Predictive-Reactive Scheduling to Increase Robustness of Production Systems

In order to enhance the robustness and flexibility of production systems and improve their responsiveness to unforeseen events, it is essential to devise appropriate strategies. This paper aims to develop and evaluate a predictive-reactive scheduling method to increase the robustness of production planning and control. As disruptions can often be anticipated, but not always prevented, a scheduling method that combines pre-dictive planning and reactive control reduces the impact of such occurrences on the production system. In order to achieve this, a real-world use case is introduced, and the relevant disruptions are identified. To address the predictive aspect of the methodology, an allocation problem and subsequently a sequencing problem are first solved using heuristics and mathematical optimization in order to schedule the jobs. Reactive scheduling strategies are then developed and implemented into a discrete event simulation model for the purpose of evaluating the improvement of the results. The results demonstrate a significant enhancement in production planning, as evidenced by more balanced processing times and a sufficient buffer for rescheduling. Therefore, this method enables the generation of a robust and cost-minimized production plan. By implementing the rescheduling strategy for production control, the impact of disruptions are minimized, as evidenced by the obtained key figures which serve as a robustness indicator.