Energy-Efficient Lot-Streaming Scheduling Method of Multi-Resource Constrained Flexible Job Shop

Abstract

To cope with the problems of high-computational complexity and multiple locally optimal solutions induced by the coupling of multiple subproblems, conflicting objectives, and integration of resource constraints of the energy-efficient lot-streaming scheduling of multi-resource constrained flexible job shop ( $\gamma $ -shop for short), an energy-efficient lot-streaming scheduling optimization approach based on the knowledge-based lot-splitting method (KLSM) and the improved multiobjective evolutionary algorithm (IMOEA) is presented. First, a flexible job shop lot-splitting scheduling model with the optimization objectives of total energy consumption, makespan, and total processing cost is formulated. Second, a hybrid approach of the KLSM and the IMOEA is designed to solve the model. The solution space of the problem is fully explored based on the moth-flame operator. Co-evolutionary operators are performed to promote information interaction among populations, hence both the population diversity and the convergence effect of the algorithm are improved. Moreover, a post-adjustment strategy based on adjacent processes is developed to reduce unnecessary fixture changes. Finally, extended experiments between some KLSM-based well-known and novel algorithms, including the proposed IMOEA, MOEA/D, NSGA-II, MOPSO, SGECF, SCEA, and SLMEA are conducted in benchmark problems and a real-world case of machine tool plant. The results show that the proposed method outperforms its competitors on co-optimization of lot-splitting, machine allocation, operation sequencing, and fixture assignment of the $\gamma $ -shop scheduling, which can effectively reduce total energy consumption, makespan, and total processing cost.