machine learning–based evaluation of digitalization and port performance in iran

The rapid digital transformation of ports is reshaping maritime logistics, operational efficiency, and global trade competitiveness. this study evaluates the digitalization and performance of iranian ports using a comprehensive machine learning approach, integrating operational, economic, digital, and environmental indicators. data from 2018–2023, including throughput, turnaround times, ict adoption, port community systems, operating costs, and co₂ emissions, were analyzed using k-means clustering and gaussian mixture models to identify performance patterns and group ports into distinct clusters. results reveal three clusters—advanced, medium, and low-performing ports—with shahid rajaee port leading in digital adoption, efficiency, and sustainability. medium-performing ports demonstrate partial digitalization, while smaller ports face significant operational and environmental challenges. the findings highlight the strategic importance of targeted investment, capacity-building, and technology integration to foster a smart port ecosystem in iran. machine learning-based evaluation provides actionable insights for policymakers and port authorities, enabling data-driven decision-making and sustainable operational improvements.