Project Leaders
The project is led by the research team at the School of Engineering at the University of British Columbia – Okanagan (UBCO).
Project Description
TrajReducer is an artificial intelligence system designed to enhance the accuracy of vessel trajectory prediction at Canadian ports. The system analyzes thousands of vessel movements using advanced spatial clustering techniques and metadata classification. It considers factors such as vessel type, size, speed, and direction, comparing the current route with similar past voyages to accurately estimate the destination and estimated time of arrival (ETA), even shortly after departure.
During its validation phase, the project demonstrated a significant improvement in port operational efficiency, reducing vessel waiting times by up to 21% (with results recently published in Science Direct). The system has been successfully tested in simulated environments and is currently being prepared for implementation in real ports.
This system represents a major step forward compared to traditional models, which can be inefficient or imprecise. TrajReducer stands out for its computational efficiency and its ability to learn and improve continuously as it processes new data. This ongoing learning enables the system to adapt to global changes in trade patterns and maritime operations.
Project Goals
The core goal of this project is to improve the operational efficiency of Canadian ports and the global supply chain by providing highly accurate predictions of vessel arrival times (ETA) and trajectories.
Moreover, the system adapts to global trade shifts —such as changes in infrastructure or trade agreements— and improves over time as it ingests more data, supporting better planning and responsiveness in the face of global supply chain disruptions.
Publication Date
July 2, 2025