Delay bound of inference-based decentralized diagnosis in discrete event systems

Abstract

We previously introduced an inference-based decentralized diagnosis framework for discrete event systems, where inferencing over the ambiguities of the self and the others is used to issue local decisions, and a global decision is taken to be the one with the least ambiguity level. In this setting, we introduced the notion of N-inference diagnosability to characterize the detection of any failure within a bounded delay, using at most N-levels of inferencing, and subsuming both disjunctive and conjunctive ways of decision fusions. In this paper, we compute the delay bound within which the occurrence of any failure can be detected for an N-inference diagnosable system. Computing the delay bound is important to execute mitigation actions in a timely manner, and is a figure of merit of a diagnosis scheme.