TOWARDS THE INTEGRATION OF DIFFERENT KNOWLEDGE SOURCES IN MODEL-BASED DIAGNOSIS

Over the last few years several attempts have been made to design diagnostic systems which combine more than one model of the system to be diagnosed or more than one reasoning mechanism. In this paper we analyze the integration of different knowledge sources in model-based diagnosis, and, in particular, abductive diagnosis. We consider behavioral models in which the interaction between processes can be represented, enriched with constraints and taxonomic relationships among diagnostic hypotheses. We provide several insights into the role of such forms of knowledge in model-based diagnosis, showing how they can be accommodated both in a meta-level definition of abduction with constraints and in an object-level framework for abduction