Dynamic Aggregate Planning of a Make-to-Order Production System

Abstract In Make-to-Order (MTO) production, products and production are non-standard and incoming orders arrive variably, with varying work content and due dates. Aggregate Production Planning (APP) takes place in the medium run, and the purpose of it is to match resource capacity with demand. In APP, orders are roughly scheduled for processing so that the resource capacity constraints and deadlines can be met. The static resource constrained project scheduling problem is the conventional research approach, but real-world production planning requires dynamic scheduling to integrate new orders over time. The dynamic model in this study employs periodic scheduling, treating it as sequential static mixed integer linear optimization problems, focusing on minimizing tardiness and rescheduling disruptions. The model is applied in experiments concerning a Flexible Manufacturing System (FMS) -centered MTO production system. The experimentation focuses on the effects of replanning frequency, order arrival rate, delivery date slack, forward planning vs. backward planning, restrictions in order processing activity timing and activity workload allocation method on the planning results. Based on the results, conclusions and guidance applicable to practical production planning are provided.