AI in Shipping Ports: Predictive Visibility
- Canadian ports may soon see smoother operations thanks to an artificial intelligence system developed at UBC Okanagan.
- Zheng Lui, a professor at UBCO's School of Engineering, and doctoral student Chengkai zhang, created the system.
- According to Lui,the research,published in Ocean Engineering,addresses a critical need in maritime logistics.
AI innovation Helps Shipping Ports See What’s Coming
Canadian ports may soon see smoother operations thanks to an artificial intelligence system developed at UBC Okanagan. The system, dubbed TrajReducer, accurately predicts ship destinations and arrival times, possibly mitigating global supply chain disruptions.
Dr. Zheng Lui, a professor at UBCO’s School of Engineering, and doctoral student Chengkai zhang, created the system. TrajReducer enhances both prediction accuracy and computational efficiency by analyzing ship trajectories using spatial clustering and metadata ranking.
According to Lui,the research,published in Ocean Engineering,addresses a critical need in maritime logistics. Accurate predictions of vessel arrival times are essential for improving port operations.
Customary methods often lack precision,with about 30% of data missing estimated arrival and departure times,Lui said.
TrajReducer overcomes these issues by studying patterns from thousands of voyages, incorporating vessel details like type, size, speed, and direction.
The system compares a ship’s current path to similar past voyages, calculating its destination with high accuracy, even early in the journey. This technology is especially relevant for major Canadian ports like Vancouver, Prince Rupert, Montreal, and Halifax, which are critical gateways for North American trade.
Lui noted that even small improvements in operational efficiency can yield significant economic benefits, as Canadian ports handle hundreds of millions of tons of cargo annually.advance knowledge of ship arrivals allows ports to optimize berth assignments, equipment arrangements, and coordination with rail and trucking networks, speeding up product delivery.
Zhang added that the system becomes more accurate as it processes more data. TrajReducer adapts and improves its predictions as global shipping patterns evolve due to new trade agreements, infrastructure changes, or climate considerations.
“Maritime shipping represents more than 80% of global trade and the ability to accurately predict where cargo vessels are headed and when they will arrive, has never been more important,” lui said.
“This is about more than efficiency—it’s about resilience,” Zhang said. “When Canadian ports can anticipate and prepare for the arrival of ships accurately, they’re better positioned to maintain the flow of goods that Canadians depend on, even when global trade faces unexpected challenges.”
What’s next
The researchers plan to further refine TrajReducer, incorporating real-time weather data and expanding its predictive capabilities to address a wider range of maritime challenges. They envision applications extending beyond port operations to include maritime safety, environmental monitoring, and comprehensive supply chain optimization.