Author(s)

P.P. Parra & A. Ligeza

Abstract

Causal reasoning is a practical support for model-based diagnosis [1]. The paper addresses the issues of extending diagnostic reasoning based on ab- ductive analysis of causal structures. The extension is aimed at admitting fuzzy characterization of faults; binary evaluation (i.e. faulty/correct) is no longer necessary. The degree of faultyness is expressed with use of basic fuzzy notions. This extension uses a uniform model for representing causal behaviour of diagnosed systems: it has the form of an AND/OR/NOT causal graph allowing for specification of causality types reflecting the basic logical operations [3]. The graph can be used to search for potential (possible) di- agnoses. Validation of generated diagnoses is performed by propagat

Keywords