基于群体编码方式的遗传算法求解装箱问题  

Genetic algorithm based on group coding for bin-packing problem

作　　者：张大斌[1] 刘桂琴[1] 王婧[1] 朱侯[1] 

机构地区：[1]华中师范大学信息管理系,湖北武汉430079

出　　处：《计算机工程与设计》2008年第12期3154-3156,共3页Computer Engineering and Design

摘　　要：就传统装箱问题提出了一种改进的混合遗传算法。该改进的遗传算法主要通过基于群体的编码方法,将每一个箱子视为一个群体,使得在进化过程中,只对群体部分进行操作,从而大大减少了冗余,克服了传统的基于物品编码或基于箱子编码的遗传算法的不足。仿真实验结果表明,该编码方式的遗传算法比基于物品编码的遗传算法显著提高了优化效果。An improved hybrid genetic algorithm is proposed to solve the traditional bin-packing problem. By using a new group-based coding method and considering every bin as a group, the proposed algorithm only operate on the group during the evolvement, to avoid the disadvantages in the bin-based representation or object-based representation. According to computational experiment result, the improved hybrid genetic algorithm can improve the optimized effect of the traditional bin-packing problem obviously in comparison with object-based genetic algorithms.

关 键 词：遗传算法 装箱问题 编码 群体操作 组合优化 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]