Software Engineering

1
Software Engineering

• Part1 - Introduction

2
Software Engineering Part 1

• State what is meant by the term software
  engineering?
• Describe the software process or lifecycle
• Explain the importance of Software Development
• List the attributes of well designed software
• Describe the following Software Development
  process models
• Waterfall
• Evolutionary or iterative development
• Prototyping
• Extreme programming
• Spiral

3
Problems with Software

• Developing software for large-scale systems is
  much more difficult than development for smaller
  projects
• Complexity the more complex a task the more
  likely errors will be introduced. Modularisation
  of a large, complex task into smaller sub-tasks
  can reduce the overall complexity
• Sizeas the size of a project increases, the
  number of individuals involved increases leading
  to increased communicational complexity and thus
  higher probability of errors occurring. This
  requires a clear medium of communication that is
  reliable and complete.
• TimescalesQuality of software versus cost of
  lateness
• Costsoftware is now the predominant factor in
  computer system cost, far exceeding hardware
  cost.
• ManagementGood project planning vital to
  development of quality software on time.

4
Case Study from SommervilleTherac-25

• Linear accelerators create high- energy beams
  that can destroy tumors with minimal impact on
  the surrounding healthy tissue
• Therac 25 was the first linear accelerator with
  dosage controlled solely by software (as opposed
  to hardware)

Ref sommerville
5
1983 Pre-release Safety Analysis

• Programming errors have been reduced by extensive
  testing on a hardware simulator and under field
  conditions on teletherapy units. Any residual
  software errors are not included in the analysis.
  
• Program software does not degrade due to wear,
  fatigue, or reproduction process.
• Computer execution errors are caused by faulty
  hardware components and by "soft" (random) errors
  induced by alpha particles and electromagnetic
  noise.

Ref sommerville
6
and then

• 1983 First Therac 25 installed
• 1985-1987 Six massive-overdose accidents due to
  software error are reported. Overdoses caused
  severe burns and death.
• 1987 Recalled for extensive design changes,
  including hardware to safeguard against software
  errors in dosage.

Ref sommerville
7
US Federal Aviation Administration

• 1981 FAA announced plans to modernize
  air-traffic control.
• 1985 IBM awarded contract. System estimate to
  have 1.5 million lines of code, cost 2.5
  billion, and be deployed by 1991.
• 1987 Revised cost 4.3 billion, deployment
  slipped to 1995.
• 1994 FAA decided that the project would never be
  completed, and cancelled it. Net loss 1.5
  billion
• Main issue causing project delay and cost growth
  was the lack of coordination between the FAA and
  air traffic controllers and service technicians
  during the early stages of development - led to
  design that could significantly impact the
  controllers ability to control aircraft.

Ref sommerville
8
Stats on software projects

• 31.1 are cancelled before they are finished
• 52.7 overrun their cost estimates by at least
  189
• 33.3 overrun their time estimates by 100-200
• 94 of all projects do a restart

J. Johnson, Creating Chaos, American
Programmer, July 1995
Ref sommerville
9
So why do projects fail

• Unrealistic Schedules
• Inappropriate Staffing
• Changing Requirements During Development
• Poor-Quality Work
• Believing in Magic

Ref sommerville
10
What is Software Engineering?

• Software engineering is concerned with
  theories, methods and tools for the systematic
  production and maintenance of software products
  that are developed and maintained on time and
  within cost estimates.

Ref sommerville
11
Software Engineering

• A systematic approach to the development of
  medium to large-scale software systems, usually
  involving teams of programmers.
• Good software development techniques required to
  minimise programming errors and localise the
  scope of errors.
• The development of the documentation necessary to
  install, use, develop and maintain the programs
  is crucial
• Cost-effective software development
• A technological and a managerial discipline.
• Aims to reduce cost of software development by
• Placing greater emphasis on systematic
  development
• Finding out the users requirements
• Specifying the system requirements in some
  formal/semi-formal way
• Producing prototypes
• Trying to ensure error free code

12
Software Cost!

• Bear in mind that
• it is 100 times more expensive to redesign a
  problem after delivery of the product than to
  identify the users requirements correctly at the
  start of the design,
• maintenance costs are typically twice the
  development,
• about 15 of development costs are in coding
  software is more than CODE.
• and it is obvious that effort spent in a
  systematic approach (i.e. engineering) from the
  start of a project will not be wasted.

13
Aim of software engineeringThe aim is to develop
software which is

• RELIABLE
• Software should perform as expected by the user.
  This is particularly important in safety or
  life-critical areas such as Avionics, chemical
  plant, medical systems.
• MAINTAINABLE
• Software should be written so that changes can be
  made without undue cost.
• EFFICIENT
• Software should not make wasteful use of system
  resources.
• USABLE
• Software should have an appropriate user
  interface. For example, ease of use, ease of
  learning, presentation, attractive.
• SECURE
• This includes confidentiality and access rights.

14
What is a Software Process/Software LifeCycle?

• A set of activities whose goal is the development
  or evolution of software.
• Generic activities in all software processes are
  
• Requirement specification/analysis
• Design
• Implementation
• Testing
• Evolution

15
Life Cycle Model

• Several different activities may be carried out
  in each life cycle phase.
• Entry and exit criteria defined for every phase.
• A phase can start only if its phase-entry
  criteria have been satisfied.
• A phase is considered to be complete only when
  all its exit criteria are satisfied.

16
Software Development Lifecycle Models

• There are a number of models for software
  development lifecycle, for example
• Build and fix model
• Classic Waterfall Model
• Incremental model
• Spiral model
• Rapid prototyping
• Extreme Programming

17
Build and fix model

• trial and error approach
• adequate for simple software programs
• unacceptable for large and complex systems

18
Waterfall Model
19
Waterfall Model

• Advantages
• Organised approach, provides robust separation of
  phases
• Reflects common engineering practice
• Disadvantages
• Doesnt cope well with changes required by the
  client
• Development teams might wait for each other
• A working version of the product is available
  only late
• Applicability
• When requirements are well known and few changes
  are likely to be needed
• Can be used also for parts of larger software
  systems

20
Process Iteration

• System requirements ALWAYS evolve in the course
  of a project so process iteration where earlier
  stages are reworked is always part of the process
  for large systems.
• Iteration can be applied to any of the generic
  process models.

21
Iterative Development

• Waterfall model breaks down a project based on
  activities
• requirement analysis, design, coding and testing.
• Iterative development breaks down a project by
  subsets of functionality.
• Overall lifecycle composed of several iterations
• each iteration a self-contained mini-project
  composed of activities such as requirement
  analysis, design, coding and testing.
• Iteration Release
• At the end of each iteration
• a stable integrated and tested system with a
  subset of the final functionality completed.
• provide a baseline for the developer
• usually not released to the customer until the
  final iteration.

22
Iterative Development Approaches

• Evolutionary
• Incremental
• Spiral

23
Evolutionary development

• Main characteristics
• The phases of the software construction are
  interleaved
• Feedback from the user is used throughout the
  entire process
• The software product is refined through many
  versions

24
Evolutionary development

• Exploratory development
• Objective is to work with customers and to evolve
  a final system from an initial outline
  specification. Should start with well-understood
  requirements and add new features as proposed by
  the customer.
• Throw-away prototyping
• Objective is to understand the system
  requirements. Should start with poorly understood
  requirements to clarify what is really needed.

25
Evolutionary development

• Advantages
• Deals constantly with changes
• Provides quickly an initial version of the system
• Customer input throughout development yields
  system closer to their (immediate) needs
• Problems
• Lack of process visibility, not well-supported by
  documentation
• Systems structure can be corrupted by continuous
  change
• Special skills (e.g. in languages for rapid
  prototyping) may be required.
• Applicability
• For small or medium-size interactive systems
• For parts of large systems (e.g. the user
  interface)
• For short-lifetime systems.

26
Incremental delivery

• Rather than deliver the system as a single
  delivery, the development and delivery is broken
  down into increments with each increment
  delivering part of the required functionality.
• User requirements are prioritised and the highest
  priority requirements are included in early
  increments.
• Once the development of an increment is started,
  the requirements are frozen though requirements
  for later increments can continue to evolve.

27
Incremental delivery advantages

• Customer value can be delivered with each
  increment so system functionality is available
  earlier
• e.g. ATM 1)withdraw cash 2)lodge cash 3)show
  balance 4)print statement 5)change PIN number
  6)Top-up mobile phone
• Early increments act as a prototype to help
  elicit requirements for later increments.
• Lower risk of overall project failure.
• The highest priority system services tend to
  receive the most testing.

28
Spiral development

• Combines features of the rapid prototyping and
  waterfall models.
• Process is represented as a spiral rather than as
  a sequence of activities with backtracking.
• The project starts at the centre of the spiral
  and builds progressively more complete version of
  the software as the spiral expands outwards.
• Each loop in the spiral represents a phase in the
  process and is split into 4 sectors.
• No fixed phases such as specification or design -
  loops in the spiral are chosen depending on what
  is required.
• Risks are explicitly assessed and resolved
  throughout the process leading to a risk based
  development model.

Customer review
29
Spiral model of the software process
30
Spiral model sectors

• Objective setting
• Specific objectives for the phase are identified.
• Risk assessment and reduction
• Risks are assessed and activities put in place to
  reduce the key risks.
• Development and validation
• A development model for the system is chosen
  which can be any of the generic models.
• Planning
• The project is reviewed and the next phase of the
  spiral is planned.

31
Spiral model features

• Advantages
• Risk reduction mechanisms are in place
• Supports iteration and reflects real-world
  practices
• Systematic approach
• Disadvantages
• Requires expertise in risk evaluation and
  reduction
• Complex, relatively difficult to follow strictly
  
• Applicable only to large systems
• Applicability
• Typically used for large, expensive projects.

32
Prototyping

• Construction of prototypes is standard practice
  in engineering.
• Requirements can be validated by prototyping,
  thus reducing risk.
• Prototypes are used during feasibility studies
  and are often constructed from reusable parts.
• Typically used for 4GL development projects.

33
Extreme Programming

• This is a so called agile method developed in
  the 1990s for smaller systems which have to be
  developed quickly
• allow the development team to concentrate on the
  software rather than on the design and
  documentation
• Agile methods rely on an iterative approach as in
  the Incremental Model but with the end user
  involved very closely with the developers
  throughout the development process the user is
  a member of the development team.
• The user is continually monitoring output,
  providing feedback and modifying his requirements
  specification as required.
• The customer is involved to extreme levels in
  frequent iterative development - hence the title.
  
• It is approached through the use of scenarios-
  what will happen if?.