Modeling the supervision of manufacturing system considering diagnosis and treatment of fault

Currently, manufacturing system is being restructured to face the global competitiveness. As result, new manufacturing strategies need to be considered, taking into account aspects such as costs, quality, delivery times and flexibility, among other. However, in practice, none of these strategies can be effectively implemented if they disconsider the occurrence of faults in the manufacturing system. In this sense, this paper introduces a procedure for supervision of manufacturing systems including diagnosis and treatment of faults, i.e., a supervision strategy that considers not only the normal behavior of systems components, but also abnormal (faulty) conditions of them. The present approach uses Bayesian networks for the diagnosis and decision-making purposes, and interpreted Petri net for the manufacturing process synthesis, modeling and control purposes. The integration of these techniques guarantees the specified functionality of the system. Special emphasis laid on methodological issues and industrial systems, where a hierarchical structure can be adopted. A flexible assembly system is presented as an application example.