VDTTE: A Vessel Travel Time Estimation Method Based On DeepTTE

Maritime transportation plays a critical role in global cargo logistics. As a key mode of international transportation, its efficiency and reliability are crucial for maintaining global economic stability. Accurate vessel travel time estimation (TTE) is critical not only for optimizing port operations but also for reducing transportation costs, improving operational efficiency, and enhancing customer satisfaction. Unlike land transportation, maritime transport is not constrained by fixed routes and is influenced by uncertain environmental factors, including wind speed, ocean currents, and weather conditions. To tackle these challenges, this paper introduces enhancements to GPS-based TTE methods initially developed for land vehicles. We adapt land-based TTE methods to process Automatic Identification System (AIS) data specific to maritime vessels. A Multi-Layer Perceptron (MLP) model is introduced to capture multiple factors affecting vessel travel time, such as weather conditions, sea states, and vessel characteristics. Experimental results show that the proposed model outperforms traditional approaches in prediction accuracy.