PhD Candidate: Timo LAUDAN, EADS Innovation Works

1
Context-Oriented Product Development
Collaboration between the Business and
Engineering DomainAn Investigation with a Focus
on Project- and Engineering-Based
Organisations Defence 14th November 2008,
Grenoble

• PhD Candidate Timo LAUDAN, EADS Innovation
  Works
• President Prof. Manfred BROY, Technical
  University of Munich
• Reviewers Prof. Benoit EYNARD, Université de
  Technologie Compiègne Prof. Pascal Le
  MASSON, Ecole des Mines de Paris
• Industrial Supervisors Philippe HOMSI, Airbus
  SAS (Unfortunately not here today)
• Dr. Michel DUREIGNE, EADS Innovation Works
• Axel MAURITZ, EADS Innovation Works
• Scientific Supervisors Prof. Michel TOLLENAERE,
  University of Grenoble, G-SCOP
• Prof. Mickaël GARDONI, University of
  Strasbourg (INSA), LGeCo

2
Presentation Outline
1
2
3
4
IndustrialApplicationCases
ResearchContext
ConceptualSolution
IndustryObservation
Results Perspectives
5
3
Presentation Outline
gtgt Establish the research question and conduction
of research proposal, synopsis on reviewed
theories
1

1. Research Organisation
2. Research Scope
3. Research Opportunity
4. Research Question
5. Research Methodology
6. Theory - Reviewed Concepts

ResearchContext
4
Research Organisation

• Working Environment
• EADS Innovation Works
• Technical Capability Centre Simulation, IT and
  Systems Engineering
• Team Systems Engineering lead by Axel MAURITZ
• Doctoral Contract since 2004 Since May 2008
  employee
• Supervising Universities
• Grenoble, G-SCOP Prof. Michel Tollenaere
• Strasbourg, LGeCo Prof. Mickaël Gardoni
• Sponsor of the Thesis
• Funded by Airbus Philippe Homsi, Head of
  Modelling and Simulation
• Empirical Study Environment
• European Integrated Project (FP6) that integrates
  major aeronautical key-players
• Project task Formalise and harmonise aero
  industries business intents and engineering
  definitions
• Empirical study Frame to study challenges on the
  scale of a collaborative project

5
Research Scope

• Collaboration between Business and Product
  Development (PD) Teams establishing the top-level
  product definition (requirement analysis)
• in context Project-and engineering-based
  Organisation

Business
BusinessIntents
Top-levelProductDefinition
PD Teams
Did you get the message?
ProductRequirements
Understand related works and challenges to put
forward for research opportunity
6
Research Opportunity (1/2)

• Context
• Traditional requirements analysis models focused
  on system- and user interactions
• Much efforts in particular industrial - on
  logical breakdown, management, dissemination and
  proof on the level of implementation of
  requirements without being strongly connected to
  the organisation and its business intents
• Related Work
• Focus on front-end negotiations early
  requirements analysis activities concerned with
  reconciling business problems, opportunities and
  product (high-level) requirements
• Intentional modelling semi-formal and formal
  approaches using the concept of goals to develop
  coherent requirement models that aims at
  increasing rationalisation and confidence in
  engineering definitions

7
Research Opportunity (2/2)

• Challenges - Literature shows that
• Coordination and communication gap in project-and
  engineering based organisations, especially
  between stakeholders and developers (Karlson et
  al. 2007)
• Lack of means that enables to perform stakeholder
  cooperation within the product development
  process (Kavakli/Loucopoulos 2003)
• Problem of supporting cross-community in
  knowledge creation and sharing is relatively
  under-investigated (Novak/Wurst 2004)
• Formalisms in early requirements analysis fall
  short in establishing usable intentional
  structures to be used non-experts (Lamsweerde
  2004)

Research Opportunity Communication, coordination
and knowledge representation in cross-community
constellations in the phase of requirements
analysis
8
Research QuestionProblematic

• Research Question
• How to organise collaboration and knowledge
  conversion between business management and PD
  Teams concerned with the elaboration of top-level
  product requirements?
• In context of the research question
• How to find coherence (adequacy, completeness and
  consistency) within the evolution of the
  projects product on the level of business and
  engineering?
• How to maintain and trace knowledge evolutions in
  context of the projects product between business
  and engineering?

9
Research Methodology
Action Research - Coupling both research and
action A closed loop of inductive and deductive
reasoning, i.e. a cognitive shift from research
design to design practice and conversely using
synergies of three channels of cognition
(spontaneous observations, organized
observations, and experimentations) and by this
activate a hybrid form of action research.
10
Theory (1/3) Reviewed Concepts to Study
Collaboration and Knowledge Conversion in
Cross-Communities Key Features

• Collaboration To understand principles of social
  interaction
• Two individuals or larger collectives of
  individuals (communities) OED 2003, Bahrdt 2000
• Modes of communication, cooperation and
  coordination towards the established objective
  for collaboration (group awareness, emergence)
  cf. Elsen 2007
• Knowledge To understand the object of
  collaboration
• Knowledge should be exchanged within the
  objective of collaboration Elsen 2007
  (different natures of knowledge, different
  knowledge conversion modes)
• Context To understand how other people can
  understand and learn from knowledge in its
  initial meaning
• Essential for knowledge conversion of
  organisation and its actors for taking
  appropriate and valuable actions Klemke 1999
  Kivijärvi 2004
• Ontology To understand how to coordinate and
  organise knowledge conversion
• Offering a skeletal and relational organisation
  for knowledge bases associated to represent
  different viewpoints based on the organisational
  level and area Huettenegger 2006 Swartout et al
  1996

11
Theory (2/3) Reviewed Concepts to Study
Collaboration and Knowledge Conversion in
Cross-Communities Key Features

• Organisation To understand the frame in which
  collaboration appears
• Organisations are immense interpretation
  systems cf. Daft/Weick 1984 Baumard 1999
• Organisation and the smaller unit community
  provide the frame in which collaboration could
  occur
• Challenges reaching knowledge conversion in
  cross-communities (Novak/Wurst 2004)
• Different thought worlds
• Different knowledge perspectives
• Establish a shared Context of Knowing
• Perspective Making and Perspective Taking
• Boundary Objects (Interpretable Knowledge
  artefacts)
• Visualisation of cross-community knowledge
  perspectives

12
Theory (3/3)Conclusion on Reviewed Concepts

• All concepts fulfil a particular role within the
  organisational context essential to study
  cross-community collaboration and knowledge
  conversion
• How are cross-community challenges characterised
  in the real-world?

13
Presentation Outline
gtgt Develop the industrial problematic and need
against the background of the research question
2

1. Empirical Context
2. Conduction of Empirical Study
3. Results of the Empirical Study

IndustryObservation
ExpertsInterviews
14
Empirical Context

• VIVACE (Acronym Value Improvement through a
  Virtual Aeronautical Collaborative Enterprise)
• Integrated European Research Project
• Resources lasted 4years (2004-2007), 75m
• Project Coordinator Philippe HOMSI (Airbus)
• 62 partners Aero Companies (29), IT Technology
  Vendors (12), Research Centers (5), universities
  (14), Others (2)
• 4 Sub projects 23 work packages, 102 tasks, 249
  sub-tasks
• 300 deliverables

15
Conduction of the Study (1/2)Collect Data

• Methodological Choice Qualitative research, a
  field observing and open facilitated research
  method (semi-structured incl. pre-tests)
• 17 Experts interviews (mostly at the place) June
  06 Feb 07
• Interview Profile
• 6 nationalities (France, Germany, Sweden,
  England, Netherlands, Italy)
• Project roles (project office, exploitation
  manager, sub-project, work-package and task
  leader)
• Types of organisations (Aero companies, Research
  Centres, project office)
• Interview time Total 14h
• Transliterated Material 70pp. (interview
  protocols), 3 weeks for transliteration

Captured a broad basis of data in the perimeter
of the empirical study environment VIVACE
16
Conduction of the Study (2/2)Structure
Interpret Data

• Evaluation of surveyed information in a
  pre-selected and existing categorization system
  (4dimensions Technical, Social, Resource,
  Environment 17variables) found in theory (2
  month, 40pp.)

17
Results of the Study (1/3)
Interview data documented in common interview
protocols
0
Interview data
InterviewProtocols
0
Structured hypothesis in context of an
inter-organisational European research project
1
Structured Interpreted
1

• Empirical Challenges Abstracted Result
• Different Backgrounds (variety of business
  approaches, objectives, cultures, working
  principles, )
• Missing common/shared perspective (visibility on
  the project shared view on business intents)
• Environmental Influences (external factors, lack
  of early involvement of business customer and
  end-user)

Abstracted generalised
2
2
3
Outline industrial problematic need based on
results from Step 0,1,2 and against the
background of the research question
18
Results (2/3)Industrial Problematic in Context
of the Research Question
3

• P.1 Business intents are stored in different
  information formats and spaces
• P2. Flat and non-contextualized representation
  (macro-viewing on documents) of business intents
• P3. PD teams often loose the justifying
  connection to business intents throughout the PD
  process
• P4. PD teams are often unsure if they implemented
  business intents completely and consistent in
  forms of requirements
• P5. Difficult to prove and trust the correct
  implementation of business intents in engineering
  processes and information spaces

PLC - Project Life Cycle
Reconciliation process of business intents and
specified requirements is challenging
19
Results (3/3)Industrial Need in Context of the
Research Question
3

• N1. Methodological approach to structure,
  organise and specify perceived business intents
  in alignment with specified requirements
• ? To answer Problem P1. P2. in terms of
  High-level product orientation and collaboration
  baseline towards which business and PD teams can
  activate all their efforts
• N2. Traceability mechanisms
• ? To answer Problem P3. P4 in terms of Trace
  and update business intents and requirements
• N3. Goal conflict and resolution mechanisms
• ? To answer P5. in terms of Relaxation and
  stabilisation of business intents before entering
  into heavy specifications
• N4. Evaluation engine
• ? To answer P5. in terms of Measure and
  estimate business intent fulfilment in relation
  with assigned requirements

? Avoid late and heavy iterations in progressive
stages of the PD process ? Increase Transparency
improve cross-community collaboration
20
Presentation Outline
gtgt Combine empirical needs from industry
observation with theoriesand come out with a
conceptual solution serving collaborative projects
3
ConceptualSolution

1. Construction of Conceptual Solution

ResearchFramework
21
Conceptual Solution Model (1/8)Confluence of
Reviewed Concepts
Knowledge-CoCoOn (Collaboration, Context,
Ontology) is a formation consolidating different
concepts to establish an environment for
knowledge conversion (share create) in context
of cross-community collaboration.
22
Conceptual Solution Model (2/8)Structure,
organise, specify and deploy business intents on
principles provided by Knowledge-CoCoOn
Our solution proposition 0perationalising the
conceptual model of Knowledge-CoCoOn within a
Business Needs Expectation Perspective (BNE-P)
Business Needs Expectation Perspective (BNE-P)
Model
Goal-tree structure
23
Conceptual Solution Model (3/8)The Role of
Collaboration in the BNE-P Model
Contribution in BNE-P Model

• Collaboration - Define principles of social
  interaction
• Answer to Need N1. Methodological approach to
  organize and specify high-level product
  definition
• Solution Element
• Principle collaboration in context of
  project-and engineering-based organisation

24
Conceptual Solution Model (4/8) The Role of
Collaboration in the BNE-P Model
Contribution in BNE-P Model

• Collaboration - Define principles of social
  interaction
• Answer to Need N1. Methodological approach to
  organize and specify high-level product
  definition
• Solution Element
• How organisational structures consider the
  environment (delimitation)

25
Conceptual Solution Model (5/8) The Role of
Ontology in the BNE-P Model
Contribution in BNE-P Model

• Ontology - Coordination of Knowledge Conversion
• Answer to Need N1. Methodological approach to
  organize and specify high-level product
  definition N2. Traceability Mechanisms
• Solution Element
• Providing structure and relationships organizing
  the synthesis of complex business information
  spaces (Perspective Making)

26
Conceptual Solution Model (6/8) The Role of
Ontology in the BNE-P Model
Contribution in BNE-P Model

• Ontology - Coordination of Knowledge Conversion
• Answer to Need N1. Methodological approach to
  organize and specify high-level product
  definition N2. Traceability Mechanisms
• Solution Element
• Defining transition points (that is
  collaboration objective) in a cross-community
  (Perspective Taking) allowing to trace and update
  knowledge between business engineering

27
Conceptual Solution Model (7/8) The Role of
Context in the BNE-P Model
Contribution in BNE-P Model

• TPBV Total Perceived Business Value
• Expectation Value Degree level of change in
  product feature capable to create the expected
  value, benefit for business management and its
  actors

• Context - Communication of Knowledge
• Answer to need N3. Goal Conflict and resolution
  mechanisms, N4. Evaluation Engine
• Solution Element
• Definition of Business Intent as basis for
  prioritising decompositions of product features

28
Conceptual Solution Model (8/8) The Role of
Context in the BNE-P Model
Contribution in BNE-P Model

• Context - Communication of Knowledge
• Answer to need N1. Methodological to approach to
  organize and specify high-level product
  definition
• Solution Element
• Nature of Message anchoring the concept of BNE
  for specifying contents (classes, attributes) of
  a business intent

29
Presentation Outline
gtgt Experimentation and Feedbacks applying the
conceptual solutionof the Business in Practice
4

1. Overview experiments
2. Experiment 1- Test the implementation of BNE-P
   Model Classes and Attributes
3. Experiment 2- Test valuation and traceability
   made possible by the BNE-P Model
4. Induced Developments

IndustrialApplicationCases
ApplicationCase
30
Three Different Experiments
Test the implementation of BNE-P Model Classes
and Attributes through an integration process
1
VIVACE
Test Traceability and valuation made possible by
the BNE-P Model on the scale of one business
intent
2
Study the possible application of BNE-P model for
the development of coherent requirements
specifications upstream in an EADS product
development project
EADS(ProspectiveResearches)
3
31
Experiment 1 (1/5)VIVACE Context

• In the second half of VIVACE (first year of the
  thesis) a project task was launched which allowed
  us to apply the developed BNE-P model
• At that stage (partially) business intents and
  engineering definitions were established, but in
  an individual mode (different structures, spread
  over different types of documents, )

32
Experiment 1 (2/5)VIVACE BNE-P Model Definition
BusinessCommunity
EngineeringCommunity

• Aero companies addressed business intents
  encompassed a large scope of investigation
• Differentiated a BNE-Perspective within
• Context (BNE-C) establishing the overall
  business context
• Focus (BNE-F) establishes and details a specific
  part of BNE-C in forms sub-objectives and
  benefits
• Allowed Aero companies to concretise their
  BNE-Perspective within a context and one or more
  focuses to be answered by PD teams (research
  centres, universities).

33
Experiment 1 (3/5) VIVACE BNE-P Model
Implementation

• Process of Integration
• Specifying business intents in BNE-Ps and its
  relating implementation in forms of engineering
  definitions
• Mixture of interviewing, document analysis,
  documentation and reviewing cycles with
  responsible BNE-P leaders

34
Experiment 1 (4/5)VIVACE Methodological Review

• Achievements
• Valid proof of model classes and attributes
  defined for the extended consideration of a BNE-P
  differentiated in context and focus
• Integration process was applicable on the full
  scale of VIVACE, with some challenges
• Support of people (missionary work)
• Availability and adequacy of information
• Limitations
• Model not experienced in the set-up and execution
  phase of the project
• Transversal analysis across BNE-P for matters of
  coherency analysis

The absence establishing business perspectivemay
led to over- under, or miss-specification
35
Experiment 1 (5/5)VIVACE Feedbacks against
empirical findings
Summary of Feedbacks (gained from interviews)
against empirical findings (problems and needs)
and in context of the research question
N1. Methodological approach to structure, organise and specify perceived business intents in alignment with specified requirements
P1. Business intents are stored in different information formats and spaces P2. Flat and non-contextualized representation (macro-viewing on documents) of business intents Helps to get common understandings of and transparency on business intents addressed and functional components developed. The specification of business intents in BNE-P helps in communication towards business management inside the partners company. Identified as key deliverable supporting exploitation phase enabling a logic of business value and engineering capability view The model could help to reach common value-oriented understandings more efficiently
36
Experiment 2 (1/8)VIVACE Context

• Test 1/ evaluation and 2/ traceability mechanisms
  made possible through the BNE-P Model on one
  VIVACE BNE-P specified within Experiment 1
• Performed together with a representative project
  member under study environment conditions

BNE-P Model
Objective
Benefit
Sub-Objective
1/ Evaluation of a goal tree
Requirements
2/ Trace consequences of evaluations towards
engineering structures
37
Experiment 2 (2/8)VIVACE 1/ Evaluation of
Soft-Goal Tree

• Characterisation of evaluation criteria using
  utility value function and uncertain information
  (interviews)
• Evaluation principle using BNE-P goal tree
  structures
• Temporal distinctions t0 (current business),
  t1 (intermediate), T End (targeted situation)
  using uncertain information
• Zones Current Zone, Improvement Zone, Targeted
  Zone

38
Experiment 2 (3/8)VIVACE 1/ Evaluation of
Soft-Goal Tree
Implementation of Goal Tree (In-house Software
for Probabilistic Evaluations)
39
Experiment 2 (4/8)VIVACE 1/ Evaluation of
Soft-Goal Tree
Illustrations of Probabilistic Evaluation Results
in the Goal-Tree
40
Experiment 2 (5/8)VIVACE 2/ Tracing Mechanisms
Establish captured relationships between Boundary
objects, i.e. benefits and specified requirements
Evaluatedgoal-tree
SpecifiedRequirements
CreateRelation(based oninterviews)
SaveRelation
41
Experiment 2 (6/8)VIVACE 2/ Tracing Mechanisms
Scenario to perform tracings on unsatisfied
business intent areas
BNE-P Evaluation Model
Specified Requirements
1
Goal-Tree Analysis
2
Trace consequences of evaluations towards
engineering structures
42
Experiment 2 (7/8)VIVACE Review
Summary of Feedbacks (gained from interviews)
against empirical findings (problems and needs)
and against the background of the research
question
N2. Traceability mechanisms N3. Goal conflict resolution mechanisms N4. Evaluation engine
P3. PD teams often loose the justifying connection P4. PD teams are often unsure if they implemented Capable to perform critical path analysis Prepare a situational picture in context of a business intent (e.g. serve moving business targets (Top-down) or difficulties in implementing engineering definitions (bottom.-up))
43
Experiment 2 (8/8)VIVACE Review
N2. Traceability mechanisms N3. Goal conflict resolution mechanisms N4. Evaluation engine
P5. Difficult to prove and trust the correct implementation Implicit proof only BNE-P provide evaluation structures In real case situation higher negotiation effort characterising evaluation criteria Concept of uncertainty helpful, but time-consuming defining it More confidence providing figures using uncertain information
44
Induced DevelopmentsUpdated Prototypical
Environment
Deploy requirements
Structure, specify organise business intent
Trace update
Perform situational Analysis
Proof Coherency
45
Presentation Outline

1. Contributions and Criticisms
2. Complementary Research Issues
3. Concluding Remark

Results Perspectives
5
46
Contribution and Criticisms (1/3)

• Result 1 BNE-P Model
• Contribution to Theory
• A knowledge-driven proposal to intentional
  modelling structuring, organising and deploying
  business intents informally
• Identification of goal-trees and resulting
  specified requirements to serve collaboration
  scenes at the interface of business and
  engineering
• Enables a first proof of coherency before
  entering into heavier formalisms
• Can provide attributes that relate to project
  management information and keep the link to
  business engineering information spaces
  (documents)
• Perspectives (limitations and open issues)
• Clear lack of integration with stronger
  formalisms in intentional modelling.
• ? Integrate strategies to perform transversal
  proof of coherencyamongst a number of goal-trees
  organised in various BNE-Ps
• Proof of different interdependency types amongst
  BNE-P organised goal-structures, e.g. semantic,
  cost/value, time

47
Contribution and Criticisms (2/3)

• Result 1 BNE-P Model (Experiment 13)
• Contribution to Practice
• VIVACE Increased visibility / transparency on
  the VIVACE project (from the inside outside) in
  the closure phase of the project
• Perspectives (limitations and open issues)
• VIVACE Clear lack of experimentations in set-up
  and execution phase of project
• ? Investigate the BNE-P Model along the PD
  process and include phase-specific surveys
  interview cycles, questionnaires, etc.
• ? To investigate business and engineering domain
  members behaviours during collaborations and
  knowledge conversions

48
Contribution and Criticisms (3/3)

• Result 2 BNE-P Evaluative Model including
  Traceability
• Contribution to Theory/Practice (Experiment 2)
• Orientation on value adding activities improved
  indication of expected functional qualities and
  orientation on prioritised business communities
  intentional structures
• Reflexive Traceability The BNE-P evaluative
  model provides channels for cross-domain
  associativity (introduced as boundary objects)
  offering a logic to follow in bottom-up or
  top-down fashions throughout intentional (BNE-P)
  and engineering information structures
• Perspectives (limitations and open issues)
• Clear lack of proof under operational conditions
• Open issues
• Integration with stakeholder analysis approaches
• Integration with higher level metrics (on
  enterprise level)
• Scalability (balancing appropriate method and
  tool assembly in context of the collaboration
  challenge)

49
Complementary Research Issues

• BNE-P Model
• Investigate BNE-P in ideation phase that allows
  researches (and RT strategy) to prepare
  argumentation baseline towards new innovations.
• Investigate BNE-P on a different scale, e.g. SME
  
• In particular, what are relevances of BNE-P if
  we face a non collaborative situation? That means
  business intent and formalisation in forms of
  specified requirements is done by the same
  person.
• Empirical Study Results
• Validate surveyed results in context of other
  projects (same or different context)
• Use surveyed material for quantitative research
  (e.g. compiling questionnaire with closed
  questions) and proof hypothesis and establish
  generalised statements

50
Concluding Remark

• The presented work advocates the point that if
  current intentional models fall short in
  establishing usable intentional structures that
  are able to provide the transparency for
  supporting continuously business-engineering
  evolutions within collaboration and knowledge
  conversions along a PD process, then it could be
  valuable to have a mediating instance that
  organises collaboration and knowledge
  conversions.
• It can act (promote) in front of stronger
  formalisms in terms of coherency development in
  requirements.
• It could strengthen negotiation forces and
  group-awareness among business and engineering
  community.
• It provides organisation of knowledge bases, i.e.
  community-related information spaces and anchors
  a value-oriented definition of business intent.
• It supports not only front-end negotiations, but
  also establishes continuous interactivity
  structures and strengthens product development
  performance in terms of increasing reactivity and
  group-awareness between business-engineering.

51

• Many Thanks!