摘要
随着电子商务的迅速发展,商城配送中心选址已成为提升配送效率、降低成本和增强客户满意度的关键。文中将密度峰值聚类(Density Peak Clustering,DPC)算法应用于商城配送中心选址问题中,通过分析历史订单数据,自动识别订单分布的密集区域,同时引入相似度策略处理复杂数据结构并提高聚类准确性。实验结果表明,改进的DPC算法在Flame数据集上聚类效果优越,相比于K-Means、I-DBSCAN和MeanShift算法,能更有效地识别高密度区域,为商城配送中心选址提供科学依据,从而优化配送中心位置,提高配送效率和客户满意度。
With the rapid development of e-commerce,the site selection of mall distribution center has become the key to improve distribution efficiency,reduce costs and enhance customer satisfaction.This paper applies the Density Peak Clustering(DPC)algorithm to the site selection problem of mall distribution center,which automatically identifies the dense region of order distribution through the analysis of historical order data.At the same time,a similarity strategy is introduced to deal with the complex data structure and improve the clustering accuracy.Experimental results show that the improved DPC algorithm has superior clustering effect on Flame dataset,and it can identify high-density regions more ef‐fectively than K-Means,I-DBSCAN and MeanShift algorithms,which provides a scientific basis for the site selection of mall distribution centers,thus optimizing the location of distribution centers,and improving the distribution efficiency and customer satisfaction.
作者
林泓安
王鑫鑫
LIN Hongan;WANG Xinxin(School of Management,Wuhan University of Science and Technology,Wuhan 430065,China)
出处
《物流工程与管理》
2025年第2期7-10,共4页
Logistics Engineering and Management
关键词
密度峰值聚类算法
相似度策略
商城配送选址
P-中心问题
Density Peak Clustering algorithm
similarity strategy
mall distribution site selection
P-center problem
