An improved differential evolution algorithm for solving a distributed assembly flexible job shop scheduling problem

• Published in: Memetic computing Vol. 11; no. 4; pp. 335 - 355
• Main Authors: Wu, Xiuli, Liu, Xiajing, Zhao, Ning
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2019; Springer Nature B.V
• Subjects: Algorithms; Applications of Mathematics; Artificial Intelligence; Assembly; Bioinformatics; Complex Systems; Computer simulation; Control; Crossovers; Engineering; Evolutionary algorithms; Evolutionary computation; Globalization; Integer programming; Job shop scheduling; Job shops; Linear programming; Mathematical and Computational Engineering; Mechatronics; Mixed integer; Mutation; Production scheduling; Regular Research Paper; Robotics; Simulated annealing; Earliness/tardiness; Distributed assembly flexible job shop scheduling problem; Improved differential evolution algorithm; Cost; Balanced algorithm
• ISSN: 1865-9284; 1865-9292
• DOI: 10.1007/s12293-018-00278-7

The single-factory manufacturing is gradually transiting to the multi-factory collaborative production with the globalization. The decentralization of resources and the heterogeneity of the production modes make it difficult to solve this kind of problem. Therefore, the distributed assembly flexible job shop scheduling problem (DAFJSP) is studied. DAFJSP can be decomposed into several flexible job shop scheduling problems and several single machine factory scheduling problems. To begin with, a mixed integer linear programming model for the DAFJSP is formulated to minimize the earliness/tardiness and the total cost simultaneously. Then, an improved differential evolution simulated annealing algorithm (IDESAA) is proposed. The balanced scheduling algorithm is designed to trade off the two objectives. Two crossover and mutation operators are designed. Due to its strong robustness, simulated annealing is integrated to local search the best Pareto solutions. The greedy idea combined with the Non-Dominated Sorted selection is employed to select the offspring. Finally, comprehensive experiments are conducted and the results show that the proposed algorithm can solve DAFJSP effectively and efficiently.