摘要
在对遗传算法、生成树遗传算法和混合进化方法进行比较的基础上,提出组合遗传算法来解决大规模基本物流网络设计问题。该问题抽象为“选址—分派问题”,并进一步分解,且将“分派”镶嵌在“选址”中。“选址”和“分派”染色体分别使用二进制编码和矩阵编码、适值采用物流费用。运算结果显示该方法比混合进化方法结果更精确,且在大规模问题求解方面速度优于通常的遗传算法,因此是一种设计大规模物流网络的较好方法。
On the basis of comparison among GA, Spanning Tree-based Genetic Algorithm （ST-GA）and Hybrid Evolutionary Method（HEM）, the article put forward the idea of Combinatorial Genetic Algorithms（C-GA） and designed the program in detail for the design,work of large scale logistics network.. Such work could be abstracted as Location-Allocation Problems. With further separation analysis, the paper made Allocation Problem inserted in the Location part. The Allocation ＆ Location chromosomes could respectively use the binary coding form and the matrix coding form. The Fit Value is the whole cost of logistics. Result shows that C-GA method could get more precise result than HEM, and could save much more time than general GA, under the condition of solving large scale problem. Therefore, the C-GA is a kind of better methods to design large scale logistics network at present.
出处
《工业工程与管理》
2006年第4期46-49,共4页
Industrial Engineering and Management
基金
国家自然科学基金项目(70573063)
中国博士后科学基金资助项目(2004036264)
关键词
物流网络
选址-分派问题
遗传算法
logistics networks
location-allocation problems
genetic algorithms
