A metamodeling and transformation approach for knowledge extraction Inaya Lahoud 1 . Davy Monticolo 2 , Vincent Hilaire 1 , Samuel Gomes 3 1 SET Laboratory, University of Technology of Belfort-Montbeliard, Belfort, France {Inaya.Lahoud, Vincent.Hilaire}@utbm.fr 2 ERPI Laboratory, Polytechnic National Institute of Lorraine, Nancy, France {Davy.Monticolo@ensgsi.inpl-nancy.fr} 3 M3M Laboratory, University of Technology of Belfort-Montbeliard, Belfort, France {Samuel.Gomes@utbm.fr} Abstract. Among the goals of Knowledge Management (KM) we can cite the identification and capitalization of the know-how of companies in order to or- ganize and disseminate them. KM is recognized as a non-trivial task. In the context of extended enterprises it is even more difficult because they are geo- graphically distributed and use heterogeneous information systems. Indeed, in- formations from which knowledge is derived from are stored in different data- bases, distributed in different sites across the entire network of the extended en- terprise. This paper proposes a semi-automatic approach for knowledge extrac- tion from several databases. This knowledge will be stored in an organizational memory (OM). The approach is based upon the definition of ontologies for knowledge exchange and Model Driven Engineering concepts such as meta- models and transformation to process requests. Keywords: Ontology; Semantic Web Service; Knowledge Management; Model Driven Engineering; Interoperability 1 Introduction: The process of designing manufactured products (called products from now on) fre- quently requires the use of several heterogeneous business tools (calculation tool, CAD tool, a tool for production management, PLM, PDM, etc.). Each of these tools deals with specific aspects of product design such as: CAD, project management, resource planning and PLM. In the nowadays context, these tools are distributed on different sites across the entire network of the extended enterprise. Moreover, to survive in an increasingly competitive business environment, manu- facturing enterprises are under unprecedented pressure to become leaner and more agile. The product range must be updated permanently and production costs the low- est possible. Product leadership companies must continue to enter new market with innovative products. This requires optimizing process and methodologies used by engineering department. The design process has to be rationalized in capitalizing