Multi-objective discrete strength pareto evolutionary algorithm II for multiple-product partial U-shaped disassembly line balancing problem

Abstract

In disassembly industry, the disassembly efficiency closely depends on a reasonable disassembly mode and an efficient disassembly line. In the current manufacturing, the profit of multi-product partial U-shaped disassembly-line-balancing (MPUD) is relatively low and the energy consumption is high. According to the characteristics of the MPUD problem, a mathematical model for maximizing profit and minimizing energy consumption of human-robot disassembly is established. To solve the MPUD problem, a multi-objective strength Pareto evolutionary algorithm II is developed and crossover and mutation operators are improved to promote the convergence and feasibility of the algorithm. We perform the experiments on a ballpoint pen and a washing machine to simulate the disassembly process and compare the proposed algorithm with the nondominated sorting genetic algorithm II and the multi-objective evolutionary algorithm based on decomposition. Experimental results show that the proposed algorithm has obvious advantages in the most important metrics.