Mastering Race Strategy Optimisation
With a long-standing love of motorsport,
Hughes Raisin decided to study on the National Motorsport Academy’s MSC Advanced Motorsport Engineering degree to help him convert his existing knowledge and skills and make them relevant to motorsport.
With a background in software and diagnostics, Hughes was specifically interested in the data acquisition and analysis side of motorsport. The degree culminates in a final project where students can focus on an area of particular interest and Hughes was even given the opportunity to apply his project to real-world racing by European Le Mans Series LMP3 team RLR Msport.
The aim of Hughes’s final project was to develop software that would allow race engineers to create the best strategy for an upcoming race. The strategy would need to be instantly adaptable due to factors occurring in-race, such as changeable weather, a full-course yellow or a safety car. Data was mined from various sources to allow Hughes to then simulate an endurance race based on an equation.
At the end of each lap, the state of the car (driver, fuel, tyres) was updated. Pitstops were decided by a decision-tree based algorithm that took into account tyre wear, fuel consumption and minimum driver time.
“I believe that, to win races, you need three things: good drivers, a good team and a good strategy,” Hughes explains. “While working on this tool, I started implementing the main parameters everyone knows: fuel, tyres… But, I realised there was a missing one, and another one, and I missed a link between some of them. So I continue developing my tool, going further into what makes a race strategy. This is also what’s fascinating to me: the number of different parameters and the fact the race engineer has to get everything clear in his/her mind to make a split-second decision that could be a winning move. And I think this tool could be a way to support the race engineer achieving this.”
Hughes believes one of the advantages of studying the NMA’S flexible course is that students have the ability to focus on the topics that are of most interest to them and, given the wide-ranging nature of the modules, it means there is a “huge” choice for the final project. Hughes has used his software to create an optimised strategy for a race already run – the 2018 opener at Paul Ricard. When comparing his strategy to the actual race (see chart above, where the red lines show the strategy predicted by Hughes’s tool, compared with cars in the actual race), he found that the software provided highly accurate results. The software was then used by Hughes, working with RLR Msport at both Monza and Barcelona in support of the team’s existing race optimisation strategy.
The scope for the software has engineers excited about the prospect of being able to adapt a strategy mid-race. Motorsport is full of unforeseen events and this software would allow engineers to quickly modify tactics if there is a racing incident or change in track conditions.
“Hughes has done a great job with this project – an easy-to-use tool which is of use to a strategy/ race engineer for pre-race planning, race decision making and post-race analysis,” says Hughes’s tutor Tim Mullis. “From the beginning, this was to be a live project, and having Hughes involved in the ELMS races has provided him with the environment to understand first hand the requirements of the race engineer. This is much more than just a ‘university project’, it is a useful tool which Hughes intends to develop further even though his NMA studies have finished.”
The feedback on the software has been very encouraging. And, although there’s still some work to do, Hughes is hopeful that it can be used in a real-world scenario as a default method of race strategy optimisation.