McLaren Accelerates Formula 1 Performance – On and Off the Track

In Racing, “Data is King”

Wind. Rain. Temperatures. Fluid dynamics. Tyre degradation. It all matters in a sport where one tenth of a second can be the difference between a good or bad race.

For McLaren Formula 1 Team, it’s not enough to know the current state of all these factors—they need to predict the future.

“This is a race where data is the king,” said Andrew McHutchon, Senior Data Scientist, McLaren Racing. “We collect thousands of different readings from each race track and then we turn this into something that we can understand. Anything we can do to move the car forward, to advance what we are putting into it, then we can increase the chances of victory on track and tasting that sweet champagne.”

Driving Minute-to-Minute Decisions

At each race location, data pours in from track sensors along with McLaren’s own weather stations. The team back at the McLaren Technology Centre in Woking, UK must interpret and relay insights to those on the ground to help make decisions, particularly regarding tyres.

“As weather conditions change throughout the day, we have to change the car to compensate,” said Nick Snyder, Performance Director at McLaren Racing. “We’ll change tyre pressures, suspension, and downforce levels. The key is to keep tyres in the right operating window to maximize performance.” 

Yet, there’s more data than the team can possibly process on their own. For help, McLaren turned to DataRobot AI Platform for machine learning that accelerates the time to understand all the data points, more accurately predict conditions, and make decisions for faster, safer racing.

“Even if we filled all the seats in a stadium, we simply wouldn’t have enough people to look at all the data points coming in. We’re looking to AI to act as extra engineers on our team to aid our analysis,” McHutchon said. “And to do that, we found DataRobot AI Platform brought a lot more power and depth than any of the other AutoML platforms that we’d seen before.”

Forecasting for Optimal Tyre Performance

To make the most accurate predictions possible, McLaren relies on the DataRobot Automated Time Series to forecast air and track temperatures-both big influencers on tyre degradation. Then, they deploy models via a StreamLit application.

Predicting the temperature on the track just a few minutes into the future allows the team to change their tyre strategy to prevent harmful degradation. The team finds that, with the time series capabilities and quick predictions, the actual values and predicted values are nearly identical.

The two models looking at track temperature and air temperature play a pretty critical role on a race weekend. DataRobot AI Platform allows our data scientists and team members to produce those models and roll them into production really quickly so we can see the benefit out on track.

“By leveraging DataRobot for predicting weather, we’re able to more accurately predict changes in track temperature—which affects tyre strategy—helping us complete two-plus additional laps before a pit stop,” said Joe Ray, Strategy Analyst.

Beyond track temperatures, the team turns to AI to help decide the level of cooling to run on the car to maximize performance. DataRobot’s Customer Facing Data Scientists bring knowledge and experience to guide the team in exploring their challenging use cases.

Bringing Horsepower to the McLaren Technology Centre

AutoML equips McLaren’s data scientists to analyze more data points and do so considerably faster. They expedite model training and experimenting time by approximately five times, savings days in the process. With that additional time, they can investigate more use cases.

At the same time, the platform’s ease of use allows users with no data science experience to run models independently. That frees data scientists to work on what’s most important to the racing team.

DataRobot AI Platform is the glue that brings us all together to lift the whole company’s ML and AI capabilities

Historically, McLaren’s dedication to technology has helped the organization win 183 races, 12 Drivers’ Championships, and eight Constructors’ Championships in Formula 1. Now with the AI platform and DataRobot team, the entire racing organization looks forward to pushing the limits of what’s possible—on and off the track.

Finally, Green points out that “We rely on our partners and our technology stacks to be platforms and help us deliver, and DataRobot was a real natural way for us to make sure we could operationalize models, but more importantly, roll them into a race weekend. And having expertise from the DataRobot team behind us was a real advantage.”