Modeling Uncertain Temporal Evolutions in Model-Based Diagnosis

TL;DR

This paper introduces a probabilistic approach using Markov chains to model and diagnose uncertain temporal evolutions in dynamic systems, enhancing accuracy by filtering unlikely diagnoses.

Contribution

It proposes a novel method integrating Markov chain theory into model-based diagnosis for time-varying systems with uncertain temporal knowledge.

Findings

Effective filtering of unlikely diagnoses using Markov chain probabilities

Improved diagnostic accuracy for systems with uncertain temporal behavior

Framework applicable to reliability-based temporal diagnosis

Abstract

Although the notion of diagnostic problem has been extensively investigated in the context of static systems, in most practical applications the behavior of the modeled system is significantly variable during time. The goal of the paper is to propose a novel approach to the modeling of uncertainty about temporal evolutions of time-varying systems and a characterization of model-based temporal diagnosis. Since in most real world cases knowledge about the temporal evolution of the system to be diagnosed is uncertain, we consider the case when probabilistic temporal knowledge is available for each component of the system and we choose to model it by means of Markov chains. In fact, we aim at exploiting the statistical assumptions underlying reliability theory in the context of the diagnosis of timevarying systems. We finally show how to exploit Markov chain theory in order to discard, in the diagnostic process, very unlikely diagnoses.