International Journal of Production Management and Engineering | 2026
Authors: Alzate P.; Guevara A.; Isaza G.A.; Toro E.M.; Jaramillo-Garzón J.A.
DOI: 10.4995/ijpme.2026.23227
Journal: International Journal of Production Management and Engineering
Year: 2026
Publisher: Universitat Politecnica de Valencia
Document Type: Article
Open Access: All Open Access; Gold Open Access; Green Open Access
Cited by: 0
Efficient maritime route planning is pivotal for optimising global logistics operations in an increasingly competitive environment. This study aims to determine the optimal number of clusters among 90 georeferenced ports across six continents, minimizing distances and enhancing operational efficiency through advanced clustering techniques. Methods such as geographic k-means, Gaussian Mixture Models (GMM), and hierarchical clustering were implemented and evaluated using quality metrics like the Calinski-Harabasz, Silhouette, and Davies-Bouldin indices. The results identified 13 optimal clusters that reflect logical and geographically coherent segmentation, highlighting geographic k-means for its consistency and replicability. Additionally, integrating an intelligent routing model enabled the design of optimised routes within each cluster, reducing distances and maximizing logistical efficiency. The proposed methodology, developed in Python, demonstrates not only its applicability to maritime operations but also its potential for extrapolation to other contexts, showcasing robustness and relevance for global strategic planning. This approach represents a significant advancement in integrating clustering and optimisation techniques to enhance maritime supply chain management. © 2026 Universitat Politecnica de Valencia. All rights reserved.
intelligent routing; Logistics clustering; maritime route optimisation; maritime transportation; operational efficiency