Improved dual-population genetic algorithm to solve human–robot collaborative assembly line balancing problem
研究以工作站数量、完工时间和平衡比为优化目标，构建人机协作装配线平衡模型，并设计改进的双种群遗传算法，实验表明该算法在求解精度和收敛速度上优于传统启发式算法。
The human–robot collaborative assembly line has been applied as an effective strategy in production to further improve the efficiency, adaptability, and flexibility. The assembly line balancing is an important part of assembly line design and optimisation. However, the current research often neglects the key factor of the number of stations, which lacks better guidance for the actual assembly line design and optimisation. To provide a better practical reference, this research first takes the number of stations, the makespan and the balance ratio as the optimisation objectives, and constructs a corresponding model. Then, an improved dual-population genetic algorithm is designed to solve the problem more efficiently. This algorithm optimises multiple operators based on the traditional genetic algorithm. The efficacy of the proposed algorithm is substantiated through numerical experimentation with three traditional heuristic algorithms. The experimental results show that the improved algorithm has superior performance in solving the human–robot collaborative assembly line balancing problem. On the one hand, the average improvement of fitness in different scale cases reaches 3.14% (up to 52.33%), and the distribution of solutions is more stable. On the other hand, the convergence speed is faster, the average reduction of convergence time is 11.50% (up to 88.24%).