concepts and explains the role of OOAD
in software development.

Objectives
In this chapter, you will learn to:
 Identify the various software development
approaches
 Identify the role of OOAD in SDLC

Chapter 1

Introducing
Object-Oriented
Analysis and Design



Understanding Software Development Life Cycle
(SDLC)
Before the acceptance of software development as an engineering stream, the process of

developing software was an ad hoc activity with no formal rules or standards. As a result,
software projects faced serious problems in terms of schedule slippage, cost overrun, and
inferior quality of software.
SDLC was introduced to address the problems faced during the software development
process. SDLC is a disciplined and systematic approach that divides the software
development process into various phases, such as requirement analysis, design, and
coding. The phase-wise development process helps you track schedule, cost, and quality
of the software projects.

Overview of SDLC Phases
There are six phases in SDLC:
Feasibility analysis: Includes analysis of project requirements in terms of input data
and desired output, processing required to transform input into output, cost-benefit
analysis, and schedule of the project. The feasibility analysis also includes the
technical feasibility of a project in terms of available software tools, hardware, and
skilled software professionals. At the end of this phase, a feasibility report for the
entire project is created.
„
Requirement analysis and specification: Includes gathering, analyzing, validating,
and specifying requirements. At the end of this phase, the Software Requirement
Specification (SRS) document is prepared. SRS is a formal document that acts as a
written agreement between the development team and the customer. SRS acts as
input to the design phase and includes functional, performance, software, hardware,
and network requirements of the project.
„
Design: Includes translation of the requirements specified in the SRS into a logical
structure that can be implemented in a programming language. The output of the
design phase is a design document that acts as an input for all the subsequent SDLC
phases.
„

Coding: Includes implementation of the design specified in the design document into
executable programming language code. The output of the coding phase is the source
code for the software that acts as input to the testing and maintenance phase.
„
Testing: Includes detection of errors in the software. The testing process starts with a
test plan that recognizes test-related activities, such as test case generation, testing
criteria, and resource allocation for testing. The code is tested and mapped against the
design document created in the design phase. The output of the testing phase is a test
report containing errors that occurred while testing the application.
„

©NIIT

Introducing Object-Oriented Analysis and Design 1.3


„

Maintenance: Includes implementation of changes that the software might undergo
over a period of time, or implementation of new requirements after the software is
deployed at the customer location. The maintenance phase also includes handling the
residual errors that may exist in the software even after the testing phase.

Software Development Approaches
Different types of projects have different requirements. Therefore, it may be required to
tailor the SDLC phases according to the specific needs of the project. This gives rise to
various software development approaches. Some of these approaches are:
„
Waterfall approach
„

Prototyping approach
„
Spiral approach
„
Win-win spiral approach
„
Incremental approach

Waterfall Approach
The waterfall approach describes the software development process in a linear sequential
flow. This means that any phase in the development process begins only if the previous
phase is complete. The waterfall approach does not define the process to go back to the
previous phase to handle changes in requirement. Therefore, different projects may follow
different approaches to handle such situations.
The waterfall approach is the earliest approach that was used for software development.
Initially, most projects followed the waterfall approach because they did not focus on
changing requirements.
The waterfall approach defines the software development process in seven phases:
„
Conception: Triggers when a problem is perceived. This phase involves identifying
goals to be achieved after the problem is solved, estimating benefits in the new
system over the current system, and identifying other areas that are affected by the
solution. This phase also involves developing the business case for the project. A
business case provides the information that a manager needs to decide whether to
support a proposed project, before resources are committed to its development.
„
Initiation: Involves a macro level study of the customer requirements. This phase
also involves defining alternative solutions to the customer requirements and
cost-benefit justification of these alternatives.
„

Analysis: Involves carrying out a detailed study of the customer requirements and
arriving at the exact requirements of the proposed system. The phase involves
freezing the requirements before the design phase begins.
1.4 Introducing Object-Oriented Analysis and Design

©NIIT


„
„
„
„

Design: Involves translating the identified requirements into a logical structure,
called design that can be implemented in a programming language.
Construction: Involves converting the design into a machine-readable form.
Integration and testing: Involves integrating and testing all the modules developed
in the previous phase as a complete system.
Implementation and maintenance: Involves converting the new system design into
operation. This may involve implementing the software system and training the
operating staff before the software system is functional.

The following figure shows the various phases in the waterfall approach.
Conception

Initiation

Analysis

Design

Construction

Integration and
Testing

Implementation
and
Maintenance

Waterfall Approach

The waterfall approach assumes that requirements are stable and frozen across the project
plan. However, this is usually not true in case of large projects where requirements may
evolve across the development process.

©NIIT

Introducing Object-Oriented Analysis and Design 1.5


Prototyping Approach
The prototyping approach, also known as evolutionary approach, came into picture
because of failures that occurred in the final version of the software application developed
using waterfall approach. The failures generally occurred because of the change in
requirements of the proposed system or because of the gap in understanding the customer
requirements by the development team. A gap in the first version of a developed
application, inevitably leads to the need for redoing the application. To overcome these
limitations, the concept of prototyping was introduced.
A prototype is a sample implementation of the system that shows limited and main

functional capabilities of the proposed system. After a prototype is built, it is delivered to
the customer for evaluation. The prototype helps the customer determine how the features
will function in the final software. The customer provides suggestions and improvements
on the prototype. The development team implements the suggestions in the new
prototype, which is again evaluated by the customer. The process continues until the
customer and the development team understands the exact requirements of the proposed
system. When the final prototype is developed, the requirements are considered to be
frozen.
The prototyping approach is used in the requirements gathering and analysis phase to
capture the exact requirements of the proposed system. After the requirements are frozen,
the remaining phases of the development process need to be executed to complete the
development of the software system.
An e-commerce Web site, such as a shopping site, is an example where you can
implement the prototyping approach. You can develop a prototype of the various Web
pages of the shopping site, such as catalogue page, product order page, and credit card
authentication page, and present it to the customer for approval. If the customer approves
the prototype of the site, requirements are stated again and the design of the Web site is
initiated. If the customer does not approve the Web site, the development team revisits the
prototype and resubmits it to the customer for approval. This process continues until the
prototype is approved.
Prototypes are of two types:
„
Throwaway prototypes: Prototypes that are eventually discarded rather than
becoming a part of the finally delivered software. Examples of throwaway prototypes
include screen mock-ups and storyboards.
„
Evolutionary prototypes: Prototypes that evolve into the final system through
iterative incorporation of user feedback.

1.6 Introducing Object-Oriented Analysis and Design

©NIIT


Although prototyping is a very useful technique to obtain accurate requirements of the
system and to speed up the development process, it has some disadvantages associated
with it. Some disadvantages of prototyping are:
„
It gives clients the false impression that a few minor changes to the prototype will
give them the required system.
„
It may compromise on the overall quality of the software in the rush to develop the
prototype. For example, the developer may use inefficient algorithms or
inappropriate programming languages for developing the prototype quickly and the
same may find place in the final application thus leading to inefficient code running
in the final application.

Spiral Approach
The spiral approach includes the iterative nature of the prototyping approach and the
linear nature of the waterfall approach. This approach is ideal for developing software that
are released in various versions.
In each iteration of the spiral approach, software development process follows the
phase-wise linear approach. At the end of first iteration, the customer evaluates the
software and provides the feedback. Based on the feedback, software development
process enters into the next iteration and subsequently follows the linear approach to
implement the feedback suggested by the customer. The process of iteration continues
throughout the life of the software.
An example of a spiral approach is the evolution of Windows operating system from
Window 3.1 to Windows 2003.You may refer to Windows 3.1 operating system as the
first iteration in the spiral approach. The product was released and evaluated by the

customers, which include the market at large. After getting the feedback from customers
about Windows 3.1, Microsoft planned to develop a new version of Windows operating
system. Windows 95 was released with enhancements and graphical flexibility. Similarly,
other versions of Windows operating system were released.
The spiral approach includes six phases:
„
Customer communication: Includes understanding the system requirements by
continuous communication between the customer and the system analyst.
„
Planning: Includes estimating schedule, cost, and resource for the iteration.
„
Risk analysis: Includes identifying, estimating, and monitoring technical and
management risks, such as schedule slippage and cost overrun.
„
Engineering: Includes requirement gathering and design of the software system.
„
Construction and release: Includes coding, testing, and deploying software at the
customer site and providing user-support documents.
„
Customer evaluation: Includes evaluation of software by the customer and
implementing feedback in the next iteration of the software development.
©NIIT

Introducing Object-Oriented Analysis and Design 1.7


The following figure shows the various phases in the spiral approach.

Spiral Approach

Win-Win Spiral Approach
The win-win spiral approach is an extension of the spiral approach. The phases in this
approach are same as the phases in the spiral approach. The only difference is that at the
time of identifying the requirements, the development team and the customer hold
discussions and negotiate on the requirements that need to be included in the current
iteration of the software.
The approach is called win-win because it is a winning situation for the development team
and also for the customer. The customer wins by getting the product that fulfills most of
the requirements while the development team wins by delivering software which is
developed with all the requirements established after negotiations with the customer. The
win-win spiral approach is generally used when you have time-bound releases.

1.8 Introducing Object-Oriented Analysis and Design

©NIIT


Incremental Approach
In an incremental approach, software requirements are broken down into various
functional units. These functional units contain a group of similar tasks. A project activity
list is created to prioritize the functional units with various activities to be performed in
each functional unit. Each functional unit is implemented in an increment and the final
product is achieved after all the functional units are implemented in the development
process.
Each increment includes three phases: design, implementation, and analysis. During the
design phase of the first increment, the functionality with topmost priority from the
project activity list is selected and the design is prepared. In the implementation phase, the
design is implemented and tested. In the analysis phase, the functional capability of the
partially developed product is analyzed. The development process is repeated until all the
functions of the project are implemented. The following figure shows the incremental

approach with various phases.

Design

Design

Design

Implementation

Implementation

Implementation

Analysis

Analysis

Analysis

Increment 1

Increment 2

Increment n

Incremental Approach

Consider an example where a bank wants to develop software to automate the banking
process for insurance services, personal banking, and home and automobile loans. The

bank wants the automation of personal banking system immediately because it will
enhance the customer services. You can implement the incremental approach to develop
the banking software. In the first increment, you can implement the personal banking
feature and deliver it to the customer. In the later increments, you can implement the
insurance services, home loans, and automobile loans features of the bank.
The limitation of the incremental approach is that it is applicable only to large
applications. This is because it is difficult to break small projects into smaller functional
increments.

©NIIT

Introducing Object-Oriented Analysis and Design 1.9


Just a minute:
Which of the following approaches includes the iterative nature of the prototyping
approach and the linear nature of the waterfall approach?
1.

Incremental approach

2.

Spiral approach

3.

Evolutionary approach

Answer:

2. Spiral approach

1.10 Introducing Object-Oriented Analysis and Design

©NIIT


Understanding Object-Oriented Concepts
In the design phase of any software development approach, you can follow two
approaches to software development. These are the function-oriented and object-oriented
development approaches. The function-oriented approach is the erstwhile approach that
evolved before the object-oriented approach.
The function-oriented approach is module-centric and concentrates on functions of the
software. This approach does not restrict the creation of modules that are tightly coupled
and expose data to other parts of the program. Therefore, software constructed using the
function-oriented approach may be so tightly coupled and interdependent that it may be
difficult to separate the construction of one module from another. In addition, data may be
visible and accessible throughout the program. Reusability of the software in this case
becomes difficult.
The limitations of the function-oriented approach gave way to the evolution of the
object-oriented approach. The fundamental idea behind object orientation is to portray
things as they exist in the real world. This helps to reduce the complexity of software
systems. OO approach introduces the concept of inheritance which allows reuse of
existing code components. In addition, OO approach supports encapsulation of data,
abstraction, and polymorphism. The OO approach guides you to think in terms of
reusability and therefore, helps in creating a better design.

Overview of OO Concepts
The various OO concepts are:
„

Classes and objects: A class is an abstract data type that contains a set of attributes
and functions. An object is an instance of a class. All the objects share the same copy
of the member functions, but maintain a separate copy of the member data. To
understand classes and objects, consider the analogy of a car. Cars such as BMW,
Ferrari, and Chevrolet share certain similar features and, therefore, can be grouped to
form the objects of the class, Cars.

©NIIT

Introducing Object-Oriented Analysis and Design 1.11


The following figure depicts a class and its objects.
Class
member data1
member data2
member function1
member function2

Object

Object

member data1
member data2

member data1
member data2

A Class and its Objects

„

Inheritance: Inheritance refers to sharing of attributes and behaviors among classes
based upon hierarchical relationship. The inheritance feature allows you to define a
new class by extending an existing class. The original class is called a base class or
super class and the new class obtained through inheritance is called a derived class or
a sub class. A base class is a generalization of the derived class, which means that the
base class contains only those properties that are common to all the derived classes.
Consider an example. Motorcycles, cars, and trucks have certain common properties
– all have wheels, engines, and horns. Therefore, they can be grouped under a class
called automobiles. Apart from sharing these common features, each subclass has its
own particular characteristics, such as trucks run on diesel while cars may run on
diesel or petrol.

1.12 Introducing Object-Oriented Analysis and Design

©NIIT


The following figure depicts inheritance of a derived class from the base class.
Automobile Class

Wheels
Engines
Horn

Car Class

Model
Wheels

Engines

Defined in derived
class
Accessible by
derived class

Horn

Inheritance of a Derived Class from a Base Class

Note
Inheritance of classes allows you to reuse code. If certain data or methods are similar
in a set of classes, then instead of defining those methods and data in each class
separately, you can define the common methods and data in the base class. The derived
class can then simply inherit the common methods and attributes from the base class.

„

©NIIT

Abstraction: Abstraction focuses on essential, inherent aspects of an entity ignoring
its implementation details. An abstract class specifies the abstract operation but does
not include its implementation. The implementation is provided in the derived classes
of the abstract class. Consider an example of the Polygon class that specifies an
abstract operation, ComputeArea(). The Polygon class does not provide the
implementation of the operation, ComputeArea(). The derived classes of the Polygon
class, Rectangle and Square, implement the ComputeArea() operation to compute the
area of a rectangle and square.

Introducing Object-Oriented Analysis and Design 1.13


The following figure shows the Polygon class with its derived classes.
Polygon
compute_area();
//declaration

Rectangle
compute_area();
//implementation

Circle
compute_area();
//implementation

Abstraction among Classes
„

„

Encapsulation: Encapsulation means preventing access to non-essential details. For
example, when you plug in the cord of the vacuum cleaner and switch it on, the
vacuum cleaner starts smoothly. You cannot see the complex processes needed to
actually convert electricity into suction power. In other words, the exact working of
the vacuum cleaner has been encapsulated. Encapsulation is also called information
hiding because it involves hiding attributes and operations of a class from other
classes.
Polymorphism: The concept of using operators or functions in different ways
depending on what they are operating on is called polymorphism. Polymorphism in

the context of operators refers to the difference in the functionality of an operator
depending upon the type of the operands. When polymorphism is used in the context
of functions performing different tasks, it is referred to as function overloading. An
example of function overloading is an add() function, which can take different
signatures at run time. For instance, the function can be overloaded as add(int a, int
b) or add(float a, float b) depending upon the type of parameters passed at run time.

Consider the example of Countryside Markets Inc., which has no formal system to store
its employee’s information. The organization now wants to store detailed information
about its employees, such as name, age, date of birth, e-mail id, department, employee
code, salary drawn, and date of joining.

1.14 Introducing Object-Oriented Analysis and Design

©NIIT


The following figure shows six classes for the Countryside Markets employee information
system.
Employee
empID
empName
empAddress
empDateOfBirth
empContactNos
mgrID
empDesignation
empJoinDate

Attributes with

Protected Access

display(int)
addEmployee()
removeEmployee()

Part Time

Full Time
BasicSalary
OtherPerks
display(int)
addEmployee()
removeEmployee()
modifySalary()

Attributes
with
Private
Access

Inheritance

Manager

Abstract Functions

Fee
timeWorked
display(int)

addEmployee()
removeEmployee()
modifyFee()

Attributes
with
Private
Access

Multiple Inheritance

Engineer

Developer

teamList
projectList

computerList
timeWorked

SkillArea
projectID

display(int)
addTeamMember()
removeTeamMember()
addProject()
removeProject()

display(int)
addComputer()
removeComputer()
installSoftware()
replaceHardware()

display(int)

Inheritance among Classes

The preceding figure shows that classes are designed at three levels. The class at each
subsequent level is derived from one or more classes of the previous level. For example,
the Employee class is the parent class of the Full Time and Part Time classes, which
inherit the properties of the Employee class. This is called single inheritance. In addition,
the Developer class inherits the properties of both the Full Time and Part Time classes.
This is called multiple inheritance.

©NIIT

Introducing Object-Oriented Analysis and Design 1.15


The Employee class is an abstract class as it includes abstract functions, such as
display(int), addEmployee() and removeEmployee(), which are implemented by the
derived classes, Full Time and Part Time.
The attributes of the Employee class are declared with protected access. This means that
the attributes of this class are accessible to the Full Time and Part Time classes.
Therefore, Full Time and Part Time classes can modify the attributes of the Employee
class. The attributes BasicSalary and OtherPerks of the Full Time class have private
access. Therefore, the derived classes, Manager, Engineer, and Developer cannot access

these attributes directly. The attributes of the Part Time class, fee and timeWorked, also
cannot be accessed by the Engineer or Manager class.
The display(int) function is provided in all classes. The display(int) function specified in
the Employee class is an abstract operation. The implementation of the display(int)
function is provided in the Full Time and Part Time classes. The derived classes, Manager
and Engineer, override the display(int) function of the Full Time class to display
information for the Manager and Engineer classes. This represents the polymorphism
concept of object-oriented design.

Just a minute:
Which phase of SDLC includes translation of the requirements specified in the SRS
into a logical structure that can be implemented in a programming language?
1.

Feasibility analysis

2.

Design

3.

Coding

4.

Maintenance

Answer:
2. Design

1.16 Introducing Object-Oriented Analysis and Design

©NIIT


Role of Object-Oriented Analysis and Design (OOAD) in
SDLC
To understand OOAD, you need to understand where OO approach fits into software
development life cycle. OO approach does not replace the standard approaches, such as
Data Flow Diagrams (DFD) or Entity Relationship (ER) diagrams. OO approach is only
an addition to the existing toolkit.
OOAD uses the OO approach to solve the real world problems. It uses OO approach to
analyze the system requirements and break a large and complex system into smaller and
simpler components.
OOAD is analysis of requirements and design of software system in terms of the objects,
classes, encapsulation, inheritance, polymorphism, abstraction, and dynamic binding. In
addition, OOAD is a methodology that can be applied to any software development
approach. If you compare the definition of analysis involved in any of the software
development approaches with the definition of OOAD, the only aspect that is new is
analyzing the requirements in terms of objects and object classes.

Note
An ER diagram is a modeling technique used to create a graphical representation of
the entities and their relationship. A DFD is a graphical technique used to depict data
flow throughout the system.

©NIIT

Introducing Object-Oriented Analysis and Design 1.17

Practice Questions
1.

2.

3.

4.

Match the models and diagrams with the modeling techniques.
a. Prototyping approach

i. Protecting access to non-essential details

b. Inheritance

ii. Iterative nature

c. Encapsulation

iii. Evolutionary approach

d. Spiral approach

iv. Sharing of attributes and behaviors
among classes based upon hierarchical
relationship

Which of the following phases of Waterfall approach involves a macro level study
of the customer requirements?
a. Conception
b. Initiation
c. Analysis
d. Construction
Which of the following approaches describes the software development process in a
linear sequential flow?
a. Waterfall approach
b. Prototyping approach
c. Spiral approach
d. Incremental approach
Which document is produced at the end of the requirement analysis phase of SDLC?
a. Software Requisition Statement
b. Software Requirement Specification
c. Software Requirement Statistics
d. System Requirement Statistics

1.18 Introducing Object-Oriented Analysis and Design

©NIIT


5.

©NIIT

Which of the following options focus on essential inherit aspects of an entity
ignoring its implementation details?
a. Inheritance

b. Abstraction
c. Encapsulation
d. Polymorphism

Introducing Object-Oriented Analysis and Design 1.19


Summary
In this chapter, you learned that:
„
SDLC is a disciplined and systematic approach towards software development.
„
SDLC divides the software development process into six phases:
z

Feasibility analysis

z

Requirement analysis and specification

z

Design

z

Coding

z

Testing

Maintenance
Different types of projects have different requirements. Therefore, it may be required
to tailor the SDLC phases according to the specific needs of the project. This gives
rise to various software development approaches. Some of these approaches are:
z

„

z

Waterfall approach

z

Prototyping approach

z

Spiral approach

z

Win-win spiral approach

Incremental approach
The function-oriented approach is module-centric and concentrates on the functions
of the software.

The object-oriented (OO) approach supports reusability and encapsulation of data
and concepts such as inheritance, abstraction, and polymorphism.
OOAD is analysis of requirements and design of software system in terms of objects,
classes, encapsulation, inheritance, polymorphism, abstraction, and dynamic binding.
z

„
„
„

1.20 Introducing Object-Oriented Analysis and Design

©NIIT


Exercises
Exercise 1
Technology Systems Inc. needs to develop software for a bank that will offer new service
to its customers. The service enables the customers of the bank to pay telephone bills
through the Internet. The bank is offering this service as an experiment, and as a result,
the requirements of the software are not consistent.
Being an experimental project, the requirements of the software could also evolve during
the software-development process. The new requirements of the software include
sophisticated user interaction and easy-to-use interface. Another important requirement
for the software is maintaining security and privacy of customer information because the
software will involve authorizing and authenticating credit-card information provided by
the customer.
The bank is also planning to provide the same service on mobile phones. Other
competitors of the bank are also planning to offer the same service and, therefore, the
bank requires a working system for this software as soon as possible. Only a few top-level

managers of the bank, who have identified this new service, can provide the full
requirements on this software. However, these employees are not frequently available due
to their workload and prior appointments.
Identify the approach to be used for developing the software for the bank and justify your
choice of approach.

©NIIT

Introducing Object-Oriented Analysis and Design 1.21


1.22 Introducing Object-Oriented Analysis and Design

©NIIT