Fault prognosis of timed stochastic discrete event systems with bounded estimation error

This article deals with the problem of fault prognosis in timed stochastic discrete event systems. For that purpose, partially observed stochastic Petri nets are considered to model the system with its sensors. The model represents both healthy and faulty behaviors of the system. Using a timed measurement sequence issued from the sensors, an approach denoted (ρ,δ)-prognosis is proposed to estimate the probability of a future fault occurrence. The method is based on two input parameters: the error bound ρ and the prognosis horizon δ. The main contribution is to bound the estimation error by ρ when the prognosis horizon does not exceed δ. An example is presented to illustrate the results.