Problem

Global approaches are built for coverage, not accuracy

As vessels move into constrained waters, predictions from global providers become less reliable. Routes diverge from reality and ETAs drift, right at the most operationally critical stage of the journey.

The reason is straightforward: every port, waterway, and coastal region has its own navigational patterns and constraints. Global models can't capture all of this detail, so they simplify. That simplification is manageable in open water - but as vessels approach their destination, inaccuracies compound fast.

Accuracy requires local detail. And it’s exactly where global approaches struggle.

Solution

Region-specific AI models built on real behaviour

We build AI models trained on local AIS data that learn how different vessel types move in specific waterways. These models are not only specialised by vessel type and operational context, but by geography, ensuring predictions are finely tuned to the reality of how vessels actually move in an area.

This approach allows our models to perform a variety of important prediction tasks:

• Trajectory - predict the most likely route a vessel will take between two locations.
• ETA - estimate arrival times with data-driven confidence bounds by vessel type.
• Plan - generate fuel-efficient passage plans aligned with a target transit time to meet schedules.

Our approach delivers clear advantages over existing approaches:

Use Cases

Our models support any application that requires reliable vessel movement predictions within a region.

• Port operator - optimise berth scheduling and reduce operational disruption with continuously updated trajectory predictions, ETAs, and confidence bounds.
• Fleet manager - improve fuel efficiency and schedule adherence by generating data-driven passage plans and monitoring route deviations.
• Commodity trader - reduce scheduling uncertainty with data-driven ETAs and confidence bounds during the critical final hours of a voyage, even with old AIS broadcasts.
• Autonomous vessel - enable safe navigation in complex environments by embedding localised navigational practices and validating navigation systems against modelled vessel behaviour.

Workflow

Predictions without changing how you work

Our models slot into existing workflows and systems with minimal effort - self-contained, simple to integrate, and ready to use without specialist knowledge or additional infrastructure. They can also be deployed on edge devices, allowing predictions even when offline.

Generate predictions with:

• Viewer - a no-code, browser-based interface for generating and visualising individual predictions powered by our APIs.
• APIs - low-code integration into planning, monitoring, or decision-support systems. See the Docs for details.

Start by supplying:

• Vessel type and length category - predictions account for specific vessel characteristics.
• Locations - a live or historical vessel position, plus a current or intended destination, supplied as simple coordinates.

Vessel positions and other information can be obtained directly from MarineTraffic.

Models

Tested against reality

Our models undergo comprehensive testing before deployment, using thousands of randomly selected real vessel transits with results aggregated for robust evaluation. To ensure reliability, each model is benchmarked on two key tasks:

• Trajectory reconstruction - assessing how accurately the model can recreate vessel tracks from start to finish, which forms the core foundation of our predictions.
• ETA prediction - evaluating how accurately the model estimates vessel arrival times by building on the predicted route and capturing vessel behaviour throughout the journey.