Vessel arrival predictions you can act on
Region-specific AI that learns how vessels move, delivering reliable arrival predictions where other models fall short
Request a Free TrialProblem
Other models are built for coverage, not accuracy
As vessels move into constrained waters, predictions from other models become less
reliable - ETAs drift right at the most operationally-critical stage of the voyage.
The reason is that every port, waterway, and coastal region has its own
navigational patterns and constraints. Other models with global coverage 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.
Accurate arrival predictions require local knowledge. And it’s exactly
where other models struggle.
Solution
Arrival predictions grounded in regional behaviour
We build AI models that learn how vessels move in specific waterways from AIS
data. 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 move
in an area.
The result is data-driven arrival predictions with confidence bounds, providing the most
accurate and reliable view throughout the operationally-critical stage of a voyage.
Regional Intelligence
Traffic patterns, navigational constraints, and local operating practices captured at a level of detail that other models can't reach.
AIS-Driven Learning
Real vessel behaviour, across vessel types and operating contexts, learned directly from carefully processed AIS data.
Continuously Updated
Rapidly retrained and redeployed to stay aligned with changing traffic patterns, regulations, and conditions.
Use Cases
Wherever ETAs matter, our models deliver
Our models support any application that requires reliable vessel arrival predictions within a region.
-
Port operator - improve berth scheduling and reduce operational disruption with continuously updated arrival predictions to flexible locations, such as pilot boarding stations or individual berths.
-
Commodity trader - reduce scheduling uncertainty in the critical final hours of a voyage, even with infrequent AIS broadcasts.
-
Autonomous vessel - enable safe navigation in complex environments by embedding localised navigational practices and validating navigation systems against modelled vessel movement.
Workflow
Arrival predictions without changing how you work
Our models are self-contained and straightforward to integrate and use. They slot into existing planning, monitoring, and decision-support systems with minimal effort - no specialist knowledge or additional infrastructure required. Edge deployment is also supported, enabling predictions even when offline.
Generate predictions with:
-
Viewer - no-code, browser-based interface for generating and visualising individual predictions powered by our API.
-
API - low-code integration into existing systems. See the Docs for details.
Start by supplying:
-
Vessel characteristics - vessel type, length, and speed over ground.
-
Transit information - live or historical vessel position and a destination, supplied as simple coordinates.
Vessel positions and other required information can be freely obtained directly from MarineTraffic.
Models
Tested against real arrivals
Every model is validated against thousands of real vessel transits before deployment, with results aggregated for robust evaluation. Each model is benchmarked on two tasks:
-
ETA prediction - how accurately the model estimates arrival time, benchmarked against a historical baseline for the same transit.
-
Trajectory reconstruction - how accurately the model recreates a vessel transit from start and end coordinates alone.
