A Data-Driven Framework for the Evaluation of Autonomous and Traditional Berthing Maneuvers in Maritime Operations

Climate change is among the most urgent challenges of our era, and maritime transport represents a significant source of GreenHouse-Gas and pollutant emissions despite its relative efficiency. To drive both operational and environmental sustainability, a unified, data-driven framework for evaluating ship Berthing Manoeuvre (BM) is presented. The aim of this work is to provide a decision-support system for comprehensive BM evaluation. A structured set of economic, operational, and environmental KPIs is assessed, and both CRITIC and Entropy methods are applied to determine their weights, yielding a single Berthing Efficiency Index (BEI). A modular Python implementation collects IoT and onboard sensor data, cleans and segments maneuvers, computes KPIs, detects anomalies, and presents findings via an interactive dashboard. Validation employed datasets from twin vessels - one performing conventional manual berthing and the other assisted by automated systems. Clustering analysis revealed distinct performance profiles, while BEI aggregation demonstrated the automated system's superior efficiency: lower CO2 emissions and reduced fuel consumption. Cross-fleet benchmarking further confirmed the framework accuracy, reusability, and scalability across diverse operational scenarios. By identifying inefficiencies and delivering evidence-based insights, this approach facilitates optimized berthing practices, targeted crew training, and informed technology adoption, thereby contributing to smarter, greener, and more resilient maritime operations.