Job shop rescheduling with rework and reconditioning in Industry 4.0: an event-driven approach

In this paper, we investigate the impact of rescheduling policies in the event of both rework and reconditioning in job shop manufacturing systems. Since these events occur in unplanned and disrupting manner, to address them properly, it is required to manage real-time information and to have flexible reaction capacity. These capabilities, of data acquisition and robotics, are provided by Industry 4.0 Technologies. However, to take full advantage of those capabilities, it is imperative to have efficient decision-making processes to deliver adequate corrective actions. In this sense, we propose an event-driven rescheduling approach. This approach consists of an architecture that integrates information acquisition, optimization process, and rescheduling planning. We study the performance of the system with several algorithms with two performance criteria, namely, (i) relative performance deviation (RPD) in terms of objective function and (ii) schedule stability. We also propose a hybrid policy that combines full rescheduling regeneration with stability-oriented strategies aimed to balance both criteria. We conducted extensive computational tests with instances from the literature under different scenarios. The results show that a sophisticated algorithm can obtain better quality schedules in terms of the objective function but at the expense of sacrificing stability. Finally, we analyze and discuss the results and provide insights for its use and implementation.