CREATING DELIVERY ROUTES USING CLUSTERING ALGORITHMS

Abstract

Clustering is the process of organizing delivery points into groups (clusters) by certain attributes based on their characteristics. As a result, the route is planned not by the whole set of points, but in parts within each cluster, which significantly reduces the complexity of the problem. The aim of the study is to identify the optimal clustering method of delivery objects and use it to create delivery routes for the city's retail network.

In this article, the work of the DBSCAN and k-means algorithms is considered. A comparative analysis showed that DBSCAN does not provide a stable result with a high density of points, whereas the k-means method demonstrates clear clusters. The optimal number of clusters is determined by the “elbow” and “silhouette” methods. The implementation of algorithms in the Python programming language and the subsequent use of the “Branch-and-bound” method will reduce the mileage of cars on routes and reduce transportation costs for the delivery of finished products.