Digital twins for talent identification in professional cycling
Elite professional cycling is an exceptionally complex sport operating at the forefront of technological innovation. The scale and richness of performance and race data generated at the highest level present a unique opportunity for modern data science and AI to support decision-making in an increasingly competitive global landscape.
This PhD project, funded by the INEOS Grenadiers, combines proprietary performance metrics with publicly available race data to develop digital twins — data-driven representations of riders that capture the key characteristics underpinning success at the elite level. By distilling large, noisy datasets into a focused set of performance indicators, the project aims to track and compare rider parameters in a systematic and interpretable way.
Blending advanced statistical modelling with expert domain knowledge, the research seeks to create a robust talent identification framework capable of highlighting riders with the potential to succeed in the world’s biggest races. As professional cycling continues to globalise and intensify, such tools will play an increasingly important role in identifying, developing, and supporting the next generation of elite performers.