多冷链物流中心选址分配及其自适应免疫算法  

Adaptive Immune Algorithm for Multiple Cold Chain Logistics Center Location-Allocation Problem

作　　者：胡利利[1] 夏扬坤[2] HU Li-li;XIA Yang-kun(Management Engineering Department,Hunan Vocational College of Engineering,Changsha 410151,China;School of Traffic and Transportation Engineering,Central South University,Changsha 410075,China)

机构地区：[1]湖南工程职业技术学院管理工程系,湖南长沙410151 [2]中南大学交通运输工程学院,湖南长沙410075

出　　处：《数学的实践与认识》2018年第16期31-39,共9页Mathematics in Practice and Theory

基　　金：中国物流学会课题(2016CSLKT3-077);湖南省教育厅科学研究项目(17C0419);中南大学中央高校基本科研业务费专项资金资助项目(2017zzts198)

摘　　要：物流中心选址是现代物流战略规划的核心,为了提升冷链物流服务能力,实现物流资源整合优化,研究了一种带需求距离积的多冷链物流中心选址-分配问题.针对冷链物流中心的时效性特点,将节点需求量与辐射半径纳入约束,并以需求量作为距离的权重放入目标函数之中,构建了相应的选址-分配数学模型.多物流中心选址-分配属于NP-Hard问题,针对以往精确算法的不足,设计了一个自适应免疫算法.借助生物免疫系统的多样性与自我调节能力,在免疫算法中采用部分匹配原则来计算抗体之间的亲和度,设计自适应惩罚机制、免疫变异和精英保留策略等来保持群体的多样性与精英的优良性.通过实验,找到了交叉概率与变异概率的适宜组合取值,文献对比表明了算法的有效性.The logistics center location is the core of modern logistics strategic planning. In order to improve the service capability of cold chain logistics and integrate the logistics resources, a multiple logistics center location-allocation problem with demand distance prod- uct is studied. According to the timeliness characteristics of the cold chain logistics center, the node demand and radiation radius are included in the constraints. By adding the node demand into the objective function as the weight of the distance, a corresponding mathemat- ical model of location-allocation problem is constructed. Multiple logistics center location- allocation problem belongs to NP-Hard problem. According to the shortcomings of previous accurate algorithm, an adaptive immune algorithm is designed to solve the problem. With the illumination of the biological immune system diversity and self-adjustment ability, the new algorithm uses the partial matching principle to calculate the affinity of the antibodies, and takes some strategies such as adaptive penalty mechanism, immune mutation and elitist retention to maintain the population diversity and the elite excellent. Through experiments, the suitable combination values of crossover probability and mutation probability are found. The comparison of the literature shows the effectiveness of the algorithm.

关 键 词：物流中心 选址 免疫算法 自适应 资源整合 冷链物流 

分 类 号：F252[经济管理—国民经济] TP18[自动化与计算机技术—控制理论与控制工程]