Multistage operation-based genetic algorithms: Optimizing advanced planning and scheduling

The planning and scheduling of manufacturing systems always require resource capacity constraints, disjunctive constraints and precedence constraints, owing to the tight due dates, multiple customer-specific orders, and flexible process strategies. Advanced Planning and Scheduling (APS) mainly supports the integrated, constraint-based and optimal planning of the manufacturing system to reduce lead times, lower inventories, and to increase throughput. It is very important role for treating a range of capabilities from finite-capacity scheduling at the shop floor level through to constraint-based planning in a multi-plant chain. In this paper, a novel approach for designing a chromosome is proposed to improve the effectiveness, which called multistage operation-based genetic algorithm (moGA). The objective is to find the optimal resource selection for assignments, operations sequences, and allocation of variable transfer batches, in order to minimize the makespan, considering the setup time, transportation time, and operations processing time. The plans and schedules are designed considering flexible flows, resources status, capacities of plants, precedence constraints, and workload balance. Moreover, we combined moGA with an effective Fuzzy Logic Controller (FLC) for improving the efficiency of an evolutional process in GA. The experimental result shows that proposed hybrid moGA+FLC can converge to optimal solution in higher speed. Furthermore we compare our approach with Moon-Kim-Gen's approach under the same experimental data. The results of various sizes of numerical experiments will demonstrate the efficiency of the hybrid moGA+FLC by comparing with the previous methods.