May-June 2001 ISISTAN Research Institute – Tandil, Argentina Software Design Methodologies: UML in Action Dr. Mohamed Fayad, J.D. Edwards Professor Department

May-June 2001 ISISTAN Research Institute – Tandil, Argentina

Software Design Methodologies: UML in Action

Dr. Mohamed Fayad, J.D. Edwards Professor

Department of Computer Science & Engineering

University of Nebraska, Lincoln

Ferguson Hall, P.O. Box 880115

Lincoln, NE 68588-0115

http://www.cse.unl.edu/~fayad

May-June 2001 ISISTAN Research Institute – Tandil, Argentina -- M.E. Fayad

L4-S2 Object Identification-1

2

Lesson 4:Object Identification - 1



Lesson Objectives

Objectives

3

Understand object identification & class classification Learn how to identify:

Objects and classes

Textual Specification Analysis

Data Analysis

Behavior Analysis

Use Case Analysis Associations and aggregations using Abbott's approach

Understand how to use the following approaches: Data Analysis Use Case



4

Classification Involves Ordering Knowledge

Finding Similarities– Common Attributes– Common Behaviors

OO Classifies Software– Exposes existing commonalities– Invents stable abstractions

Goal is a simple & natural design



5

No Single Best Classification Structure

Depends upon

– Domain

– Application

– Experience

– Creativity Water Animals

Air AnimalsLand Animals

?



6

Same Object Can Be Perceived from Several Perspectives

Data-Drivenhead, tail, body, leg

Behavior-Drivenwalk, run, eat

Responsibility-Drivencarry things, communicate,maintain its living system



Textual Specification Analysis

Data Analysis

Behavior Analysis

Use-Case Analysis

Responsibility Analysis 7

Object/Class Identification Techniques



Abbott’s Noun Approach– Use noun, pronoun, and noun phrases to identify

abstract objects and classes.– Use singular proper nouns (e.g., sensor number 5) and

nouns of direct reference (e.g., the fifth sensor) to identify abstract objects.

– Use plural and common (e.g., sensor) nouns to identify classes.

– Use verbs and predicate phrases (e.g., are simultaneously activated) to identify the associated operations.

Comments– This approach is the oldest approach 8

Textual Analysis Approach



Benefits:– Easy for beginners to use– Abbott’s mapping should usually work– Can be used with pre-existing textual requirements

specifications– Does not require a complete paradigm shift

Risks– Indirect– Many software engineers are weak in grammar– English is vague, Examples

• Some nouns can be used as verbs and vis versa• Some words (e.g., purchase, record) can be used as both nouns and

verbs– Assumes user’s requirements are coherent, complete and

correct– No tool support 10

Textual Analysis Approach



Classes should make sense in the problem domain.

Good classes classify the objects which need to be modeled in the system.

Classes often correspond to NOUNS.

Avoid redundant or irrelevant classes which add no value in the problem domain.

Remove classes which have no attributes. 12

Selecting Good Classes



After initial pass, discard classes which are:

Redundant

Irrelevant to the problem domain

Vague

Attributes

If class name has no attributes of its own, it is probably an attribute.

11

Elimination of Inappropriate Classes



The Problem Statement:

A simple cash register has a display, an electronic wire with a plug, and a numeric keypad which has keys for subtotal, tax, and total. This cash storage device has a total key which triggers the release on the drawer. The numeric buttons simply place a number on the display screen, the subtotal displays the current total, the tax key computes the tax, and the total key adds the subtotal to the tax.

• Identify all the classes in this problem statement• Use the class elimination rules to eliminate the

unnecessary classes. 12

Example 1: Simple Cash Register



We are going to use nouns to find classes Nouns (initial)Register Display WirePlug Keypad KeysDevices Release DrawerButtons Screen NumberTotal Tax

Nouns (General Knowledge)0-9 keysMoneySubtotal KeyTax KeyTotal Key

13

Classes in the initial pass



14

Eliminating Unnecessary Nouns

Register

Display

Wire ---> Irrelevant

Plug ---> Irrelevant

Keypad

Keys

Devices ---> Vague

Release ---> Irrelevant

Drawer

Buttons ---> Redundant

• Screen ---> Redundant

• Number ---> Attribute

• Total ---> Attribute

• Tax ---> Attribute

• 0-9 Key

• Value ---> Attribute

• Money

• Subtotal Key

• Tax Key

• Total Key



15

Data Analysis Approach

Name Breed Owner Name

Name Address

Happy Poodle JoeTasha Pit Bull EricKing, Jr. Shepherd Carol

Carol 2601 Lake St.

1. Identify abstract objects as table

2. Identify instances as rows in table

3. Identify relationships between objects

Dog Owner

Dog

But how do you find your abstract objects?



Tangible things -- airplane, book, table

Roles -- doctor, professor

Incidents -- accident, flight

Interactions -- purchase, marriage

Specifications -- insurance policy

People -- humans who carry out some function

Places -- areas set aside for people or things

Organizations -- formally organized collections of people, resources. and facilities 16

Analyzing the Domain for Abstract Objects



Associations often refer to verbs of verb phrases – Examples: next to, contains, part of, works for,

married to, downstream from, connected to, etc.

These may be explicit in the problem statement or implicit in the knowledge of the problem domain

Write down all candidates, then eliminate unnecessary ones and add others

Aggregation is just a common type of association 17

Identifying Associations/Aggregations



Data Flow Diagrams (DFDs)

State-Transition Diagrams (STDs)

Semantic Nets or Object-Interaction

Diagrams (OIDs)

Message or communication Diagrams

18

Behavior Analysis



19

Looking for Behavior

Examine required Processing or behavior of system components

Objectify common behavior

– Use inheritance

DFDs may help identify processesSwimmers

Flyers

Runners



DFD Approach

– Use each data store on a DFD to identify an abstract object or class

– Use (all or part of) the transforms associated with data store to identify associated operations

Candidate Objects

– external entities

– data stores

– control transformations

Candidate Classes

– data flows

20

Data Flow Diagrams (DFDs)



Benefits– Very well-known approach– Many requirements analysis methods are

based on DFDs.– Tool support exists– Does not require paradigm shift

Risks– Data abstraction – Indirect– Traditional DFDs have the wrong scope 21

Data Flow Diagrams (DFDs) (cont’d)



State Approach

– Identify an object or a class for each entity that has a state

Benefits

– Can be used at any time that an object or a class has a finite number of obvious states

Risks

– The state can belong to the entire subsystem or an operation

– An object or a class may have an infinite number of states.

– An object or a class may have states that are not all obvious.

– An object or a class may have only a small number of trivial states. 22

State-Transition Diagrams (STDs)



23

Use Case Analysis: Baseball System

Pitch a Ball

Run the Bases

Hit a Home Run

Runner on First Base

Batter

Pitcher

Scored a Run

1

1

1

1



Base hit Single base hit

• In field base hit• Outfield base hit

Double base hit Triple base hit Home run

• A home run• Two home runs• Three home runs• Grand Slam 24

Some Use Cases in Baseball Game



25

A Use Case Description

Use Case: Hit a Home Run When a batter hits a home run, the runners run the bases

and reach home plate and score When a batter hits a home run, the batter runs the bases

until reaching the home plate and scores When the runners score, the score board updates the

score board, the game announcer is going crazy and saying “ a big home run ... holy cow ... holy cow “ and the fans are going wild and screaming a lot.

When ..



1. Define: use case model, use case, specification objects, incident objects, organization objects.

2. List all the object identification techniques

3. What are the benefits and risks of each of these techniques?

4. Describe how do you identify associations and aggregations

5. Explain:

a. The same object can be perceived from several perspectives.

b. No single best classification structure.

c. Classification involves ordering knowledge.

d. How to select good classes

e. How to eliminate inappropriate classes 26

Discussion Questions



Check the definition of: Classes, Objects, Attributes, Interfaces, Abstraction, Behavior, Responsibility, Roles, Associations, Aggregation.

Describe the following Models: Use Case Model, DFD, STD, OID, OCD.

27

Questions for the Next Lecture

Documents

May-June 2001 ISISTAN Research Institute – Tandil, Argentina Software Design Methodologies: UML in Action Dr. Mohamed Fayad, J.D. Edwards Professor Department