GEP Algorithm-based Optimization Method for PCs Production Scheduling under Due Date Variation

Production scheduling plays a critical role in the on-time delivery of precast components (PCs) during the assembly of prefabricated construction. However, previous studies focused less on considering the changing scenarios during production process. And existing models are not adequate for dynamic environments where construction projects may have due dates change in actual PCs demand that occur due to on-site assembly schedules. As a direct experience, dispatching rules are suitable for solving complex scheduling problems, but existing manually configured scheduling rules are not guaranteed to be universal across multiple scenarios. The paper combines discrete-time simulation methods to establish a due date variation response model and by using dynamic scheduling rules constructed by Gene Expression Programming (GEP) algorithm to cope with multiple due date variation scenarios and achieve dynamic production planning. The effectiveness of the model and the superiority of the algorithm are tested by a real case. 25.8% and 7.5% cost savings can be achieved by comparison to no response and other existing excellent dispatching rules, respectively. The results show that the method can effectively address the occurrence of due dates variations and enhance the dynamic precast production management.