Advancing computational analysis through AI
A core challenge in hull form design is not only identifying better-performing geometries, but doing so in a way which preserves engineering understanding. Johnathan’s research recognises designers must be able to interrogate results, understand trade-offs and apply insight across programmes - not simply receive an optimised shape. A key focus of the work is therefore on combining machine learning with physics-based simulation in a way which remains transparent, explainable and grounded in engineering judgement.
Rather than treating AI as a replacement for CFD, the research explores how it can act as a force multiplier - guiding where and how computational effort is applied. AI models are used to identify patterns across large volumes of data, highlight dominant performance drivers, and prioritise regions of the design space which warrant deeper investigation. This enables engineers to explore more concepts, earlier in the design process, without the prohibitive time and cost traditionally associated with high-fidelity analysis.
Integrating AI into real-world design workflows
A significant emphasis of the research is on practical implementation. Hull form design workflows are well established, and meaningful innovation must integrate seamlessly with existing processes. Johnathan’s work focuses on embedding AI techniques into familiar design stages - supporting concept development, refinement and trade-off analysis - rather than introducing parallel or disconnected tool chains.
By accelerating early-stage exploration, the approach helps reduce downstream design churn and late-stage rework. Designers are better equipped to make informed decisions when performance sensitivities, constraints and trade-offs are understood earlier. This leads to more resilient designs and greater confidence when transitioning from concept to detailed design and physical validation.
"Advanced analysis only delivers value when it informs real decisions - AI gives us the opportunity to explore design space faster, and with greater confidence."
Supporting digital twins and adaptive design
As digital twins, continuous monitoring and adaptive design approaches become more prevalent, the ability to rapidly assess and update hull performance will be increasingly valuable. Johnathan’s research contributes toward future-ready design environments where AI-enabled models can support near-real-time assessment of design changes, operational conditions or evolving requirements.
This capability is particularly relevant in contexts where operational profiles may change over time, or where regulatory, environmental or commercial pressures drive new performance priorities. By enabling faster and broader exploration of alternatives, the research supports more agile decision-making across the asset lifecycle - not just at initial design.
Driving sustainability through better design insight
Efficiency is a central consideration in modern maritime design. By enabling detailed exploration of hull forms optimised for reduced resistance, improved propulsion performance and lower fuel consumption, the research advances practical design objectives. Importantly, this is achieved without narrowing focus to a single metric - allowing designers to balance efficiency gains with stability, safety, operability and build constraints.
The work provides tools and methods which support evidence-based reductions in emissions and lifecycle impact, while maintaining robust performance and compliance with industry standards.
Building capability through collaboration
The research is inherently multidisciplinary, drawing on close collaboration between hydrodynamics specialists, data scientists, naval architects and software engineers across BMT. This ensures AI techniques are developed with a deep understanding of the physical problem space and real-world engineering constraints.
For Johnathan, the Fellowship provides the space to pursue this forward-looking research while embedding its outcomes into our broader analytical capability. It enables knowledge transfer, mentoring of emerging analysts, and the development of repeatable, scalable approaches which can be applied across future projects. The work is about ensuring advanced computational methods continue to translate into practical, trusted engineering outcomes - delivering lasting value for customers and the wider maritime sector.
