Swarm algorithms for sustainable dynamic flexible job shop rescheduling under machine breakdown in smart manufacturing plants

Poultry manufacturing plants employing Dynamic Flexible Job Shop Scheduling Problems (DFJSSP) face workflow disruptions due to unexpected machine failures. Efficient rescheduling algorithms are essential to reallocate operations and minimize disruptions. This study proposes two machine failure handling strategies utilizing Grey Wolf Optimization (GWO) and Particle Swarm Optimization (PSO) techniques. Initially, the proposed algorithms generate a healthy state schedule that is executed and monitored in the manufacturing plant. In the event of a machine failure, the digital model of the system triggers a rescheduling process. In the first scenario, the finished operations are excluded out of the schedule, and the remaining operations are optimally rescheduled, excluding faulty machines. The results showed that GWO emphasizes aggressive makespan reduction but at higher energy costs, while PSO provides a more balanced trade-off with slightly longer makespan but lower energy consumption. In the second scenario, an operation-shifting technique is applied; disrupted operations are rescheduled on alternative machines, while following operations retain their initial assignments but are delayed. Both scenarios incorporate operational constraints and evaluate energy consumption to ensure efficiency. Accordingly, an energy consumption analysis report is provided to the decision makers to select the best scenario. The algorithm is implemented and tested under varying failure conditions. Both scenarios, with the proposed optimization algorithms, demonstrate effective rescheduling, with energy consumption analysis confirming rational energy use. As a confirmative step, the proposed GWO and PSO algorithms have been applied to the standard Brandimarte benchmark test cases with different problem sizes. The results proved the validity of the algorithms. Then, the system’s performance has been investigated under different disruption times and failure scenarios. The proposed rescheduling algorithm proves robust and superior in handling machine failures. It minimizes workflow disruptions, ensures operational feasibility, and optimizes energy consumption, making it a reliable solution for poultry manufacturing plants. Finally, to validate the efficiency and effectiveness of the proposed algorithm, we compare the outcomes of the two scenarios by applying them to the same problem and analyzing the behavior of each rescheduling strategy. Two powerful optimization techniques, GWO and PSO, were employed to assess the robustness of the rescheduling plan. Both techniques produced very similar results, demonstrating the consistency and reliability of the proposed approach.