Multi-resource constrained scheduling considering process plan flexibility and lot streaming for the CNC machining industry

The CNC machining scheduling problem is a classic flexible job shop scheduling problem (FJSSP) that has been proven to be NP-hard. Although there has been extensive research on FJSSP, its applications in real-world CNC machining scheduling problems have received limited attention, with few related papers. In reality, several critical characteristics within the CNC machining industry can significantly impact the scheduling problem, such as process plan flexibility, multiple resource constraints, and lot streaming. This study aims to establish a scheduling system for the CNC machining industry that integrates order information, work in process information, and the aforementioned characteristics, using a hybrid multi-objective genetic algorithm (HMOGA) combined with local search. The optimal parameter combination for HMOGA is determined through a design of experiments, while the effectiveness of local search is evaluated using computational analysis. Finally, real data from a company is used to validate the proposed method. The results demonstrate that the scheduling system developed in this study effectively addresses real-world CNC machining scheduling problems.