KNOWLEDGE MANAGEMENT SUPPORT FOR SYSTEM

ENGINEERING COMMUNITY

Olfa CHOURABI, Pr. Mohamed BEN AHMED

RIADI Research Laboratory, ENSI University, la Manouba, Tunis,TUNISIA

Yann POLLET

Chaire d’intégration de systèmes, CNAM, Paris, France

Keywords: Knowledge Management, Corporate Memory, Project Memory, Knowledge Processes, System Engineering

Abstract: Knowledge is recognized as a crucial resource in today’s Knowledge Intensive Organizations. Creating

effective Knowledge Management systems is one of the key success factors in Process Improvement

initiatives like the CMMI, SPICE etc. This contribution aims to provide a starting point for discussion on

how to design a Knowledge Management System to support System Engineering activities. After

motivating the problem domain, we introduce a conceptual model supporting continuous learning and reuse

of all kinds of experiences from the System Engineering domain and we present the underlying

methodology.

1 INTRODUCTION

In today’s highly dynamic environment the effective

use of all available Corporate Knowledge is

indispensable for success. Knowledge Management

(KM) tries to tackle this problem by providing

methods and tools to support the creation,

acquisition, capitalization, sharing and effective use

of knowledge in social settings. To this end, a

promising approach is the development of a

Corporate Memory (CM).

A Corporate Memory is an explicit, disembodied

and

persistent representation of knowledge and

information in an organization, in order to facilitate

their access and reuse by members of the

organization, for their tasks (Rabarijaona et al.,

2000). The main objective of building a Corporate

Memory Management System is the coherent

integration of this dispersed knowledge in a

corporation with the objective to promote

knowledge growth, promote knowledge

communication and in general preserve knowledge

within an organization (Steels, 1993).

System Engineering (SE) is a knowledge-intensive

ocess encompassing requirement gathering,

design, development, testing, deployment,

maintenance, and project coordination and

management activities. The System Engineering

process could be defined as an iterative process of

problem resolution aiming at transforming user’s

requirements into a solution satisfying the

constraints of: functionality, cost, time and quality

(Meinadier, 2002).

It is highly unbelievable that all members of a

devel

opment team possess all the knowledge

required for the activities mentioned above. This

underlies the need for KM support to enable SE

organizations to:

-Effectively share domain expertise within

devel

opment team;

- Identify the requirements of a system;

-Reuse non-externalized knowledge of the

devel

opment team members;

-Bring together knowledge from distributed

ndividuals to form a repository of organizational

knowledge;

-Retain knowledge that would otherwise be lost due

o the loss of experienced staff; and Improve

organizational knowledge dissemination to support

quality and process improvement initiatives.

Meeting these demands requires fast, continuous

earning and reuse of past experiences within project

teams or Communities of Practice (

Wenger, 1998).

339

CHOURABI O., BEN AHMED M. and POLLET Y. (2005).

KNOWLEDGE MANAGEMENT SUPPORT FOR SYSTEM ENGINEERING COMMUNITY.

In Proceedings of the Seventh International Conference on Enterprise Information Systems, pages 339-342

DOI: 10.5220/0002520803390342

 SciTePress

We here focus on system process improvement

through transfer of experience, to create what we

could call “learning System Engineering

organizations”.

In this paper we argue that an interesting approach

for System Engineering activity support lies in the

adoption of a process-oriented Knowledge

Management strategy (POKM) A POKM strategy

aims to provide project members with task-relevant

knowledge in corporate operational processes (

Maier

et al., 2001).. As opposed to a purely document-

centric perspective for KM, it examines what

processes create knowledge assets and how to

represent this contextual knowledge in a Corporate

Memory.

The paper is structured as following: the first section

introduces System Engineering domain and presents

our motivations for supporting SE Community with

a Knowledge Management System. The second

section gives an overview of existing Corporate

Memory construction approaches that are relevant

for SE domain. Based on existing approaches and

industrial needs, we propose a KM System for SE

community in section three. Finally potential issues

for future research directions are outlined.

2 CONTEXT AND MOTIVATION

2.1 System engineering domain

A cartography of SE processes according to the

Standard ISO 15288 includes:

Agreement Processes: Acquisition, and Supply

Enterprise Processes: Enterprise Environment

Management, Investment Management, System Life

Cycle Process Management, Resource Management,

and Quality Management

Project Processes: Project Planning, Project

Assessment, Project Control, Decision-making, Risk

Management, Configuration Management, and

Information Management

Technical Processes: Stakeholder Requirements

Definition, Requirements Analysis, Architectural

Design, Implementation, Integration, Verification,

Transition, Systems Analysis, Validation, Operation,

Maintenance, and Disposal

2.2 Motivation for KM in SE

We constantly gain experience with each

development project. “Knowledge emerges in work

practices, often being defined by the first project to

address the issue involved”. (Henninger, 2003). In

this section, we discuss the advantages of KM

support for SE community.

Various attempts have been made to Quality

Improvement in SE domain. According to the

standard IEEE 1220, continuous process

improvement in SE could be addressed through:

The application of an auto-evaluation program that

determines the degree of maturity in SE practices

like CMMI

Capitalization and transfer of experiences gained

with each development project (Meinadier, 2002).

We here focus on se Process Improvement through

knowledge capitalization. We propose a Corporate

Memory structure to support SE activities by

arguments that create a result. Expressions can be

used as values in any command. providing pieces of

relevant experiences, competencies, tools,

methodologies, process models and reusable

components.

3 RELATED WORKS AND

POSITIONING IN THE STATE

OF THE ART

Knowledge Management is a very complex problem

and can be tackled from several viewpoints: socio-

organizational, financial and economical, technical,

human and legal (Barthes, 1996)

One approach for managing knowledge in

organizations is to set up a corporate memory. The

corporate memory is in charge of insuring persistent

storage, indexing and diffusion of relevant pieces of

knowledge within the organization. (Gandon, 2002)

In the following, we present an overview of some

relevant methodologies to construct Corporate

Memories. (Dieng et al., 1998)

Non Computational Corporate Memory ;

Document-based Corporate Memory (Dieng et

al., 1998); Competence Management systems

(Rus et al., 2001) ; Knowledge-based Corporate

Memory (Strohmaier, 2003) ; Case-based

Corporate Memory ; Process-oriented Corporate

Memory (Abecker et al., 2001),

4 CONCEPTUAL MODEL FOR

SYSTEM ENGINEERING

KNOWLEDGE MANAGEMENT

SYSTEM (SEKMS)

4.1 Basic principles of SEKMS

This section discusses basic principles guiding the

design of our approach

ICEIS 2005 - ARTIFICIAL INTELLIGENCE AND DECISION SUPPORT SYSTEMS

340

Process centered approach for KM: the

hypothesis stating that Knowledge is tied to action

leads us to consider knowledge capture inside

processes. We argue that knowledge management

activities cannot be separated from other business

processes, because knowledge emerges within

business processes (Maier, 2001

). Thus, knowledge

that emerges in a process can be stored together with

its process context.

Multi-perspective approach: typically, corporate

memories are supported by centralized approaches.

Thus, knowledge about people, knowledge about

processes, and domain knowledge are represented

and maintained as information in global repositories

which serve as sources to meet a knowledge

worker’s knowledge needs.(Abecker, 2004). We

propose to better balance between the needs of

smaller organizational units and the more global KM

concerns, by providing each community the

capability of managing its own knowledge. We

propose to design a distributed organizational

memory as a constellation of set specialized OMs.

Proactive delivery of relevant knowledge: a

proactive knowledge delivery methodology

addresses the dissemination of knowledge assets in a

just-in- time manner in the context of its applicable

targeted processes. (Abecker,2004) It allows

knowledge retrieval when it is applicable to the task

in which a user is currently engaged. To achieve this

goal, we propose to adopt intelligent agent

technology to procure crucial knowledge to users.

4.2 Architecture of SEKMS

In the context of SE activities, both informal

knowledge (such as documents) and formal

knowledge (such knowledge represented by a

knowledge model) are needed. Therefore our

approach aims to design a CM architecture where

the CM can be composed of different sorts of

memories: documents, knowledge bases etc. And

thus, combining several techniques for CM

construction is needed. Figure 1 shows the different

components of the SE Knowledge Management

System (SEKMS). We distinguish three principal

perspectives.

Business process perspective: provides the

description of the real business process.

In a business process, knowledge exists in the form

of data and information in combination with

experience, communication, reflection, expertise,

techniques and cognitive abilities.

Our approach begins with a careful analysis of the

existing business processes, in order to identify what

kind of knowledge is created/used by whom in the

context of his daily work. This analyze includes both

the explicit and the tacit form of knowledge. It

includes the identification of tools, documents,

models, -defined roles, and codified procedures. But

it also includes all the implicit relations, the tacit

conventions, the recognizable intuitions, the specific

perceptions, the embodied understandings, the

underlying assumptions, the shared worldviews, the

sequences of collaboration, which may exist in

implicit form, connected to the activities of

employees in the context of running activities. The

output of this business process analyze, is a set of

raw experiences captured and then enriched with

context attributes.

Knowledge structure perspective: Provides a

formal structure to describe knowledge assets. A

common approach to model knowledge in corporate

memory is to take document-centric approach. As

SE experiences are complex and diverse, our CM is

defined in term of knowledge components (KC)

instead of documents. A KC can be a process model

fragment, a best practice, a lesson learned, a product

model, a requirement, or comment on a business

process et. Providing that we work on different KC

structures, we propose to rely on several knowledge

representation techniques for representing KC in the

CM. Thus, we propose using: conceptual graphs,

lattices, semi-structured documents etc., in order to

provide more flexibility and consisted ness in KC

representation.

Meta Knowledge Model perspective The Meta

knowledge model is a unified view that structures

the content of the CM. It provides explicit formal

specifications of the concepts and terms in the SE

domain and the relationships among them. This meta

knowledge model is not yet well specified, we plan

to generate it through synergies between standards

Meta models proposed in SE literature.

KNOWLEDGE MANAGEMENT SUPPORT FOR SYSTEM ENGINEERING COMMUNITY

341

Business process analysis / Assessment of Knowledge

Needed/generated in the context of SE activities

Business

rocess

ers

ective

REFERENCES

Abecker A., 2004, An Agent-based Framework for

Distributed Organizational Memories. Knowledge

Management Department German Research Center for

Artificial Intelligence DFKI Kaiserslautern, Germany

Barthès J.-P., 1996, ISMICK and Knowledge

Management. In Proc. of ISMICK’96, J. F.

Schreinemakers ed, Knowledge Management:

Organization, Competence and Methodology

Dahlberg S., 1998, INCOSE SE terms glossary. INCOSE

Concepts and Terms WG, International council on

system engineering, version 0.

Dieng R.,Corby O., Giboin A., Ribière M., 1998, Methods

and Tools for Corporate Knowledge Management, in

Proc. of KAW’98, Banff, Canada

Gandon F.,2002, Distributed Artificial Intelligence and

Knowledge Management: Ontologies and Multi-Agent

Systems for a Corporate Semantic Web, In 6th

International Workshop, CIA 2002, volume 2446 of

LNCS, Berlin, Germany, Springer-Verlag

Henninger S., 2003, Tool Support for Experience-Based

Software Development Methodologies, In Advances in

Computers, Vol. 59, pp. 29-82

Meinadier J., 2002, Le métier d’intégration de systèmes.

Hermès Science Publications, Paris.

Maier R., Remus U., 2001,Towards a Framework for

Knowledge Management Strategies: Process-

Orientation as a New Strategic Starting Point. In

HICSS-34 Hawai

Martin P. and Eklund P., 2000, Knowledge retrieval and

the world wide web. In R. Dieng ed, IEEE Intelligent

Systems, Special Issue on Knowledge Management

and Knowledge Distribution Over the Internet, p. 18-

Rus I., Lindvall M., 2001, Knowledge Management in

Software Engineering A State-of-the-Art-Report.

Produced by Fraunhofer Center for Experimental

Software Engineering Maryland and The University of

Maryland

Rabarijaona A., Dieng R., Corby O., Ouaddari R., 2000,

Building a XML-based Corporate Memory. In IEEE

Intelligent Systems, Special Issue on Knowledge

Management and Internet. 56-64.

Steels L., 1993, Corporate Knowledge Management ,

Proceedings of the 1

International Syposium on the

Management of Industrial and Corporate Knowledge

(ISMICK'93), pp. 9-30

Wenger E., 1998, Communities of practice. Learning,

meaning, and identity. Cambridge University Press,

Cambridge, UK

Figure 1: Conceptual model of SEKMS

Raw ex

erience ca

tured

Context modeling: (phase V cycle, activity, process, role, out put, input,

constraints)

stem En

ineerin

unified ontolo

(

meta-model

)

Meta knowledge perspective

Model perspective

Categorize knowledge

components

Explicit: technical documents,

process fragment, lessons

learned, requirement statement

Tacit: know how, design

rationale, informal

collaboration

Formalize knowledge

Conceptual graphs, lattices,

semi structured documents,

cases. Semantic annotations,

process models,

Knowledge Mining

Clustering

/Generalizing/

Refinement

Knowledge

Structure

perspective

ICEIS 2005 - ARTIFICIAL INTELLIGENCE AND DECISION SUPPORT SYSTEMS

342