An NSGA-II-based multiobjective approach for real-time routing selection in a flexible manufacturing system under uncertainty and reliability constraints

Abstract

Routing flexibility is one of the most common types of flexibilities of manufacturing systems. It allows the system to continue producing given part types despite uncertainties. Its main purpose is to maintain a high level of performance so that the system can deal with disturbances (failures, maintenance actions,…). This type of flexibility occurs when there are alternative or redundant machine tools in the system. However, due to resource and alternative routing limitations, the scheduling problems in such systems can become very complex. Furthermore, though routing flexibility aims to enhance system responsiveness, it still depends on the reliability and availability of machines and individual components in a system. The present paper aims to investigate the scheduling problem in a flexible manufacturing system (FMS) with routing flexibility under uncertainties related to the random arrival of parts orders and machines failures, by considering reliability and maintenance constraints. The real-time decisions for part routing selection are made using a non-dominated sorting genetic algorithm (NSGA-II), by considering the workload, utilization level, and reliability of machines in a workstation, in order to minimize the deadlocks and maximize the overall system reliability. The simulation results obtained showed that, for an overloaded system, the proposed NSGA-II algorithm induces the best performance in terms of total profit, system productivity, and machines utilization.