Abstract
This paper presents an intelligent support of decision making to improve security analysis called Adast. The adopted approach in this research is based on acquiring and reusing past accident scenarii, historically validated on other homologated transport systems. The advantage of this approach, based primarily on Case-Based Reasoning (CBR) and ontologies, lies not only on the capitalization of knowledge from experience feedback, but also on benefit in order to provide assistance to domain experts in their crucial task of analyzing and improving security. Also, the problems of knowledge and case representation are considered in this paper. A prototype of intelligent decision support system based on Adast has been implemented to better integrate the proposed CBR cycle and ontology.