<![CDATA[Modern electronics controller manufacturing operates in highly dynamic environments where production schedules must rapidly adapt to newly arriving customer orders while maintaining production efficiency and delivery performance. However, conventional predictive scheduling alone is often insufficient because schedule revisions are required for unexpected order arrivals, which may increase production completion time and customer lateness. Therefore, this study aims to develop a two-stage predictive–reactive scheduling framework for a no-wait flow shop with sequence-dependent setup times (SDST), involving 20 initial jobs with varying release times and 7 dynamically arriving jobs with distinct due dates. In Stage 1, a Genetic Algorithm (GA) is employed to maximize the number of initial jobs completed before a 300-minute production cut-off, establishing a predictive baseline schedule. In Stage 2, the proposed GA performs reactive rescheduling by integrating the unprocessed initial jobs with the newly arriving jobs to minimize the makespan deviation from the predictive schedule and reduce the new jobs’ lateness. Based on a case study, the results demonstrate that the proposed model and GA optimization effectively balance schedule stability and responsiveness. The results show that while reactive rescheduling introduces a makespan deviation of 183-time units, incoming job lateness is substantially reduced, providing a trade-off between operational continuity and service-level performance.]]>
Copyrights © 2026
