The modern Formula 1 car is an intelligent and connected data system that can travel at over 200mph. More people have been to space than have driven one.
These F1 racers represent the pinnacle of vehicular technology. Every aspect of every car is monitored by hundreds of sensors, measuring lap times, tyre and brake temperatures, air flow and engine performance. As you watch a race, hundreds of stories are playing out behind the scenes, such as tyre wear, engine health and driver responsiveness.
There are few sports where data analytics is used as comprehensively as it is in F1 – it dictates how the cars are built, how they are driven, and how the race is broadcast.
Before data analytics permeated the sport, success or failure in a race was solely down to the driver and the split-second decisions he made on the track. By the 1970s, however, telemetry systems had become sufficiently small and sophisticated that they could be fitted to cars to provide an insight into how they were functioning.
By the 1980s, electronic systems were routinely fitted onboard F1 cars. Initially storage was limited to just a lap’s-worth of data and drivers would be given a signal to turn the telemetry on when the team wanted to capture information. This data would then be taken off the car and transferred to garage based computer systems for later analysis.
By the end of the decade ‘burst’ telemetry was developed, which fired radio signals from the car to the garage during a race, giving the crew advanced warning of the physical situation of the car for pit stops. These bursts gave way to streaming data, piped back to the garage and on to the factory. It’s since become a vital part of running and planning F1 races.
Today, real time data streams drive the sport in every aspect, including pre-race simulations, real-time decision making by analysts and pit crew, post-race analysis, and the broadcast experience.
Multiple sensors across car and driver are constantly monitoring and transmitting information. These streams of data give teams hidden insights that would be invisible to the human eye.
Speed, exhaust and tyre temperatures, clutch fluid pressure, oil and water levels, engine RPMs, G-force, and driver biometrics are just some of the elements measured by the car’s array of sensors. These are collected in the cars Electronic Control Unit and transmitted via a telemetry antenna to the pits in real time.
The Mercedes AMG* F1 W08 EQ Power+ cars, for example, are loaded with 200 sensors that transmit millions of data points over the course of a race weekend. Over 300GB of data is reportedly transmitted from the car. Williams* claims it has over a thousand channels of data being collected at any one time during a race. While Red Bull’s* RB12 car is fitted with around 100 sensors gathering data on 10,000 components.
Cloud computing has a role to play here too. Some of this crucial data is sent to the garage in bespoke feeds relevant to mechanics and engineers, while wider
Modern F1 cars are fitted with around 200 sensors
In fact, because a limited amount of people per team are allowed trackside, much of the pre- and post-race analysis work is performed by team members back at the team HQ. They parse the data in real time, combining it with GPS, weather information, and what the competition is doing, to give them an analytical overview of every race.
All of this data is subsequently pooled, analysed, and actioned to shape a race strategy. This is then fed to trackside analysts in communication with the driver.
Mercedes* works with data analytics and integration specialist Tibco Software*, creating a virtual team of data analysts to work with data streamed from races. Every detail can be poured over to analyse, for example, where a problem occurred – a car’s trajectory, tyre pressure, weight, the track conditions, and many other factors are gathered into a data reservoir for analysts to sift through and get a sense of what went wrong (or right).
Data is also used to optimise pitstop procedures. Pit stops have always been a vital part of winning or losing an F1 race – with crews drilled in military precision to perform any tyre or nose changes required.
A delay can, and often has, resulted in a driver losing a race. By sifting through video footage and data from the car and pit equipment after practice runs, teams can see where valuable seconds can be saved. In 2016, four pit crew workers at the Williams* team were kitted out with biometric sensors measuring heart rate variability, recovery times, breathing rate and estimated core temperature. By capturing this data, it allowed the team to assess the crew and adjust their fitness training, with the aim of improving performance.
The vast amount of data collected during a race can also be used to reengineer vehicles. After a race, the data is used in simulations to improve the car’s systems before the next competition cycle. McLaren* claims it can achieve more in one day than it would in a week of on track testing by doing this.
Data is heavily used in development and testing too, whether it’s through (CFD) computational fluid dynamics to simulate airflow and the interaction on surfaces, or driver ‘in-loop simulators which model realistic car behaviour based on previous gathered data.
The F1 data revolution is also a boon for broadcasters – viewers are able to see information on pit stop times, sector times, DRS charge, heat maps (showing which parts of the cars are exposed to the highest temperatures), listen into selective live radio feeds between the race engineer and the driver, and be privy to certain streams of live data transmitted to teams, giving them an insight into the decisions that are being made.
Here, more than perhaps anywhere, big data drives big decisions. As the reach and application of data analytics proliferates in the wider world, F1 is bound to take advantage of whatever the tech industry throws up – with cloud computing, predictive analytics, predictive intelligence, machine learning, and prescriptive intelligence playing larger roles in the future of the sport.
*Other names and brands may be claimed as the property of others.