考虑随机需求和时间约束LIRP联合协同优化  

LIRP Joint Collaborative Optimization under Stochastic Demand and Time Constraints

作　　者：焦玉玲[1] 张林静 邢小翠 JIAO Yuling;ZHANG Linjing;XING Xiaocui(Transportation College of Jilin University,Changchun 130022,China)

机构地区：[1]吉林大学交通学院,吉林长春130022

出　　处：《西南交通大学学报》2020年第5期963-970,共8页Journal of Southwest Jiaotong University

基　　金：国家自然科学基金(51775238)。

摘　　要：针对多节点多层次多功能的供应链管理中整体效益亟待提高的问题,结合某连锁超市的单一供应商、多分销中心、多零售店所构成二级分销网络,建立以系统总成本和供货时间为目标的多目标选址-库存-路径问题(location-inventory-routing problem,LIRP)集成规划模型.利用线性加权法将其转化为单目标规划模型,提出遗传算法和节约里程法相结合的两阶段启发式算法并求解模型,第1阶段用遗传算法求解选址-库存问题,第2阶段用节约里程算法求解车辆路径规划问题.并结合某连锁超市实例,对不同总成本权重下的不同决策方案的分销网络进行LIRP集成优化,优化后的系统方案比原文献的总里程减少了3606.9 km,系统总成本减小了6526.2元,缺货成本降低了124.6元,只有19.7元,验证了模型和算法的有效性.Aiming at improving the overall efficiency of the multi-node,multi-level,and multi-functional supply chain management,a secondary distribution network composed of a single supplier,multiple distribution centers,and multiple retail stores for a chain supermarket wasexplored to establish the multi-objective location-inventoryrouting problem(LIRP)integrated planning model with the objectives of the total system cost and supply time.The linear weighting method was used to transform the model into the single-objective programming one.A twostage heuristic algorithm combining genetic algorithm and mileage saving method was proposed to solve the model.In the first phase,the location-inventory problem was solved by the genetic algorithm,and in the second phase,vehicle routing problem was solved by the mileage saving method.A chain supermarket example was used for the LIRP integration optimization of the distribution network with different decision schemes and total cost weights.Compared the results from a reference,the optimized system scheme reduced the total mileage by 3606.9 km,the total system cost by 6526.2 yuan,and the cost of back orders by 124.6 yuan,being 19.7 yuan,which verifies the model and algorithm.

关 键 词：二级分销网络 选址模型 车辆路径 库存 两阶段启发式算法 遗传算法 节约里程法 

分 类 号：TB491[一般工业技术]