Lectures 2

1
Lectures 2 3 Software Process Models

• Neelam Gupta

2
Linear Sequential Model(Waterfall Model)
Requirement Analysis and Specification
Design and Specification
Coding and Module Testing
Integration and System Testing
3

• Contributions
• Introduces discipline into the software
  development process
• Implementation of a product should be postponed
  until after the objectives of implementation are
  well understood
• Drawbacks
• Linear
• Rigid
• Monolithic

4
The Prototyping Model

• Do it twice principle by Brooks
• First version Throwaway prototype
• Second version Actual solution using Waterfall
  model
• Drawbacks
• Quick fixes may not be evaluated
• Does not eliminate the time gap between the
  definition of requirements and final delivery of
  application
• No emphasis on need for anticipating changes

5
Evolutionary Process Model

• Software evolves
• Requirements may change
• Market competition
• Iterative in nature
• Deliver core product to the real user
• Measure the added value of the user
• Develop the next increment of the product

6
Incremental Model

• Special care is placed on useful subsets of the
  system that might be delivered to the user and on
  definition of interfaces that will later allow
  new subsystems to be added smoothly.
• Example Word processing software
• Basic file management, editing, and document
  production
• More sophisticated editing features
• Spelling and grammar
• Advanced page layout

7
Transformation Model

• Software Development

Formal Specification
Sequence of Automated Transformations
Implementation
8
Requirement Analysis And Specification
Informal Requirements
Formal Specifications
Verification
Reusable Components
Detailed Formal Specifications
Executable Formal Specifications
Optimized Executable Specifications
Test
Formal Development Record
Release
9

• In contrast to Waterfall model
• supports program evolution
• Also provides a method for proving program
  correctness.
• Program correctness proofs represent an analytic
  mathematical approach for analyzing correctness
  of program after it is developed.
• Given P and SpecificationltPre,Postgt,..,
  verify truth of Pre P Post.
• Transformation approach is constructive
• Given specification pairs ltPre,Postgt
• Derive P that satisfies all the ltPre,Postgt pairs
  in the specification.
• Automatically derives a program that satisfies
  specification.

10
Advantages

• Enable a software engineer to specify, develop,
  and verify a software system by applying a
  rigorous mathematical notation.
• Ambiguity, incompleteness, and inconsistency can
  be discovered/corrected easily by mathematical
  analysis.
• Use of formal methods serves as a basis of
  program verification.
• Offers promise of defect free software
  development.
• Programmer is constrained to operate within well
  defined formal boundaries can focus on design
  complexity kept under control.

11
Recent Research

• Use of techniques developed in Artificial
  Intelligence to support environment for
  transformational model.
• Automated Assistant mediates and supports
  activities of the software process as directed by
  Software Engineer
• Recording derivation steps and rationale
• Producing necessary information for SE to make
  decisions
• Suggesting strategies

12
Drawbacks

• Requires extensive mathematical background and
  training in applying formal methods.
• Still a research oriented approach but promising
  for software development for safety critical
  systems such as aircraft avionics and medical
  devices.

13
Spiral Model

• Provides a framework for designing software
  processes that incorporate risk assessment and
  management into each phase of software
  development.
• Meta model.
• Definitions
• Risks potentially adverse circumstances that may
  impair the development process and quality of
  products.
• Risk Management identify, address and eliminate
  software risk items before they become a threat
  to successful software operation or a major
  source of expensive software rework.

14
(No Transcript)
15

• Cyclic
• Each cycle of spiral consists of 4-6 stages
• Radius of spiral cost accumulated so far in the
  process
• Introduces prototyping and risk and cost
  evaluation in each phase.
• Couples evolutionary nature of software
  development with controlled and systematic
  aspects of linear sequential model.
• Provides potential for rapid development of
  incremental versions of software.

16

• Allows to restate the issues of robustness vs
  correctness
• After one cycle, unstated requirements checked as
  part of robustness of application
• May become part of specifications in the next
  cycle
• After each cycle, robustness approximates
  correctness more closely.

17
Spectrum of Models

• Waterfall Life Cycle Model
• Evolutionary Approaches
• Transformational Approach
• Spiral Method (meta model)

• Documentation driven
• Increment driven
• Specification driven
• Risk driven