Cet article met en avant les enjeux importants de la gestion de connaissances pour le producteur d'énergie qu'est EDF. Ces enjeux, et les besoins auxquels ils répondent, conduisent à des démarches pouvant avoir plusieurs objectifs opérationnels. Au coeur de la conception des solutions envisagées se trouvent la mise en oeuvre de systèmes d'organisation des connaissances adaptés et la manière de prendre en compte, gérer ou produire des documents. Cet article propose une analyse de démarches de gestion de connaissances existant à EDF en préalable à la spécification d'une méthodologie de gestion des connaissances incluant la caractérisation des systèmes d'organisation de connaissances (SOC) associés. ABSTRACT. This article states the most significant stakes of knowledge management for an energy producer such as EDF. These stakes and the needs they intend to answer in an operational context yield to the implementation of different approaches. Putting into operation the best suited knowledge organisation systems and dealing with documents (producing them, managing them or exploiting them) play a key role in the design of knowledge management solutions. This article proposes an analysis of EDF existing knowledge management approaches prior to the specification of a knowledge management approach including associated knowledge organisation systems (KOS) characterization. MOTS-CLÉS : gestion des connaissances, ingénierie des connaissances, système à base de connaissances, systèmes d'organisation de connaissances, cas industriel.

No abstract

The long term management of a production asset raises several major issues among which rank the technical management of the plant, its economics and the fleet level perspective one has to adopt. Decision makers are therefore faced with the need to define long term policies (up to the end of asset operation) which take into account multiple criteria including safety (which is paramount) and performance. In this paper we first remind the reader of the EDF three-level methodology for asset management. We then focus on the knowledge model and on the software tools that implement this methodology in order to gather, preserve, share, maintain and exploit the expert knowledge needed for asset management and to allow decision makers to define, evaluate and analyze long term plant operation and maintenance policies. Lastly, as the quality of the processed plant level evaluations (operation & maintenance strategies are evaluated, at a plantlevel, through a set of technical and economic indicators) and their interpretation relies on the quality of the knowledge captured in the tools, we focus on the definition of a “adaptative” user interface — based on Electronic Structured Documents — that allows technical/strategic experts and decision makers to consult the useful pieces of knowledge in a context dependent way. Such an interface, which, in a near future, should be fully implemented in the tools will facilitate the validation of the knowledge-base content and the analysis of the processed results.

No abstract