基于混合粒子群算法解决多目标装配线平衡问题

Use hybrid PSO algorithm for solving the multi-objective simple assembly line balancing problem

针对第一类装配线平衡问题,提出一种混合的粒子群算法,该算法在标准粒子群算法的基础上对其进行离散化,并提出一种基于优先权重的编码方式,考虑到粒子群算法易陷入局部最优的特点,采用遗传算法的全局搜索能力加强粒子群的全局搜索,再利用变邻域搜索算法进行局部搜索,提高算法的搜索性能。另外,在目标函数方面,在最小化工位数的基础上增加平滑指数这一目标函数,使装配线的效率进一步提高。最后通过算例比较,表明混合粒子群算法能够有效地解决第一类装配线平衡问题。

For solving the Assembly Line Balancing Problem of type 1 （ ALBP-1 ）, an improved hybrid Partical Swarm Optimization （PSO） algorithm is proposed. This algorithm is a discrete PSO and priority weight encoding method is employed. Genetic Algorithm （GA） and Variable Neighborhood Search （VNS） algorithm are used to improve the searching performance of PSO. Moreover, in terms of objective function, smoothness index is added to improve the efficiency of assembly line. At last, by comparing severial samples,the results show that hybrid PSO algorithm can solve ALBP-1 efficiently.