考虑物品异质性的联合补货-配送集成优化模型与智能求解算法  

Integrated Optimization Model and Intelligent Algorithm for Joint Replenishment Delivery Problem Considering Item Heterogeneity

作　　者：余海 吴锋艳 冷宛佳 吴小含 车木子 王林[2] YU Hai;WU Fengyan;LENG Wanjia;WU Xiaohan;CHE Muzi;WANG Lin(Material Company of State Grid Hubei Electric Power Co.,Ltd.(State Grid Hubei Tendering Co.,Ltd.),Wuhan 430014,China;不详)

机构地区：[1]国网湖北省电力有限公司物资公司(国网湖北招标有限公司),湖北武汉430014 [2]华中科技大学管理学院,湖北武汉430074

出　　处：《武汉理工大学学报（信息与管理工程版）》2024年第5期734-741,749,共9页Journal of Wuhan University of Technology：Information & Management Engineering

基　　金：国网湖北省电力有限公司科技项目(SGHBWZ00CGJS2310065)。

摘　　要：多物品联合补货策略通过协调不同物品的补货周期实现规模效益,可以有效降低物流成本。在多物品联合补货系统中,物品异质性是一个难以忽视的问题。物品异质性常常导致物品损坏、质量损失、运输风险等问题。针对这一现象,构建一种考虑资源约束的异质品联合补货-配送模型。为求解该NP-hard问题,设计一种基于自适应重组与反向学习策略的骨干差分进化算法。实验结果表明,新算法在运算时间和运算精度上比几种对比算法都有明显提升;算法表现与主要订货成本存在明显相关性,将该参数引入自适应重组策略有望进一步改进算法性能。The multi-item joint replenishment strategy achieves economics of scale by coordinating the replenishment cycle of different items,which can effectively reduce logistics costs.However,item heterogeneity is a difficult problem to ignore in such multi-item joint replenishment systems.Item heterogeneity often leads to item damage,quality loss,transportation risks,and etc.To address this issue,a capacitated joint replenishment and delivery model with heterogeneous items was proposed.A backbone differential evolution algorithm with self-adaptive crossover and opposite-learning strategy was designed to solve this NP-hard problem.The experimental results showed that the new algorithm has a significant improvement in operation time and operation accuracy compared with several comparison algorithms.There is a significant correlation between the performance of the algorithm and the main order cost.Introducing this parameter into the adaptive recombination strategy is expected to further improve the performance of the algorithm.

关 键 词：联合补货-配送问题 异质品 资源约束 自适应算子 骨干差分进化算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] F224[自动化与计算机技术—控制科学与工程]