The Experience Factory

1
The Experience Factory

• May 2004 Leonardo Vaccaro

2
Introduction

• Reuse of products, processes and experience
• originating from the system life cycle is seen
  today
• as a feasible solution to the problem of
  developing
• higher quality system at lower cost.

3
Outline

• Importance of Experience Factory in the Software
  Business
• The Experience Factory
• Software Engineering Laboratory
• Q-Labs Experience Management System
• Experience Factory Implications

4
Experience Factory(Definition)

• It is a logical and/or physical organization that
  supports project developments by analyzing and
  synthesizing all kinds of experience, acting as a
  repository for such experience, and supplying
  that experience to various project on demand

5
Importance of Experience Factory in the software
business

• The software business requires understanding,
  continuous
• improvement, and the packaging of experience for
  reuse.
• Reusing experience in the form of processes,
  products and other forms of knowledge is
  essential for improvement
• Experience needs to be packaged
• Experience must be evaluated
• Software development must support reuse of
  experience
• Packaged experiences need to be integrated

6
Experiences Factory Schema(Victor Basili)

• Envioronment
• Characteristics

• Characterize
• Set Goals
• Choose Process

• Project
• Support

• Package

• Goals, Processes,
• Tools, Products,
• Resource Models...

• Execution
• Plans

• Generalize

• EXPERIENCE
• BASE

• Tailor

• Data, Lessons
• Learned

• Execute Process

• Formalize

• Project
• Analysis

• Analyze

• Experience Factory Organization

• Project Organization

7
Project Organization

• Produce and maintain software
• Provides the analysis organization with
• project and environment characteristics
• development data
• resource usage information
• process information

8
Quality Improvement Paradigm

• The basic methodological device for the
  Experience Factory
• Developed by Victor Basili
• Is the result of the application of the
  scientific method to the problem of software
  quality improvement
• Based upon the notion that improving the software
  process requires the continual accumulation of
  evaluated experiences (learning) in a form that
  can be effectively understood and modified
  (experience models) into a repository of
  integrated experience models (experience base)
  that can be accessed and modified to meet the
  needs of the current project (reuse)

9
Quality Improvement Paradigm (cont.)

• It is articulated in six steps

• 1. Characterize

• 6. Package

• 5. Analyze

• 2. Set Goals

• 3. Choose Process

• 4. Execute

10
1. Characterize

• Understand the environment based upon
• Available models
• Data
• Intuition
• Establish baselines with the existing business
  processes in the organization and characterize
  their criticality

11
2. Set Goals

• Set quantifiable goals for successful project and
  organization performance and improvement.
• Use of Goal \ Question \ Metric Paradigm
• The reasonable expectation are defined upon the
  baseline provided by the characterization step

12
3. Choose Process

• Choose the appropriate processes for
• Improvement
• Support methods and tools
• The processes must be consistent with the goals
  that have been set

13
4. Execute

• Perform the processes
• Constructing the products
• Providing project feedback based upon the data on
  goal achievement that are being collected

14
5. Analyze

• At the end of each specific project, analyze the
  data and the information gathered to
• Evaluate the current practices
• Determine the problems
• Make recommendation for future project
  improvements.

15
Analysis Organization

• Processes the information received from the
  development organization
• Returns direct feedback to each project, together
  with goals and models tailored from similar
  project.

16
6. Package

• Consolidate the experience gained in the form of
  new, or updated models
• Store it in a experience base so it is available
  for future projects.

17
Support Organization

• Sustains and facilitates the interaction between
  developers and analysts
• Saving and maintaining the information
• Making it efficiently retrievable
• Controlling and monitoring the access to it.

18
Experience Base

• Contains an accessible and integrated set of
  analyzed, synthesized and packaged experience
  models that capture past experience

19
Packaged Experience

• The experience factory can package all kinds of
  experience
• Resource models
• Change and defect models
• Process definition and models
• Product models
• A variety of quality models

20
Experience Package

• Is the main product of the experience factory
• The content of this product vary upon the kind of
  experience clustered in the package
• There is a central element that determines what
  the package is.

21
Examples of Experience Package

• Product Packages
• Processes Package
• Relationship Packages
• Tool Packages
• Management Packages
• Data Packages

22
Product Packages

• Have as their central element a product,
  clustered with the information needed to reuse it
  and the lessons learned in reusing it
• Examples
• Programs
• Architectures
• Designs

23
Processes Packages

• Have as their central element a process,
  clustered with the information needed to execute
  it and lessons learned in executing it.
• Examples
• Processes model
• Methods

24
Relationship Packages

• Have as their central element a relationship or a
  system of relationship among observable
  characteristics of a software project. These
  packages are used for analysis and/or forecast of
  relevant phenomena
• Examples
• Cost and defect models,
• Resource models

25
Tool Packages

• Have as their central element a specific tool
  either constructive or analytic.
• Examples
• Code generator (constructive)
• Static analyzer (analytic)

26
Management Packages

• Have as their central element any container of
  reference information for project management.
• Examples
• Management handbooks,
• Decision support models

27
Data Packages

• Have as their central element a collection of
  defined and validated data relevant for a
  software project or for activities within it.
• Examples
• Project databases
• Quality records

28
Goal \ Question \ Metric Paradigm

• The mechanism used by the Quality Improvement
  Paradigm for defining and evaluating a set of
  operational goals using measurement.
• Represents a systematic approach for tailoring
  and integrating goals with models of the software
  processes.

29
Goal \ Question \ Metric Paradigm

• Development of goals
• Generation of questions that define the goals
• Indication of metrics that answer the question
• Objective
• Minimize the number of defects

• Management
• Goals or Needs

• Goals

• Questions

• Metrics

30
Software Engineering Laboratory

• A software organization that has for a long time
  recognized the value of accumulation and reuse of
  experience is the NASA Goddard Space Flight
  Centre which has developed since 1977 the
  Software Engineering Laboratory

31
Software Engineering Laboratory

• Is a very advanced examples of the concept of
  Experience Factory.
• The experience packages developed by the Software
  Engineering Laboratory have mainly focused on
  project management and control, acquisition and
  tailoring of new technologies for software
  development and maintenance

32
The Software Engineering Laboratory Organization
33
Software Management Environment

• The most interesting experience packages
  developed by Software Engineering Laboratory
• Is a set of data, tools manuals and analysis
  techniques supplied to the project management in
  order to
• control the execution of project,
• compare it with similar ones
• detect and analyze problems,
• identify solutions

34
Q-Labs

• An internationally distributed software
  engineering consulting company with office in
  Europe as well as in the United States
• Needed an infrastructure to share experience
  between employees no matter where they are
  located
• Q-Labs Experience Management System

35
Experience Management System

• A physical implementation of the Experience
  Factory
• Composed of
• Content
• Data, Information, Experience
• Structure
• The way the content is organized
• Procedures
• Instructions on how to use, package, delete and
  update the experience
• Tools
• Visual Query Interface

36
Q-Labs Experience Management System

• Stores documents and their description in the
  Experience Base
• The search is performed using a tool called
  Visual Query Interface
• Experience packages in the context of Q-Labs are
  single computer document
• Texts
• Slides
• Graphs
• Scanned images
• Numeric data

37
Experience Factory Implications

• Offers an organizational structure that
• Separates the product development focus from the
  learning and reuse focus.
• Supports learning and reuse
• Generates a tangible corporate asset in the form
  of packaged experiences.
• There are costs involved in instituting such a
  program.
• Based upon the SEL experience where a full
  measurement program has been in progress for over
  14 years, project data collection overhead is
  estimated to be about 5 of the total project
  cost.

38
References

• The Experience Factory Victor Basili,
  Gianluigi Caldiera, Dieter Rombach
• http//wwwagse.informatik.uni-kl.de/pubs/rep
  ository/basili94c/encyclo.ef.pdf
• Implementing the Experience Factory concepts
  Victor Basili, Mikael Lindvall, Patricia Costa
  http//www.cebase.org/www/researchActivities/eBase
  /websepp.pdf
• Software Engineering Laboratory
  http//sel.gsfc.nasa.gov/website/welcome.htm
• Lessons Learned about Structuring and Describing
  Experience for Three Experience Base Mikael
  Lindvall, Michael Frey, Patricia Costa, Roseanne
  Tesoriero

39
Visual Query Interface

• Visualize the content of the Experience Base
• The attributes are used to describe the
  experience packages
• Package Id
• Name
• Submitted
• X axis represents the Open Date
• Y axis represents The Person Responsible
• The color represents the Costumer