Improved whale optimisation algorithm for two-sided disassembly line balancing problems considering part characteristic indexes

• Published in: International journal of production research 2022-04, Vol.60 (8), p.2553-2571
• Main Authors: Zhang, Yu, Zhang, Zeqiang, Guan, Chao, Xu, Peiyu
• Format: Article
• Language: English
• Subjects: Algorithms; Dismantling; Integer programming; Mixed integer; mixed-integer programming model; multi-objective optimisation; Optimization; part characteristic indexes; Two-sided disassembly line; whale optimisation algorithm
• ISSN: 0020-7543; 1366-588X
• DOI: 10.1080/00207543.2021.1897178

An effective method for disassembling large-sized waste products is to utilise two-sided disassembly lines. A mixed-integer programming model for two-sided disassembly lines is constructed in this study. The model considers four optimisation objectives: the number of mated-stations, idle index, demand index, and hazard index. The model redefines the hazard and demand indexes and adds new time constraints to the mated-station. In view of the complexity of the problem, a multi-objective improved whale algorithm is proposed, in which two different crossover operators are designed to simulate the predatory behaviour of whales to improve the efficiency of the solution. Then, a disturbance factor is introduced to reduce the probability of the population falling into a local optimum. The validity of the model and the effectiveness of the algorithm are verified by comparing the calculation results of the GUROBI solver with those of the proposed algorithm for two-sided disassembly line examples of different sizes. Subsequently, the algorithm is used to solve large-scale linear problems, and the results are compared with those of other algorithms to verify the superiority of the proposed algorithm. Finally, the model and algorithm are applied to a two-sided disassembly of an engine, and several optimal allocation schemes are obtained.