A multi-decision approach for decentralized diagnosis of the presence and absence of faults in discrete event systems

We develop a multi-decision framework for decentralized diagnosis of discrete event systems (DES), where each diagnoser issues a tuple of diagnoses instead of a single diagnosis. We use the multi-decision framework to generalize existing methods for decentralized diagnosis. We study the diagnosis of the occurrence of fault as well as the diagnosis of the absence of fault. We show that the multi-decision permits to diagnose a broader class of systems. We also show how to tackle a potential undecidable problem that seems to arise with multi-decision.