Reliability analysis by Markov model and stochastic estimator of stochastic Petri nets

Abstract

In our study, we were interested in the reliability of large discrete systems. These studies can be based on Markov models or on Stochastic Petri Nets (SPNs) that are generally used for the analysis and synthesis of the models used in the different phases of a system's life. Markov models or SPNs are perfect for many cases, still, they suffer from the combinatorial explosion when analytically their state numbers increase as the complexity of the dynamic systems grows accordingly with their components. Such issue reflects itself in the slowness of these models to accomplish convergence. These different modelling tools make it possible to deduce the average behaviour and to obtain the performance indicators of the system studied, either by calculation or by estimation. We will present the Markov analysis of a system whose state space is finite as well as its estimator obtained using SPNs.