Biofouling growth is the buildup of marine organisms (e.g., algae, plants, and small animals) on the surface of vessel hulls and propellers and leads to increased drag, fuel consumption, and emissions. Biofouling growth is very hard to predict because of changes in the environment and vessel routine, or monitor, because the hull is submerged, or the vessel is deployed somewhere it cannot be inspected. Moreover, even when biofouling is discovered it is very hard to assess the impact of the biofouling on the vessel’s performance, because performance is dependent on many factors, e.g., wind, waves, currents, operational profile, and power system components. 

Forget noon reports (and ISO 19030). Research has shown this standard is not accurate or effective. We tap into the on-board data stream that collects fundamental information on location, environmental conditions, and the state of the vessel's powertrain. We recommend mechanical components to be monitored at a rate of approximately once per minute, electromechanical components at once per second, and electrical components at megahertz frequencies. Additionally, because not all ships have the same sensor networks, our methods are flexible to whatever data you have available.

More data is always better!

For assessing the hull and propeller cleaning effectiveness we recommend around 6 months of available sailing data (3 months in the fouled state and a 3 month period after cleaning to assess the effectiveness).

For scheduling we recommend around a year of sailing data to properly capture the operational and environmental varaibility.

However, our models can always be retrained and improved as more data becomes available so don't hesitate to reach out if you are just starting to collect data.

With the right datasets and data availability our models are scientifically proven to have over 99% accuracy. However, our value is proven by adapting to whatever data you have available. Our case studies have shown us that even with limited feature sets (gps, log speed, engine rpm, fuel consumption, etc.) our models are still within 3-4% error which allows us to get the best insights out of your data.

Our models leverage cutting edge hybrid AI research carried out at Delft University of Technology. We combine the best of marine engineering and AI research to get the most insights out of your data.