Two-Sided Disassembly Line Balancing Design with a Soft Computing Technique

Abstract

The aims of this research are twofold. First, a mathematical model is developed for the problem of two-sided balancing of disassembly lines for medium-sized products. Second, a metaheuristic algorithm is proposed for solving such a problem. The proposed mathematical model attempts to minimize the number of workstations, paired stations, modified index of work relatedness and workload balance between workstations. The proposed model falls in the domain of mixed-integer linear programming, which can be explained by the fact that it is a 𝑁𝑃-hard problem. We have presented the modified particle swarm optimization, which adds the concept of Pareto optimality to include the non-dominated solutions in the elite list. The proposed method is compared with two competing algorithms: the genetic algorithm and the combinatorial optimization with random algorithm on four benchmark instances. The result, which follows the concept of Pareto optimality, shows that our proposed method provides more non-dominated solutions than the competing algorithms. Moreover, the proposed algorithm exhibits better convergence and diversity than the competing algorithms. In practice, this research work could serve as a basis for designing a two-sided disassembly line for a medium-sized product.