Analysis and design of entreprise with uml


Published on

Published in: Technology, Education
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide
  • 1
  • 2
  • 3 Core Message: Modeling captures essential parts of the system Key Point 1: Computer system basically automate business processes. However, it’s not easy to build software systems on time and within budget. Key Point 2: Building a complex software system requires blueprint. You don’t construct a building without a blueprint. Visual modeling is the blueprint for software systems. Conclusion: VM is a key to successful software development
  • 4 Core Message: VM captures business process Key Point 1: Understanding business process is hard. If an architect and developers don’t understand the business process, which is the requirement for the system, you can’t build the proper system. Key Point 2: Use case is a technique to capture business process from user’s perspective. Use case is easy to understand because business process is defined in textual format, not in computer lingo. Example: Use the previous slide to describe use case. The system may handle order entry, inventory, and payroll processes. This particular use case only looks at order entry process. Conclusion: VM captures business process
  • 5 Core Message: VM is a communication tool. Key Point 1: Business analyst and domain experts define requirements. Software architects and developers build systems based on requirements. Typically, they have communication problems due to different use of terminology and different definition of concepts. Key Point 2: Take a look at the Rose development team. The team is distributed in three cities around the world; Sweden, Milwaukee, and Philadelphia. There is a common language - visual modeling. Key Point 3: With VM, there is a smooth transition between business domain and computer domain. Also, you can establish traceability from business domain to computer domain. Conclusion: VM is a communication tool.
  • 6 Core Message: VM manages complexity Key Point 1: This is a model of typical system. Sale has information about purchase order, sales person, and customer. In a system, you may get hundreds or thousands of these things (or objects). Key Point 2: Human mind can only handle 7 plus or minus things at once. VM allows you to raise your level of abstraction. There are constructs to group things into more manageable number of things. Example: One new developer joins the group. How do you describe your software system? Conclusion: VM manages complexity
  • 7 Core Message: VM defines software architecture Key Point 1:Model of a system is essentially a software architecture. You can then map it to the physical architecture. For example, this is a three tier architecture - mapping logical to physical. Key Point 2: Model your system independent of the implementation language. Technology change too much too fast. A few years ago, C++ was the language of the future but today it’s Java. How do you plan to change your system from PowerBuilder to Visual Basic? Reuse models. Conclusion: VM defines software architecture
  • 8 Core Message: VM promotes reuse. Key Point 1:On previous slide, we talked about reusing a model. Key Point 2:There is a higher leverage of reuse. Reusing parts of the system or an application. For example, a component is an application in a binary format, whether the component is made of objects or not. VM can be used as a component browser and it can also be used to model component assembly. Conclusion: VM promotes reuse.
  • 9 Core message -- second bullet UML can be used to communicate system and software design throughout the life cycle
  • 10 Walk the audience through the timeline. Point out that the UML is the natural successor to the notations. 1. Late ‘80s and early ‘90 - there are many (50+) OO methodologies 2. Among the first generation methodologies, Booch and OMT stood out 3. Around 1993, second generation methodologies came out - Booch ‘93 and OMT-II. Methodologist borrowed good concepts from each others so many concepts were the same across the methodologies, but different notations. 4. Oct. 1994 - Dr. James Rumbaugh joined Rational to unify Booch & OMT. 5. At OOPSLA ‘95, Grady and Jim announced Unified Method 0.8. 6. Use Case technique developed by Dr. Ivar Jacobson was adapted by all methodologies by then. 7. Rational acquires Objectory in fall of ‘95 - Dr. Ivar Jaconson joins Rational. 8. Jun of ‘96 - Rational submits UML 0.9 to OMG. 9. UML gains industry support from HP, Microsoft, Oracle + 16 others 10. UML is the defacto standard for OO and component technologies 11. The final submission goes in Sep. ‘97 - expect the announcement in Dec.
  • 11 Transition: So! Who should use UML? Everybody who is doing software development. Core Message: UML supports object and component-based technology. Key Point 1: UML is an expressive language that can be used to describe pretty much everything about software application. - object technology: objet, class, relationships, scenario, Use Case - component-based development: component, ActiveX/COM, CORBA - large scale system, application partitioning
  • 12 The last set of slides briefly introduces most of the UML notations. NOTE: Not all notations are covered due to time constraints
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44 Some managers have the perception that “iterative” really means “hacking” or not knowing what you’re doing or not planning anything. Nothing could be further from the truth.
  • 45 If anything, the iterative life cycle requires more planning and more thought all along the way. The difference is that instead of doing all of the detailed planning up front, it is done in a distributed fashion, a little at a time along the way, when enough information is available to make the planning worthwhile.
  • 46 These three characteristics are what make the iterative life cycle particularly attractive to managers. The iterative approach provides managers with more accurate information with which to plan and manage the project.
  • 47
  • 48 With a typical waterfall approach, no significant amounts of risk are removed until integration begins near the end of the life cycle. The highest (technical) risks are typically architectural: performance (speed, capacity, accuracy, etc.), system interface integrity, and system qualities (adaptability, portability, etc.). These cannot be verified until a significant portion of the system has been coded and integrated. With an iterative approach, the highest risk items are addressed early in the life cycle, thereby making significant reductions in the risk profile. From a management perspective, the level of confidence in the project plan should correspondingly increase significantly from iteration to iteration.
  • 49
  • 50
  • 51
  • 52 Each phase is divided into one or more iterations. Each iteration is defined in terms of the scenarios it implements. The classes needed to implement all selected scenarios are developed in a mini-waterfall development process.
  • 53 This is what one single iteration looks like. As with any waterfall representation of a process, in the mini-waterfall, many of these activities are occurring concurrently. For example, analysis is not done all in one continuous lump. And it intermingles with design and code, etc.
  • 54
  • 55
  • 56
  • 57
  • 58 Recall that packages are arbitrary groupings of classes and associations. Packages are defined for two reasons: 1. To address complexity. It is bigger than an individual class, but smaller than a system. Most systems are complex enough to need an intermediate-level “chunking” mechanism. 2. To delineate areas of responsibility, i.e., to assign ownership of portions of the system to developers.
  • 59 This is a critical step in making the iterative approach successful. If it is not done properly, the benefits of the iterative approach will be lessened. Also, remember that sometimes the right thing to do in this step is to revise the evaluation criteria rather than reworking the system. Sometimes, the benefit of the iteration is in revealing that a particular requirement is not important or is too expensive to implement or creates an unmaintainable architecture. In these cases, a cost/benefit analysis must be done. The decision is a business decision.
  • 60 There is no “right” answer relative to how many iterations are required. It depends on lots of factors such as the overall project length and the degree of risk involved. For most projects, the answer usually ends up being somewhere between 3 and 6 iterations.
  • 61 A common pitfall is to be too optimistic about how much can be achieved in the first iteration. Since the entire development environment and the entire development team have to be in place and operating effectively for the first iteration to complete, it is wise to be conservative about how much of the system can be completed. The team will be distracted by issues like how to work with a new compiler, how to implement the CM policy, how to get the CM tool tailored for the project, how to put regression testing in place, and all the million other things that must be addressed on the first iteration. For the second iteration, all these things should be in place and everyone will have experience with how to develop the release, so they can attempt larger amounts of functionality.
  • 62 Adopting the iterative approach will not make all your problems go away. It will give you more effective techniques for dealing with them.
  • 63
  • Analysis and design of entreprise with uml

    1. 1. Analysis and Design with UML
    2. 2. Agenda <ul><li>Benefits of Visual Modeling </li></ul><ul><li>History of the UML </li></ul><ul><li>Visual Modeling with UML </li></ul><ul><li>The Rational Iterative Development Process </li></ul>
    3. 3. What is Visual Modeling? “ Modeling captures essential parts of the system.” Dr. James Rumbaugh Visual Modeling is modeling using standard graphical notations Computer System Business Process Order Item Ship via
    4. 4. Visual Modeling Captures Business Process Use Case Analysis is a technique to capture business process from user’s perspective
    5. 5. Visual Modeling is a Communication Tool Use visual modeling to capture business objects and logic Use visual modeling to analyze and design your application
    6. 6. Visual Modeling Manages Complexity
    7. 7. Visual Modeling Defines Software Architecture Model your system independent of implementation language User Interface (Visual Basic, Java) Business Logic (C++, Java) Database Server (C++ & SQL)
    8. 8. Visual Modeling Promotes Reuse Multiple Systems Reusable Components
    9. 9. What is the UML? <ul><li>UML stands for Unified Modeling Language </li></ul><ul><li>The UML combines the best of the best from </li></ul><ul><ul><li>Data Modeling concepts (Entity Relationship Diagrams) </li></ul></ul><ul><ul><li>Business Modeling (work flow) </li></ul></ul><ul><ul><li>Object Modeling </li></ul></ul><ul><ul><li>Component Modeling </li></ul></ul><ul><li>The UML is the standard language for visualizing, specifying, constructing, and documenting the artifacts of a software-intensive system </li></ul><ul><li>It can be used with all processes, throughout the development life cycle, and across different implementation technologies </li></ul>
    10. 10. History of the UML Nov ‘97 UML approved by the OMG
    11. 11. UML Supports Application Development Classes application partitioning Business Objects Relationships Business Process Objects Use Cases large scale system Scenarios Components Microsoft ActiveX/COM Microsoft ORDBMS Oracle CORBA OMG
    12. 12. UML Concepts <ul><li>The UML may be used to: </li></ul><ul><ul><li>Display the boundary of a system & its major functions using use cases and actors </li></ul></ul><ul><ul><li>Illustrate use case realizations with interaction diagrams </li></ul></ul><ul><ul><li>Represent a static structure of a system using class diagrams </li></ul></ul><ul><ul><li>Model the behavior of objects with state transition diagrams </li></ul></ul><ul><ul><li>Reveal the physical implementation architecture with component & deployment diagrams </li></ul></ul><ul><ul><li>Extend your functionality with stereotypes </li></ul></ul>
    13. 13. Putting the UML to Work <ul><li>The ESU University wants to computerize their registration system </li></ul><ul><ul><li>The Registrar sets up the curriculum for a semester </li></ul></ul><ul><ul><ul><li>One course may have multiple course offerings </li></ul></ul></ul><ul><ul><li>Students select 4 primary courses and 2 alternate courses </li></ul></ul><ul><ul><li>Once a student registers for a semester, the billing system is notified so the student may be billed for the semester </li></ul></ul><ul><ul><li>Students may use the system to add/drop courses for a period of time after registration </li></ul></ul><ul><ul><li>Professors use the system to receive their course offering rosters </li></ul></ul><ul><ul><li>Users of the registration system are assigned passwords which are used at logon validation </li></ul></ul>
    14. 14. Actors <ul><li>An actor is someone or some thing that must interact with the system under development </li></ul>Student Registrar Professor Billing System
    15. 15. Use Cases <ul><li>A use case is a pattern of behavior the system exhibits </li></ul><ul><ul><li>Each use case is a sequence of related transactions performed by an actor and the system in a dialogue </li></ul></ul><ul><li>Actors are examined to determine their needs </li></ul><ul><ul><li>Registrar -- maintain the curriculum </li></ul></ul><ul><ul><li>Professor -- request roster </li></ul></ul><ul><ul><li>Student -- maintain schedule </li></ul></ul><ul><ul><li>Billing System -- receive billing information from registration </li></ul></ul>Maintain Schedule Maintain Curriculum Request Course Roster
    16. 16. Documenting Use Cases <ul><li>A flow of events document is created for each use cases </li></ul><ul><ul><li>Written from an actor point of view </li></ul></ul><ul><li>Details what the system must provide to the actor when the use cases is executed </li></ul><ul><li>Typical contents </li></ul><ul><ul><li>How the use case starts and ends </li></ul></ul><ul><ul><li>Normal flow of events </li></ul></ul><ul><ul><li>Alternate flow of events </li></ul></ul><ul><ul><li>Exceptional flow of events </li></ul></ul>
    17. 17. Maintain Curriculum Flow of Events <ul><li>This use case begins when the Registrar logs onto the Registration System and enters his/her password. The system verifies that the password is valid (E-1) and prompts the Registrar to select the current semester or a future semester (E-2). The Registrar enters the desired semester. The system prompts the professor to select the desired activity: ADD, DELETE, REVIEW, or QUIT. </li></ul><ul><li>If the activity selected is ADD, the S-1: Add a Course subflow is performed. </li></ul><ul><li>If the activity selected is DELETE, the S-2: Delete a Course subflow is performed. </li></ul><ul><li>If the activity selected is REVIEW, the S-3: Review Curriculum subflow is performed. </li></ul><ul><li>If the activity selected is QUIT, the use case ends. </li></ul><ul><li>... </li></ul>
    18. 18. Use Case Diagram <ul><li>Use case diagrams are created to visualize the relationships between actors and use cases </li></ul>Student Registrar Professor Maintain Schedule Maintain Curriculum Request Course Roster Billing System
    19. 19. Uses and Extends Use Case Relationships <ul><li>As the use cases are documented, other use case relationships may be discovered </li></ul><ul><ul><li>A uses relationship shows behavior that is common to one or more use cases </li></ul></ul><ul><ul><li>An extends relationship shows optional behavior </li></ul></ul>Register for courses <<uses>> Logon validation <<uses>> Maintain curriculum
    20. 20. Use Case Realizations <ul><li>The use case diagram presents an outside view of the system </li></ul><ul><li>Interaction diagrams describe how use cases are realized as interactions among societies of objects </li></ul><ul><li>Two types of interaction diagrams </li></ul><ul><ul><li>Sequence diagrams </li></ul></ul><ul><ul><li>Collaboration diagrams </li></ul></ul>
    21. 21. Sequence Diagram <ul><li>A sequence diagram displays object interactions arranged in a time sequence </li></ul>: Student registration form registration manager math 101 1: fill in info 2: submit 3: add course(joe, math 01) 4: are you open? 5: are you open? 6: add (joe) 7: add (joe) math 101 section 1
    22. 22. Collaboration Diagram <ul><li>A collaboration diagram displays object interactions organized around objects and their links to one another </li></ul>: Registrar course form : CourseForm theManager : CurriculumManager aCourse : Course 1: set course info 2: process 3: add course 4: new course
    23. 23. Class Diagrams <ul><li>A class diagram shows the existence of classes and their relationships in the logical view of a system </li></ul><ul><li>UML modeling elements in class diagrams </li></ul><ul><ul><li>Classes and their structure and behavior </li></ul></ul><ul><ul><li>Association, aggregation, dependency, and inheritance relationships </li></ul></ul><ul><ul><li>Multiplicity and navigation indicators </li></ul></ul><ul><ul><li>Role names </li></ul></ul>
    24. 24. Classes <ul><li>A class is a collection of objects with common structure, common behavior, common relationships and common semantics </li></ul><ul><li>Classes are found by examining the objects in sequence and collaboration diagram </li></ul><ul><li>A class is drawn as a rectangle with three compartments </li></ul><ul><li>Classes should be named using the vocabulary of the domain </li></ul><ul><ul><li>Naming standards should be created </li></ul></ul><ul><ul><li>e.g., all classes are singular nouns starting with a capital letter </li></ul></ul>
    25. 25. Classes RegistrationForm RegistrationManager Course Student CourseOffering Professor ScheduleAlgorithm
    26. 26. Operations <ul><li>The behavior of a class is represented by its operations </li></ul><ul><li>Operations may be found by examining interaction diagrams </li></ul>registration form registration manager 3: add course(joe, math 01) RegistrationManager addCourse(Student,Course)
    27. 27. Attributes <ul><li>The structure of a class is represented by its attributes </li></ul><ul><li>Attributes may be found by examining class definitions, the problem requirements, and by applying domain knowledge </li></ul>Each course offering has a number, location and time CourseOffering number location time
    28. 28. Classes RegistrationForm RegistrationManager addStudent(Course, StudentInfo) Course name numberCredits open() addStudent(StudentInfo) Student name major CourseOffering location open() addStudent(StudentInfo) Professor name tenureStatus ScheduleAlgorithm
    29. 29. Relationships <ul><li>Relationships provide a pathway for communication between objects </li></ul><ul><li>Sequence and/or collaboration diagrams are examined to determine what links between objects need to exist to accomplish the behavior -- if two objects need to “talk” there must be a link between them </li></ul><ul><li>Three types of relationships are: </li></ul><ul><ul><li>Association </li></ul></ul><ul><ul><li>Aggregation </li></ul></ul><ul><ul><li>Dependency </li></ul></ul>
    30. 30. Relationships <ul><li>An association is a bi-directional connection between classes </li></ul><ul><ul><li>An association is shown as a line connecting the related classes </li></ul></ul><ul><li>An aggregation is a stronger form of relationship where the relationship is between a whole and its parts </li></ul><ul><ul><li>An aggregation is shown as a line connecting the related classes with a diamond next to the class representing the whole </li></ul></ul><ul><li>A dependency relationship is a weaker form of relationship showing a relationship between a client and a supplier where the client does not have semantic knowledge of the supplier </li></ul><ul><li>A dependency is shown as a dashed line pointing from the client to the supplier </li></ul>
    31. 31. Finding Relationships <ul><li>Relationships are discovered by examining interaction diagrams </li></ul><ul><ul><li>If two objects must “talk” there must be a pathway for communication </li></ul></ul>Registration Manager Math 101: Course 3: add student(joe) RegistrationManager Course
    32. 32. Relationships RegistrationForm RegistrationManager Course Student CourseOffering Professor addStudent(Course, StudentInfo) name numberCredits open() addStudent(StudentInfo) name major location open() addStudent(StudentInfo) name tenureStatus ScheduleAlgorithm
    33. 33. Multiplicity and Navigation <ul><li>Multiplicity defines how many objects participate in a relationships </li></ul><ul><ul><li>Multiplicity is the number of instances of one class related to ONE instance of the other class </li></ul></ul><ul><ul><li>For each association and aggregation, there are two multiplicity decisions to make: one for each end of the relationship </li></ul></ul><ul><li>Although associations and aggregations are bi-directional by default, it is often desirable to restrict navigation to one direction </li></ul><ul><li>If navigation is restricted, an arrowhead is added to indicate the direction of the navigation </li></ul>
    34. 34. Multiplicity and Navigation RegistrationForm RegistrationManager Course Student CourseOffering Professor addStudent(Course, StudentInfo) name numberCredits open() addStudent(StudentInfo) major location open() addStudent(StudentInfo) tenureStatus ScheduleAlgorithm 1 0..* 0..* 1 1 1..* 4 3..10 0..4 1
    35. 35. Inheritance <ul><li>Inheritance is a relationships between a superclass and its subclasses </li></ul><ul><li>There are two ways to find inheritance: </li></ul><ul><ul><li>Generalization </li></ul></ul><ul><ul><li>Specialization </li></ul></ul><ul><li>Common attributes, operations, and/or relationships are shown at the highest applicable level in the hierarchy </li></ul>
    36. 36. Inheritance RegistrationForm RegistrationManager Course Student CourseOffering Professor addStudent(Course, StudentInfo) name numberCredits open() addStudent(StudentInfo) major location open() addStudent(StudentInfo) tenureStatus ScheduleAlgorithm name RegistrationUser
    37. 37. The State of an Object <ul><li>A state transition diagram shows </li></ul><ul><ul><li>The life history of a given class </li></ul></ul><ul><ul><li>The events that cause a transition from one state to another </li></ul></ul><ul><ul><li>The actions that result from a state change </li></ul></ul><ul><li>State transition diagrams are created for objects with significant dynamic behavior </li></ul>
    38. 38. State Transition Diagram entry: Register student exit: Increment count Initialization Open Closed Canceled do: Initialize course do: Finalize course do: Notify registered students Add Student / Set count = 0 Add student[ count < 10 ] [ count = 10 ] Cancel Cancel Cancel
    39. 39. The Physical World <ul><li>Component diagrams illustrate the organizations and dependencies among software components </li></ul><ul><li>A component may be </li></ul><ul><ul><li>A source code component </li></ul></ul><ul><ul><li>A run time components or </li></ul></ul><ul><ul><li>An executable component </li></ul></ul>
    40. 40. Component Diagram Course Course Offering Student Professor Course.dll People.dll Course User Register.exe Billing.exe Billing System
    41. 41. Deploying the System <ul><li>The deployment diagram shows the configuration of run-time processing elements and the software processes living on them </li></ul><ul><li>The deployment diagram visualizes the distribution of components across the enterprise. </li></ul>
    42. 42. Deployment Diagram Registration Database Library Dorm Main Building
    43. 43. Extending the UML <ul><li>Stereotypes can be used to extend the UML notational elements </li></ul><ul><li>Stereotypes may be used to classify and extend associations, inheritance relationships, classes, and components </li></ul><ul><li>Examples: </li></ul><ul><ul><li>Class stereotypes: boundary, control, entity, utility, exception </li></ul></ul><ul><ul><li>Inheritance stereotypes: uses and extends </li></ul></ul><ul><ul><li>Component stereotypes: subsystem </li></ul></ul>
    44. 44. What the Iterative Life Cycle Is Not <ul><li>It is not hacking </li></ul><ul><li>It is not a playpen for developers </li></ul><ul><li>It is not unpredictable </li></ul><ul><li>It is not redesigning the same thing over and over until it is perfect </li></ul><ul><li>It is not an excuse for not planning and managing a project </li></ul><ul><li>It is not something that affects only the developers on a project </li></ul>
    45. 45. What the Iterative Life Cycle Is <ul><li>It is planned and managed </li></ul><ul><li>It is predictable </li></ul><ul><li>It accommodates changes to requirements with less disruption </li></ul><ul><li>It is based on evolving executable prototypes, not documentation </li></ul><ul><li>It involves the user/customer throughout the process </li></ul><ul><li>It is risk driven </li></ul>
    46. 46. Three Important Features of the Iterative Approach <ul><li>Continuous integration </li></ul><ul><ul><li>Not done in one lump near the delivery date </li></ul></ul><ul><li>Frequent, executable releases </li></ul><ul><ul><li>Some internal; some delivered </li></ul></ul><ul><li>Attack risks through demonstrable progress </li></ul><ul><ul><li>Progress measured in products, not documentation or engineering estimates </li></ul></ul>
    47. 47. Resulting Benefits <ul><li>Releases are a forcing function that drives the development team to closure at regular intervals </li></ul><ul><ul><li>Cannot have the “90% done with 90% remaining” phenomenon </li></ul></ul><ul><li>Can incorporate problems/issues/changes into future iterations rather than disrupting ongoing production </li></ul><ul><li>The project’s supporting elements (testers, writers, toolsmiths, CM, QA, etc.) can better schedule their work </li></ul>
    48. 48. Risk Profile of an Iterative Development Risk Preliminary Iteration Architect. Iteration Architect. Iteration Devel. Iteration Devel. Iteration Devel. Iteration Transition Iteration Transition Iteration Post- deployment Time Transition Inception Elaboration Construction Waterfall
    49. 49. Risk Management Phase-by-Phase <ul><li>Inception </li></ul><ul><ul><li>Bracket the project’s risks by building a proof of concept </li></ul></ul><ul><li>Elaboration </li></ul><ul><ul><li>Develop a common understanding of the system’s scope and desired behavior by exploring scenarios with end users and domain experts </li></ul></ul><ul><ul><li>Establish the system’s architecture </li></ul></ul><ul><ul><li>Design common mechanisms to address system-wide issues </li></ul></ul>
    50. 50. Risk Management Phase-by-Phase (cont.) <ul><li>Construction </li></ul><ul><ul><li>Refine the architecture </li></ul></ul><ul><ul><li>Risk-driven iterations </li></ul></ul><ul><ul><li>Continuous integration </li></ul></ul><ul><li>Transition </li></ul><ul><ul><li>Facilitate user acceptance </li></ul></ul><ul><ul><li>Measure user satisfaction </li></ul></ul><ul><li>Post-deployment cycles </li></ul><ul><ul><li>Continue evolutionary approach </li></ul></ul><ul><ul><li>Preserve architectural integrity </li></ul></ul>
    51. 51. Risk Reduction Drives Iterations Initial Project Risks Initial Project Scope <ul><li>Revise Overall </li></ul><ul><li>Project Plan </li></ul><ul><li>Cost </li></ul><ul><li>Schedule </li></ul><ul><li>Scope/Content </li></ul><ul><li>Plan Iteration N </li></ul><ul><li>Cost </li></ul><ul><li>Schedule </li></ul>Assess Iteration N Risks Eliminated <ul><li>Revise Project Risks </li></ul><ul><li>Reprioritize </li></ul><ul><li>Develop Iteration N </li></ul><ul><li>Collect cost and quality metrics </li></ul>Define scenarios to address highest risks Iteration N
    52. 52. Use Cases Drive the Iteration Process Iteration Planning Rqmts Capture Analysis & Design Implementation Test Prepare Release Inception Elaboration Construction Transition Iteration 1 Iteration 2 Iteration 3 “ Mini-Waterfall” Process
    53. 53. The Iteration Life Cycle: A Mini-Waterfall <ul><li>Results of previous iterations </li></ul><ul><li>Up-to-date risk assessment </li></ul><ul><li>Controlled libraries of models, code, and tests </li></ul>Release description Updated risk assessment Controlled libraries Iteration Planning Requirements Capture Analysis & Design Implementation Test Prepare Release Selected scenarios
    54. 54. Detailed Iteration Life Cycle Activities <ul><li>Iteration planning </li></ul><ul><ul><li>Before the iteration begins, the general objectives of the iteration should be established based on </li></ul></ul><ul><ul><ul><li>Results of previous iterations ( if any) </li></ul></ul></ul><ul><ul><ul><li>Up-to-date risk assessment for the project </li></ul></ul></ul><ul><ul><li>Determine the evaluation criteria for this iteration </li></ul></ul><ul><ul><li>Prepare detailed iteration plan for inclusion in the development plan </li></ul></ul><ul><ul><ul><li>Include intermediate milestones to monitor progress </li></ul></ul></ul><ul><ul><ul><li>Include walkthroughs and reviews </li></ul></ul></ul>
    55. 55. Detailed Iteration Life Cycle Activities (cont.) <ul><li>Requirements Capture </li></ul><ul><ul><li>Select/define the use cases to be implemented in this iteration </li></ul></ul><ul><ul><li>Update the object model to reflect additional domain classes and associations discovered </li></ul></ul><ul><ul><li>Develop a test plan for the iteration </li></ul></ul>
    56. 56. Detailed Iteration Life Cycle Activities (cont.) <ul><li>Analysis & Design </li></ul><ul><ul><li>Determine the classes to be developed or updated in this iteration </li></ul></ul><ul><ul><li>Update the object model to reflect additional design classes and associations discovered </li></ul></ul><ul><ul><li>Update the architecture document if needed </li></ul></ul><ul><ul><li>Begin development of test procedures </li></ul></ul><ul><li>Implementation </li></ul><ul><ul><li>Automatically generate code from the design model </li></ul></ul><ul><ul><li>Manually generate code for operations </li></ul></ul><ul><ul><li>Complete test procedures </li></ul></ul><ul><ul><li>Conduct unit and integration tests </li></ul></ul>
    57. 57. Detailed Iteration Life Cycle Activities (cont.) <ul><li>Test </li></ul><ul><ul><li>Integrate and test the developed code with the rest of the system (previous releases) </li></ul></ul><ul><ul><li>Capture and review test results </li></ul></ul><ul><ul><li>Evaluate test results relative to the evaluation criteria </li></ul></ul><ul><ul><li>Conduct an iteration assessment </li></ul></ul><ul><li>Prepare the release description </li></ul><ul><ul><li>Synchronize code and design models </li></ul></ul><ul><ul><li>Place products of the iteration in controlled libraries </li></ul></ul>
    58. 58. Work Allocation Within an Iteration <ul><li>Work to be accomplished within an iteration is determined by </li></ul><ul><ul><li>The (new) use cases to be implemented </li></ul></ul><ul><ul><li>The rework to be done </li></ul></ul><ul><li>Packages make convenient work packages for developers </li></ul><ul><ul><li>High-level packages can be assigned to teams </li></ul></ul><ul><ul><li>Lower-level packages can be assigned to individual developers </li></ul></ul><ul><li>Use Cases make convenient work packages for test and assessment teams </li></ul><ul><li>Packages are also useful in determining the granularity at which configuration management will be applied </li></ul><ul><ul><li>For example, check-in and check-out of individual packages </li></ul></ul>
    59. 59. Iteration Assessment <ul><li>Assess iteration results relative to the evaluation criteria established during iteration planning: </li></ul><ul><ul><li>Functionality </li></ul></ul><ul><ul><li>Performance </li></ul></ul><ul><ul><li>Capacity </li></ul></ul><ul><ul><li>Quality measures </li></ul></ul><ul><li>Consider external changes that have occurred during this iteration </li></ul><ul><ul><li>For example, changes to requirements, user needs, competitor’s plans </li></ul></ul><ul><li>Determine what rework, if any, is required and assign it to the remaining iterations </li></ul>
    60. 60. Selecting Iterations <ul><li>How many iterations do I need? </li></ul><ul><ul><li>On projects taking 18 months or less, 3 to 6 iterations are typical </li></ul></ul><ul><li>Are all iterations on a project the same length? </li></ul><ul><ul><li>Usually </li></ul></ul><ul><ul><li>Iteration length may vary by phase. For example, elaboration iterations may be shorter than construction iterations </li></ul></ul>
    61. 61. The First Iteration <ul><li>The first iteration is usually the hardest </li></ul><ul><ul><li>Requires the entire development environment and most of the development team to be in place </li></ul></ul><ul><ul><li>Many tool integration issues, team-building issues, staffing issues, etc. must be resolved </li></ul></ul><ul><li>Teams new to an iterative approach are usually overly-optimistic </li></ul><ul><li>Be modest regarding the amount of functionality that can be achieved in the first iteration </li></ul><ul><ul><li>Otherwise, completion of the first iteration will be delayed, </li></ul></ul><ul><ul><li>The total number of iterations reduced, and </li></ul></ul><ul><ul><li>The benefits of an iterative approach reduced </li></ul></ul>
    62. 62. There Is No Silver Bullet <ul><li>Remember the main reason for using the iterative life cycle: </li></ul><ul><ul><li>You do not have all the information you need up front </li></ul></ul><ul><ul><li>Things will change during the development period </li></ul></ul><ul><li>You must expect that </li></ul><ul><ul><li>Some risks will not be eliminated as planned </li></ul></ul><ul><ul><li>You will discover new risks along the way </li></ul></ul><ul><ul><li>Some rework will be required; some lines of code developed for an iteration will be thrown away </li></ul></ul><ul><ul><li>Requirements will change along the way </li></ul></ul>
    63. 63. Copyright © 1997 by Rational Software Corporation