基于学习型免疫算法的物流配送管理  

Logistics distribution management on the basis of the learning-based immune algorithm

作　　者：李震[1] 马洪波[1] LI Zhen;MA Hong-bo(Engineering & Technical College of Chengdu University of Technology,Leshan 614007,China)

机构地区：[1]成都理工大学工程技术学院,四川乐山614007

出　　处：《电子设计工程》2019年第6期113-115,124,共4页Electronic Design Engineering

摘　　要：为了提高物流运营的效益,将信息处理技术融入到物流管理体系。采用学习型免疫算法进行合理规划物流配送中心,在进行抗体克隆时,克隆数的选取不是预先设定,而是取决于父抗体的亲和度;在抗体的学习训练机制中,每个抗体变异的速率与抗原的激励水平成反比。仿真实验证明:以20个城市为例,通过免疫学习算法从中筛选出3个城市作为配送中心,对应的坐标分别为:(1 752,1 600),(2 766,936),(3 358,1 559),算法的最优适应度值收敛于3.486e+5。In order to improve the efficiency of logistics operation,the information processing technology is integrated into the logistics management system.The learning-based immune algorithm is used to plan the logistics distribution center,in which during the cloning of antibodies,the selection of the number of clones is not predetermined,but depends on the affinity of the parent antibody;and in the learning and training mechanism of antibody,the rate of variation of each antibody is inversely proportional to the level of the antigen.The result of the simulation experiment shows that three cities were selected as distribution centers through the immune learning algorithm from 20 cities,and their corresponding coordinates respectively were(1 752,1 600),(2 766,936)and(3 358,1 559),and the optimal fitness value of the algorithm converges to 3.486e+5.

关 键 词：学习型免疫算法 物流管理 克隆数 抗体变异 最优适应度值 

分 类 号：TN911[电子电信—通信与信息系统]