Earliness and Tardiness Minimizing on a Realistic Hybrid Flowshop Scheduling with Learning Effect by Advanced Metaheuristic

Abstract

This study proposed a novel hybrid metaheuristic that hybridized the particle swarm optimization, simulated annealing and variable neighborhood search to solve the hybrid flowshop scheduling with sequence-dependent setup times (SDST). To address the realistic assumptions of the proposed problem, three additional traits were added to the scheduling problem. These include SDST, position-dependent learning effects (LEs), and the consideration of tardiness together with earliness penalties as objective function. According to the best of our knowledge, this problem has never been investigated in the hybrid flowshop. Considering position-dependent LEs, it is assumed that the learning process reflects a decrease in the process and setup times as a function of the number of repetitions of the production of a same operation in a same stage because in many realistic situations, the more time you practice, the better LE you obtain. To evaluate the performance of the suggested method, the hybrid simulated annealing metaheuristic (HSA) presented recently is investigated for comparison purposes and computational experiments are performed on standard test problems. Results show that the suggested method performs better than the HSA for various test problems.