The assembly line balancing ( ALB) problem is an important and difficult problem in production management. How to allocate the workers to stations to obtain the best efficiency of the line is the key to the problem. According to the problem in ALB problems with worker allocation, a multi-objective genetic algorithm based on generalized Pareto-based scale-independent fitness function (gp-siffGA) was proposed. Firstly the algorithm established a mathematical model of the ALB-wa and proposed a random key-based representation method adapting the GA. Following, advanced genetic operators adapted to the specific chromosome structure and the characteristics of the ALB-wa problem were used. Moreover, Pareto dominance relationship was used to solve the ALB-wa problem instead of using relative preferences of multiple objectives. Finally, the performance of proposed method was validated through numerical experiments. The results indicate that the proposed approach has the high convergence and efficiency, also it can improve the quality of solutions more than the other existing GA approaches.%装配线平衡问题是生产管理中重要且比较难解决的问题,其中如何分配工人到不同的工作站以提高生产效率是问题的关键.针对包含工人分配问题的装配线平衡问题,提出一种基于Pareto的问题无关的适应值计算方法的多目标遗传算法.算法中首先建立ALB-wa问题的数学模型,提出一个基于随机键编码的基因表达方式；使用匹配指定的染色体结构和ALB-wa问题的遗传操作；使用基于Pareto支配关系的评价函数来代替使用基于偏好的评价函数.最后,通过实验数值验证该方法的性能.结果表明,该方法具有较高的收敛性和效率,改进了现有的其他遗传算法.
展开▼
