Joint optimization of opportunistic maintenance and production scheduling considering batch production mode and varying operational conditions

The joint optimization of production scheduling and maintenance has been a hot topic. Most of the existing publications assume that the operational condition (OC) of a machine is constant during the entire production task. However, it is practical that a machine may experience several different OCs due to the requirement of different jobs. The varying OC may impact the deterioration of machines and their maintenance decisions. In addition, in a batch production system, a maintenance action has to be advanced before or postponed after a processing batch. Even if a machine is maintained preventively, it cannot be renewed since preventive maintenance (PM) is usually imperfect. In addition, minimal repair should be considered, which is costly and time consuming upon machine failure. The above concerns are practical but insufficiently considered in an integrated manner. Due to this integration, the complexity of the joint optimization problem is enhanced. An improved genetic algorithm (GA) based on random keys, convex set theory and the Jaya algorithm is proposed to solve the joint optimization problem of opportunistic PM and production scheduling in a batch production system under varying OCs. A series of comparative cases are conducted to illustrate the effectiveness of the proposed methods.