基于改进K-means算法的物流配送中心选址研究  

Research on Logistics Distribution Center Location Selection Based on Improved K-means Algorithm

作　　者：姚佼[1] 吴秀荣 李皓 谢贝贝 王诗璇 梁益铭 YAO Jiao;WU Xiurong;LI Hao;XIE Beibei;WANG Shixuan;LIANG Yiming(School of Management,University of Shanghai for Science and Technology,Shanghai 200093,China;China Railway Jinan Group Co.,Ltd,Jinan 250000,China)

机构地区：[1]上海理工大学管理学院,上海200093 [2]中国铁路济南局集团有限公司,山东济南250000

出　　处：《物流科技》2024年第5期10-13,19,共5页Logistics Sci-Tech

基　　金：教育部人文社会科学规划基金项目(22YJAZH131)。

摘　　要：针对传统K-means算法需要主观设定K值及无法处理类别型数据问题,文章运用肘部法及轮廓系数法确定合理K值,对类别型数据采取独热编码(One-Hot Encoding)转换为可以处理的连续型数据,并将其运用到在物流配送中心选址中;并综合考虑多种类别的影响因素,构建了相应的影响因素指标体系,提出的模型能够识别输入数据的数值型及类别型数据,实现样本的有效聚类。相关的案例分析结果表明,相比传统K-means聚类,文章的改进K-means算法选址结果可使物流总成本降低8.76%,运营成本降低14.85%,固定成本降低8.09%,效果显著。To address the limitations of the traditional K-means algorithm,such as the subjective determination of the number of clusters K and its inability to handle categorical data,this research utilize the elbow method and silhouette coefficient method to determine an optimal value for K.Categorical data was transformed into continuous data that can be processed by using one-hot encoding,and this approach was applied in the site selection of logistics distribution centers.Furthermore,a comprehensive consideration of various influencing factors was incorporated by constructing a corresponding index system for these factors.The proposed model is capable of identifying both numerical and categorical data in the input dataset,enabling effective clustering of samples.The results of the case analysis demonstrate that,compared to traditional K-means clustering,the improved K-means algorithm in this study yields significant benefits in terms of site selection for logistics centers.Specifically,the results show a reduction of 8.76%in overall logistics costs,a 14.85%decrease in operational costs,and an 8.09%decrease in fixed costs.These findings indicate a notable improvement in performance.

关 键 词：物流配送中心选址 K-MEANS聚类算法 肘部法 轮廓系数法 独热编码 

分 类 号：F252.14[经济管理—国民经济]