Performance evaluation with cost estimation for supervised systems

Supervisory control theory deals with automated synthesis of models of supervisory controllers that ensure safe and nonblocking behavior of the supervised system. Typically, (large) supervised systems cannot be guaranteed to meet elaborated performance requirements, as ensuring them during the synthesis procedure is a costly undertaking. We propose a model-based systems engineering framework that enables supervisor synthesis of stochastic (nondeterministic) discrete-event systems, and post-synthesis validation of quantitative properties of the supervised system by imposing a cost model. To this end, we develop several extension tools that interface with the supervisory controller synthesis tool Supremica. To illustrate our approach, we remodel an industrial case study involving coordination of maintenance procedures of a printing process and demonstrate how to obtain performance measures using Markovian reward model checking.