University of Strathclyde and Stone Marine Propulsion Limited KTP24_25 R1

Ship propellers are currently designed through a slow, manual process that leaves significant performance gains on the table. This project will replace that manual workflow with an automated system that uses neural networks, computational fluid dynamics, and parametric modelling. The goal is to customise and optimise each propeller design to suit a specific client’s requirements—rather than relying on off-the-shelf or slightly modified standard designs. If successful, the approach could make large ships more fuel-efficient and quieter. That matters because the global shipping fleet is a major source of carbon emissions and underwater noise, which harms marine life. Better propellers can reduce both without requiring new engines or fuels. The impact would be felt in supply chains and logistics: lower fuel costs for operators, and a quieter ocean for ecosystems. This is applied engineering, not fundamental science. The research is directly aimed at a commercial product, and the partners—a university and a marine propulsion company—intend to put the resulting design tool into use. There is no immediate consumer-facing outcome; the benefit will show up in the performance of ships that carry the world’s goods.