Research on the Application of Weighted Distance K-means Clustering Algorithm with Capacity Constraint in Express Service Location

Clustering is an unsupervised machine learning method, which aims to group data points according to the similarity of data. As a mature clustering algorithm, K-means has been widely used in the location problem. This paper discusses its application in an express service location problem, especially through the improved K-means algorithm to optimize the location decision. The advantage of the algorithm in this paper is that it considers the different influence of the receiving volume and the delivering volume on the cost of the express logistics field, the self-defined weighted distance is introduced to combine the express transportation distance with the express delivery and receipt index, which more accurately reflects the economic cost and benefit of the express delivery enterprise. On this basis of the weighted distance, the algorithm is customized and improved through a capacity constraint in order to satisfy the location demand of the express delivery enterprise. Finally, a numeric experiment shows the effectiveness and practicality of our algorithm on the express service location problem.