The University of Akron Summit College Department of Business Technology Computer Information System

1
The University of AkronSummit CollegeDepartment
of Business TechnologyComputer Information
Systems

• 2440 241 Systems Analysis and Design
• Chapter 2 Approaches to System Development
• Instructor Enoch E. Damson

2
The Systems Development Life Cycle (SDLC)

• Analysts organize their work into projects
• Project a planned undertaking that has a
  beginning and an end, and produces a desired
  result or product
• System development project a planned
  undertaking that produces a new information
  system
• Information systems development go through phases
• Phase related system development activities
  that are grouped into categories of planning,
  analysis, design, implementation, and support
• The three major information systems development
  phases are as follows
• Analysis phase
• Design phase
• Implementation phase

3
The Systems Development Life Cycle (SDLC)

• Analysis phase provides a thorough
  understanding of the businesss information needs
  and requirements
• Design phase defines the architecture and
  structure of a new system to satisfy the
  businesss requirements
• Implementation phase involves the actual
  construction, testing, and installation of the
  function information system

4
The Systems Development Life Cycle (SDLC)

• The two other systems development phases include
• Project planning phase
• Support phase
• Project planning phase the beginning of a
  systems development project that consists of
  required activities to initiate, plan, and obtain
  approval for the project
• Support phase the most expensive and longest
  phase that consists of ongoing activities to
  maintain and enhance the system (after its
  completion and installation) over its lifetime

5
The Systems Development Life Cycle (SDLC)

• The approach to defining the required phases and
  activities for a system development project is
  called the systems development life cycle (SDLC)
• The five phases of the systems development life
  cycle includes the following
• Project Planning phase
• Analysis phase
• Design phase
• Implementation phase
• Support phase
• Current approaches to system development include
  iteration phases that tend to break projects down
  into a series of mini-projects (iterations)
• Each iteration involves analysis, design, and
  implementation activities

6
The Phases of the Systems Development Life Cycle

• Identifying the objectives and primary activities
  of each of the SDLC phases is vital to
  understanding information systems development

7
The Planning Phase

• Primary objective identify the scope of the new
  system, ensure that project is feasible, and
  develop a schedule, resource plan, and budget for
  the remainder of the project
• Activities in the planning phase includes
• Defining the problem
• Producing the project schedule
• Confirm project feasibility
• Staff the project
• Launch the project

8
The Analysis Phase

• Primary Objective understand and document
  business needs and processing requirements of the
  new system
• Keywords in analysis phase are discovery and
  understanding
• Activities of the phase include
• Gathering information
• Defining system requirements
• Building prototypes for discovery of requirements
• Prioritizing requirements
• Generating and evaluating alternatives
• Reviewing recommendations with management
• Gathering information is fundamental to the
  analysis phase
• Gathering information involves learning from
  users about the problem domain
• Problem domain the area of users business for
  which a system is being developed

9
The Design Phase

• Primary objective design the solution system
  based on the requirements defined and decisions
  made during analysis
• Activities in this phase include
• Designing and integrating the network
• Designing the application architecture
• Designing the user interfaces
• Designing the system interfaces
• Designing and integrating the database
• Prototyping for design details
• Designing and integrating the system controls
• The application is the portion of the new
  information system that satisfies the users
  needs in the problem domain

10
The Implementation Phase

• Primary objective produce a reliable, fully
  functional information system and ensure that
  users are all trained and that the organization
  is ready to benefit as expected from use of the
  system
• Activities include
• Constructing software components
• Verifying and testing
• Convert data
• Train users and document the system
• Install the system

11
The Support Phase

• Primary objective keep the system running
  productively after initial installation
• Activities include
• Maintaining the system
• Enhancing the system
• Supporting the system

12
Scheduling Project Phases

• Analysts in the 1970s and 1980s used the
  waterfall approach to their projects
• Waterfall approach executing an SDLC in which
  one face leads sequentially into the next phase
• The waterfall method required rigid planning and
  final decision making at each step of the SDLC
• Analysts rarely attempt to use a straight
  waterfall approach

13
Scheduling Project Phases

• Most activities of the SDLC phases overlap and
  appear to be done concurrently
• The overlap occurs because of the following
  reasons
• Efficiency some analysis thoughts may lead to
  design thoughts
• Interdependency most components of a computer
  system depend on each other leading to both
  analysis and some design at the same time
• Not all activities overlap completely in the SDLC
  because of dependency some activities naturally
  depend on results of prior work
• Most successful projects have schedules that
  provide some flexibility but requiring completion
  of identified milestones on a timely basis

14
Understanding Iteration and Project Phases

• Overlapping schedules are best defined and
  managed using iteration across phases
• With iteration, work activities are repeated and
  each iteration is refined to be closer to the
  desired result
• There are several approaches to defining each
  iteration including
• Defining and implementing key functions the
  system must include in the first iteration before
  getting to the less crucial functions and then to
  the optional functions
• Focusing on one subsystem at a time with the
  first subsystem containing the core functionality
  and data for the other subsystems
• Using the incremental development approach that
  completes parts of the system in one or more
  iterations and puts them into operation for users
• All approaches to system development can use some
  amount of iteration.
• The object-oriented approach to system
  development is always described as highly
  iterative

15
Methodologies, Models, Tools, and Techniques

• Systems analysts have variety of aids to help
  complete activities in the SDLC
• The aids include
• Methodologies
• Models
• Tools
• Techniques

16
Methodologies

• System development methodology provides
  guidelines for completing every activity in the
  SDLC, including specific models, tools, and
  techniques
• Some methodologies are homegrown or purchased and
  contain written documentation that defines
  everything needed to produce at any point of the
  project development

17
Models

• A model is a representation (abstraction) of an
  important aspect of the real world
• Some models are
• physically similar to the real product
• graphical representations of important details
• abstract mathematical notations
• Models used in systems development include
  representations of inputs, outputs, processes,
  data, objects, object interactions, locations,
  networks, and devices

18
Models

• Most models are graphical models drawn that
  employ agreed-upon symbols and conventions called
  diagrams and charts
• Some models of system components include
• Flowchart
• Data flow diagram (DFD)
• Entity-relationship diagram (ERD)
• Structure chart
• Use case diagram
• Class diagram
• Sequence diagram
• Some models used to manage the development
  process include
• PERT chart
• Gantt chart
• Organizational hierarchy chart
• Financial analysis models NPV, ROI, BSC

19
Tools

• A tool in the context of system development is
  software support that helps create models or
  other components required in the project
• The most comprehensive tool available for system
  developers is called a computed-aided system
  engineering (CASE) tool
• CASE tools help analysts create the important
  system models to automatically check the models
  for completeness and compatibility with other
  system models
• CASE tools can also generate program code based
  on the models
• Some tools used in system development include
• Project management application
• Drawing/graphics application
• Word processor/text editor
• Computer-aided system engineering (CASE) tools
• Integrated development environment (IDE)
• Database management application
• Reverse-engineering tool
• Code generator tool

20
Techniques

• A technique in system development is a collection
  of guidelines that help an analyst complete a
  system development activity or task
• Techniques often include step-by-step
  instructions or for creating a model a more
  general advice for collecting information from
  system users
• Some techniques used in system development
  include the following
• Strategic planning techniques
• Project management techniques
• User interviewing techniques
• Data-modeling techniques
• Relational database design techniques
• Structured analysis technique
• Structured design technique
• Structured programming technique
• Software-testing techniques
• Object-oriented analysis and design techniques

21
Relationships Among Components of a Methodology

• A methodology includes a collection of techniques
  used to complete activities within each phase of
  the SDLC
• The activities include completion of a variety of
  models and other documents and deliverables
• System developers use software tools to help
  complete activities

22
Two Approaches to System Development

• Systems development is done in many different
  ways
• Different companies, groups in a company,
  individuals in a group may have different ways of
  developing systems
• The two general approaches to system development
  that form the basis of virtually any
  methodologies include
• The traditional approach
• The object-oriented approach

23
The Traditional Approach

• The traditional approach includes many variations
  based on techniques used to develop information
  systems with structured and modular programming
• The approach is often referred to as structured
  system development
• A refinement to structured development is called
  information engineering (IE), a popular variation

24
Structured System Development

• The structured approach to system development
  uses structured analysis, structured design, and
  structured programming
• The approach is also known as the structured
  analysis and design technique (SADT)
• The structured programming technique, developed
  in the 1960s, was the first attempt to provide
  guidelines to improve quality of computer
  programs
• The structured design technique was developed in
  the 1970s to make it possible to combine separate
  programs into more complex information systems
• The structured analysis technique evolved in the
  early 1980s to help clarify the requirements for
  the computer system for developers before they
  designed the programs

25
Structured Programming

• A structured program is one that has one
  beginning and one ending, and for which each step
  on the program execution consists of
• Sequence
• Decision
• Repetition constructs
• Another structured programming concept is
  top-down programming which divides more complex
  programs into a hierarchy of program modules
• With top-down programming, one module at the top
  of the hierarchy controls program execution by
  calling required lower-level modules
• When the hierarchy involves multiple programs,
  the arrangement is sometimes called modular
  programming

26
Structured Design

• The structured design technique was developed to
  provide some guidelines for deciding
• what the set of programs should be
• what each program should accomplish
• how the programs should be organized into a
  hierarchy
• The modules and arrangements of the modules are
  shown graphically using a model called a
  structure chart
• Two main principles of structured design are that
  program modules should be designed to be
• Loosely coupled each module is independent of
  other modules
• Highly cohesive each module accomplishes one
  clear task
• Newer structured design assume database
  management systems (DBMS) are used in the system
  to interact with a database

27
Structured Analysis

• The structured analysis technique helps the
  developer to define
• what the system needs to do (the processing
  requirements)
• what data the system needs to store and use (data
  requirements)
• what inputs and outputs are needed
• how the functions work together as a whole to
  accomplish tasks
• The key graphical model of the system
  requirements used with structured analysis is
  called the data flow diagram (DFD)
• The data flow diagram is a model that shows
  inputs, processes, storage, and outputs of a
  system produced in structured analysis

28
Structured Analysis

• The most recent variation of structured analysis
  defines systems processing requirements by
  identifying all of the events that will cause the
  system to react in some way
• Each event leads to a different system activity
• The analyst takes each of these activities and
  creates a DFD showing the processing details,
  including inputs and outputs
• A graphical model called an entity-relationship
  diagram (ERD) can also be created based on the
  data needed by a system, including things about
  which information (data entities) is stored and
  the relationships among them, produced in
  structured analysis and information engineering
• The data entities from the ERD correspond to the
  data storage shown on DFDs

29
Weaknesses of the Structured Approach

• The structured approach has been considered to be
  weak because critics thought that
• the techniques address only some, but not all, of
  the activities of analysis and design
• the transition from the DFD (in structured
  analysis) to the structure chart (in structured
  design) did not work well in practice
• the data modeling and the ERD were much more than
  modeling processes with the DFD
• to ensure that systems are comprehensive and
  coordinated, the development of a system should
  begin only after the organization completed an
  overall strategic planning effort
• Because of these weaknesses, some developers
  turned to a refinement of structured development
  named information engineering (IE)

30
Information Engineering

• Information engineering (IE) is a traditional
  system development methodology thought to be more
  rigorous and complete than the structured
  approach, because of its focus on strategic
  planning, data modeling, and automated tools
• IE focused much more on data
• The process model of IE process dependency
  diagram is similar to a DFD but focuses more on
  which processes are dependent on other processes
  and less on inputs and outputs
• IE provides a more complete life cycle support
  through the use of an integrated CASE tool
• The CASE tool helps analysts to follow the IE
  approach faithfully but it lacks flexibility
• By the late 1980s, the IE was popular with large
  mainframe systems but less useful for smaller
  desktop applications and client/server
  applications because of the lack of flexibility

31
The Object-Oriented Approach

• Many information systems projects today are now
  using the object-oriented technology which
  requires a completely different approach
• The object-oriented approach views an information
  system as a collection of interacting objects
  that work together to accomplish tasks
• Conceptually, there are no processes (or
  programs) and data entities (or files)
• The system consists of objects which is a thing
  in the computer system that can respond to
  messages

32
The Object-Oriented Approach

• The object-oriented approach began with the
  development of the Simula programming language in
  Norway in the 1960s
• Simula was used to create computer simulations
  involving objects like ships, etc
• In the 1970s, the Smalltalk language was
  developed to solve the problem of creating
  graphical user interfaces (GUIs) that involved
  objects like buttons, menus etc
• Other programming languages include C, Java, C

33
The Object-Oriented Approach

• Object-oriented analysis (OOA) defines all of the
  types of objects that do the work in the system
  and shows what user interactions are required to
  complete tasks
• Object-oriented design (OOD) defines all of the
  types of objects necessary to communicate with
  people and devices in the system, showing how
  objects interact to complete tasks, and refining
  the definition of each type of object so it can
  be implemented with a specific language or
  environment
• Object-oriented programming (OOP) consists of
  writing statements in a programming language to
  define what each type of object does, including
  the messages that the objects send to each other
• The object-oriented approach uses a class diagram
  to show all of the classes of objects in the
  system
• Every class may have more sophisticated
  subclasses that inherit characteristics of the
  class above them

34
Benefits of the Object-Oriented Approach

• Two major benefits of the object-oriented
  approach include
• Naturalness intuitive because people tend to
  think about the world in terms tangible objects
• Reuse many systems in organizations use the
  same objects, so the classes can be used over and
  over again when needed
• Many systems developed today combine both
  traditional and object-oriented technology
• Some IDEs also combine the two approaches in the
  same tool

35
Systems Development Life Cycle Variations

• There are some variations to the SDLC
• Some of the variations are based on the life
  cycle phases
• Some life cycles differ in their emphasis on the
  social, technical, and subsystems
• Following are some of the variations
• Variations of names for phases
• Variations based on an emphasis on people
• Variations based on speed of development

36
Variations of Names for Phases

• Early SDLC
• Feasibility study
• System investigation and analysis
• Systems design
• Implementation
• Review and maintenance
• Information Engineering (IE)
• Information strategy planning
• Business area analysis
• Business systems design and technical design
• Construction and transition
• Production
• Unified Process (UP)
• Inception phase
• Elaboration phase
• Construction phase
• Transition phase

37
Variations of Names for Phases

• SDLC with Activity Names and Phases
• Organize the project and study feasibility
• Study and analyze the current system
• Model and prioritize the functional requirements
• Generate alternatives and propose the best
  solution
• Design the system
• Obtain needed hardware and software
• Build and test the new system
• Install and operate the new system

38
Variations Based on an Emphasis on People

• Some methodologies recognize more explicitly that
  information systems are socio-technical systems
• Socio-technical systems information systems
  that include both social and technical subsystems
  designed to work well together
• Other terms for socio-technical system
  development include user-centered design and
  participatory design

39
Variations Based on Speed of Development

• Reasons for speeding up the development process
  include
• Continuing backlog of needed systems
• Constantly changing technological and business
  environment
• Some of the variations include
• Rapid application development (RAD)
• Prototype development

40
Current Trends in Development

• Below are some of the new trends in systems
  development
• Risk and the spiral model
• extreme programming (XP)
• The unified process (UP)
• Agile Modeling

41
Risk and the Spiral Model

• The spiral model is a highly iterative model
• The model shows the SDLC as spiral, starting in
  the center and working its way around, over and
  over again, until project completion
• Among the many ways of implementing the spiral
  model is the following
• Initial planning gather just enough information
  for a prototype
• The first prototype goes through the sequential
  path through analysis, design, construction,
  testing, integration with previous prototype
  components
• Planning for the next prototype
• Cycle of activities begins again
• Spiral model focuses on identifying risk factors

42
eXtreme Programming (XP)

• The XP (lightweight systems development) method
  adapts techniques from many sources and adds some
  new ideas
• Developers plan project by making users describe
  user stories
• User stories descriptions of the support users
  need from the system required functionalities
• The stories are documented quickly with informal
  descriptive models
• Developers plan a series of releases for the
  project, each including a working part of the
  final system
• The first release is worked, taking several
  iterations to complete before going to the next
  release

43
The Unified Process (UP)

• The UP is an object-oriented (OO) system
  development method offered by Rational software
  (recently purchased by IBM)
• The UP is the attempt by the three proponents
  (Grady Booch, James Rumbaugh, and Ivar Jacobson)
  of the Unified Modeling Language (UML) to define
  complete methodology that
• Provides several unique features
• Uses UML for systems modeling
• UP is not a standard OO development methodology
  yet but rapidly gaining attention because of the
  stature of the UML proponents

44
The Unified Process (UP)

• The UP is designed to reinforce six best
  practices of systems development common to many
  system development methodologies
• Develop iteratively
• Define and manage system requirements
• Use component architectures
• Create visual models
• Verify quality
• Control changes
• The UP defines four cycles
• Inception
• Elaboration
• Construction
• Transition

45
The Unified Process (UP)

• The inception phase defines the scope of the
  project by specifying use cases (similar to user
  stories) and completes a feasibility study
• The elaboration phase focuses on several
  iterations that take part of the system and
  define the requirements, design the solution, and
  implement the solution
• The construction phase focuses on building the
  system (possibly multiple releases) using
  additional iterations that include requirements,
  design, and implementations
• The transition phase involves turning the system
  over to the end users and focusing on end-user
  training, installation, and support

46
The Unified Process (UP)

• The four phases on the UP are different from the
  traditional SDLC
• Rather than defining the generic analysis,
  design, and implementation phases, the four
  phases of the UP define the project sequentially
  by indicating the emphasis on the project team at
  any point in time.

47
The Unified Process (UP)

• The UP defines the following disciplines within
  each phase to make iterative development
  manageable
• Business modeling
• Requirements modeling
• Design
• Implementation
• Testing
• Deployment
• Configuration and change management
• Project management
• Some of the defined roles of the UP include
• Designer
• User case specifier
• Systems analyst
• Implementer
• Architect

48
Agile Modeling

• The Agile Modeling (popularized by Scott Ambler)
  encourages developers to combine the best of XP
  with the best of the UP
• It increases the amount of modeling over XP but
  decreases the formality and documentation
  required of the UP
• It includes the following
• Iterative and incremental modeling applying the
  right models, creating several models in
  parallel, and modeling in small increments
• Teamwork model with others, get active
  stakeholder participation, encourage collective
  ownership, and display models quickly
• Simplicity create simple content, depict models
  simply, and use the simplest modeling tools
• Validation consider testability, then prove the
  model is right code

49
Tools to Support System Development

• Automated tools are needed to improve the speed
  and quality of systems development, irrespective
  of the methodology used
• CASE tools contain a database information about
  the project, called a repository which stores
  information about the system, including models,
  descriptions, and references that link the
  various models
• Upper CASE tools - provide suport for analysts
  during the analysis and design phases
• Lower CASE tools - provide support for
  implementation, primarily generating programs and
  database schemas based on specifications in the
  repository
• Integrated CASE (ICASE) tools - combine support
  for the full life cycle

50
Tools to Support System Development

• Below are sample CASE tools
• Microsoft Visio
• Visible Analyst
• Oracle Designer
• Rational Rose
• Together
• Embarcadero Describe
• Rational XDE Professional