Reduction of Petri net maintenance modeling complexity via Approximate Bayesian Computation

The accurate modeling of engineering systems and processes using Petri nets often results in complex graph representations that are computationally intensive, limiting the potential this tool real life applications. This paper presents a methodology to properly define optimal structure properties reduced net mimic output reference model. is based on Approximate Bayesian Computation infer plausible values model parameters rigorous probabilistic way. Also, method provides numerical measure level approximation structure, thus allowing selection among set candidates. suitability proposed illustrated simple illustrative example system reliability case study, showing satisfactory results. also show allows flexible reduction taken as reference, justification for choice structure.