Compute Maritime projects big fuel savings for AI-designed CTV

Image: Compute Maritime

A next-generation AI-designed offshore wind crew transfer vessel is projected to save over 100,000 liters of fuel a year, resulting in major CO2 emissions reductions. That’s according to London-based Compute Maritime, whose flagship product is the NeuralShipper AI platform for ship design.

The company has just reported the results of its U.K. Government-funded project, GenDSOM, which brings generative AI and additive manufacturing into ship design.

Working with consortium partners Siemens Digital Industries Software, Rapid Fusion, HP, BYD Naval Architects and the University of Southampton, the project has produced a next-generation crew transfer vessel (CTV) for the offshore wind sector that, in detailed performance modeling, saves 101,671 litters (about 26,861 U.S. gallons) of fuel and 258.7 tonnes of CO2 per vessel every year compared with a conventional baseline.

The designed vessel, a 32.5-meter twin-hull CTV designed by BYD Naval Architects and built to carry 24 offshore wind technicians and four crew, was developed using Compute Maritime’s NeuralShipper AI to optimize its hull form, then paired with a diesel-electric hybrid propulsion system, developed with Siemens Energy.

The combined result is an 11.1% reduction in annual fuel consumption and an 8.9% reduction in CO₂ emissions against a like-for-like conventional diesel vessel operating the same offshore wind duty cycle.

In other words, says Compute Maritime, the new vessel is not simply a performance bonus: it is what makes the clean electric propulsion viable at all.

One finding, says Compute Maritime, underscores why the AI optimization matters so much. Modeled across a full day of operations, the baseline vessel ends the day with a 34 kWh energy deficit, drawing the batteries beyond their safe discharge limit and falling short of the 25-knot service speed. The NeuralShipper-optimised vessel reverses this, finishing the day with a 106 kWh surplus, comfortably within the battery’s operating envelope and delivering full service speed.

AI OPTIMIZATION

At the heart of the saving is the hull NeuralShipper generated and refined a hull form that reduces the power required at the vessel’s 25-knot service speed by 6.3%, with reductions of up to 11.6% at higher speeds. Because a crew transfer vessel spends its working life making repeated high-speed transits to and from wind farms, even a single-digit cut in required power compounds into tens of thousands of liters of fuel saved across a year of operations.

ADDITIVE MANUFACTURING

GenDSOM advances manufacturing as well as design. As part of the project, Compute Maritime and Rapid Fusion developed a complete additive-manufacturing toolset that brings production constraints, such as build volume, support structures and material behavior, directly into the NeuralShipper design loop, so that components are optimized to be printable from the outset rather than reworked for manufacture afterwards. Using this approach, the consortium designed and produced a hydrofoil with Rapid Fusion’s Apollo, a robotic large-format additive manufacturing (LFAM) system that uses high-deposition pellet printing to build large composite and thermoplastic components. The result is a working demonstration of an unbroken pathway from AI-generated geometry to a finished, production-ready marine component.

“When we launched GenDSOM, we set ourselves a clear and deliberately ambitious goal: to prove that generative AI could deliver real, measurable savings in the water, not just faster drawings on a screen,” said Shahroz Khan, CEO of Compute Maritime. “This is what the future of shipbuilding looks like to us: intelligence at the core of design, delivering efficiency the industry can measure on its fuel bills and its emissions reports.”

“Truly efficient ship design now depends on smart tools working on both sides of the problem: a revolutionary generative tool to create and explore high-performing hull forms, and advanced simulation to evaluate them with confidence,” said Dmitry Ponkratov, Marine Director at Siemens Digital Industries Software. “GenDSOM brings these elements into a single loop, so that design, performance and manufacturability evolve together rather than in sequence. Results like these show what becomes possible when generative AI and high-fidelity simulation are combined.”

“The largest fuel savings in shipping come from optimizing a vessel’s form, and that is exactly what these results demonstrate,” said Tahsin Tezdogan, Professor at the University of Southampton. “Advanced simulation allowed us to validate the performance of unconventional, highly efficient hull shapes with confidence.”

The vessel has been designed as a future-proof, upgradeable platform, says Compute Maritime. By specifying an offshore fast-charging inlet at the build stage, it is ready to draw power from the charging infrastructure that wind farm operators are beginning to install at sea. As that infrastructure and the wider electricity grid continue to decarbonize, the vessel can progressively shift from diesel to battery power with no change to its core propulsion system, on a roadmap that points toward emissions reductions of around 95% over its 25-year service life.
‘
“Seeing a NeuralShipper-optimized hull translated into a complete, buildable crew transfer vessel has been one of the most rewarding parts of this project,” said Jami Buckley, CEO of BYD Naval Architects. “GenDSOM let us explore hull forms and configurations that are difficult to reach through traditional design alone, and to do so within real manufacturing and operational constraints. The result is a vessel that is efficient, practical and ready for the offshore wind market.”

GenDSOM responds directly to the U.K’.s 2025 Maritime Decarbonization Strategy, which targets emissions reductions of 30% by 2030 and 80% by 2040 for domestic maritime. Decarbonising smaller working vessels such as crew transfer vessels, which operate intensively and in growing numbers across the U.K/’s expanding offshore wind fleet, is widely seen as critical to meeting those targets.

GenDSOM is funded through the U.K .Shipping Office for Reducing Emissions (UK SHORE) programm]in the Department for Transport. Innovate UK, part of UK Research and Innovation, is the main delivery partner for UK SHORE interventions.