Abstract

Maritime trade plays a crucial role in the global economy, and recent technological developments have accelerated marine logistics. However, this increase impacted port performance, leading to port congestion in some regions and distorting the smooth flow of maritime logistics. Few studies employing AIS data have explored marine traffic congestion; hence, developing a system that makes port metrics more accessible is needed. This work employs a methodology to analyze the port congestion level of Rio de Janeiro. Three algorithms were developed using the Automatic Identification System (AIS) data to identify the geolocation area, the convex hull area, and the average vessel's proximity. These algorithms were used to calculate the Port Congestion Indicators (PCIs): spatial concentration, spatial density, average service time, and Machine Learning techniques were employed to extract knowledge from the database. As a result, this process identified the periods when ports are most congested, and the centroids of these clusters can be used to predict future congestion levels. These indicators provide resources for better management and can motivate actions such as the redistribution of ship loading and unloading locations and improving port performance measurement.