How Envision Virgin Racing team uses data science to hone performance

Enterprise

All the sessions from Transform 2021 are available on-demand now. Watch now.


Effective use of data science can help business leaders improve their decision-making processes. In the high-speed world of motorsport, those decisions have race-changing implications.

That’s certainly the case for Sylvain Filippi, managing director and CTO of the Envision Virgin Racing team, one of the leading teams in Formula E — a single-seater motorsport championship that only uses electric cars. His team produces huge amounts of data, but needs to use this information effectively to produce a competitive advantage.

To give his team every chance of success, Filippi’s team started working with global professional services firm Genpact two years ago. Envision Virgin Racing uses Genpact’s data science skills to hone performance on race day. Filippi explained to VentureBeat how the relationship works and the advantages it provides to his team.

VentureBeat: Why is data crucial on race days?

Sylvain Filippi: That’s everything to do with the race format in Formula E. Formula 1 races are spread out over three or four days; they do free practice on Thursday and Friday, qualifying on Saturday, and a race on Sunday. They have loads of time to look at the data and analyze it. In Formula E, we do all of these elements in one day: at 7:30 a.m., we start free practice one, and then free practice two at around 10 a.m., qualifying at midday, and at 4 p.m., we race. You basically have an hour or less in-between sessions to download all the data, to look at it, gather all the insights, and then make decisions on what you’re changing and what you’re modifying for the next session. So it’s a huge engineering challenge, purely because of the race format.

VentureBeat: What does this data challenge mean in terms of your business?

Filippi: You have to be super good at IT – we have a top IT infrastructure team, given the size of our business. You also need talented software engineers and data engineers because you need to be really efficient at downloading data, analyzing it, and structuring it. And then, the engineers need to know what they are looking for in terms of getting the right insights and making the right changes. Those changes are twofold – making any changes on the car like in any other racing formula, but also all the data that is related to the energy side of Formula E. There’s a gigantic amount of work to cover in less than an hour between sessions.

VentureBeat: How are you working with Genpact to help you deal with this data challenge?

Filippi: Genpact is helping us by using its expert capabilities in data science. They’re starting to play with our data, which according to them is pretty good because there’s a lot of detail and it’s really well structured, which is not the case in many companies. I’m pretty certain that they’re going to be able to find trends and correlations and links between all sorts of random stuff — from weather to temperature — and onto the tires. They’ll be able to crunch our entire set of data and find some trends that we haven’t seen and we don’t know about because it’s physically impossible for the human brain.

VentureBeat: Why is the application of data science so important to your team?

Filippi: This work is exciting because it’s never been done in motorsport before really. It’s very, very early days in terms of how to gain performance through AI and machine learning – and that’s a fun exercise. We started working with Genpact about two years ago. And we started from scratch. There was no software or a platform; nothing like that existed. So, they located some super clever data scientists and started looking at our data, and they’re starting to come up with insights and models.

VentureBeat: Where has the relationship with Genpact produced dividends?

Filippi: You have the pure performance side of the car, which is changing race by race, which is about asking “how do you make the car go faster?” There’s also the whole strategy side. So we don’t have pit stops at the moment in Formula E, even though they could come back in the future. But, we do have Attack Mode, which allows teams to temporarily raise the power output of a Formula E car. Using that boost at the right time can have race-changing implications, so when do you use Attack Mode? There’s many decisions that need to be taken in a 45-minute race around Attack Mode and what we do in any sporting situation. Data is crucial because we’re a relatively limited team by motorsport regulations, and you’re only allowed 17 operational people at the track – and that’s not many people compared to Formula 1.

VentureBeat: Where else are you using data science on race day?

Filippi: In Formula E races, instead of a number of labs, like in Formula One, there are timed races – so 45 minutes plus one lap. Managing that sounds easy, but it’s a gigantic challenge because when you start the race, you have a certain amount of energy in your battery. If you don’t know the distance you’re going to be covering, then you don’t know how much energy you can use per lap. If you misjudge it, you’ll run out of energy at the end of the race. Or on the other side, you’ll finish with too much energy, which means you could have gone faster. Genpact have worked with us to develop a piece of software that helps us evaluate that energy relationship dynamically throughout the race so we can accurately estimate the exact distance that we’re going to cover. And that’s hugely complex because, by definition, what happens in racing is an ever-changing scenario. No one can predict a race –  there’s always the impact of a safety car or rain or something else.

VentureBeat: How are you using analytics to engage the fan base?

Filippi: Fans these days, especially the younger generation, are really data-hungry – they want to understand how the sport works and why decisions are made. The strategy aspect of it, and why you make a decision, is super important. So we’re working with Genpact on how we can give even more to the fans because it makes the sport more engaging and it makes them excited. But that’s a work in progress. It’s really important – we’re a new sport. This is our seventh season, and we’ve grown at a rapid pace. Now, we are a major motorsport platform and we need to keep going, but also we need to reaffirm who we are and our values.

VentureBeat: What is your best-practice tip for making the most of data?

Filippi: Genpact has taught us that it’s really important to be very organized – structured data is everything. If you have a lot of data, but it’s completely unstructured – and you can’t access it because it’s in documents where you can’t extract the data – then it’s pretty useless. So you’ve got to really think about the outcome and the insights you want to get to, and then work backwards to what your data should look like. Never underestimate the value of your data, which in most companies around the world is much more valuable than people would think.

VentureBeat

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Products You May Like

Articles You May Like

Infosys to Develop Quantum Computing Capabilities on Amazon Web Services, Will Use Amazon Braket
Microsoft unveils four Surface computers and a new folding phone
Why Far Cry 6 Player Character Dani Rojas Has a Face, a First for Far Cry
Apple’s New iCloud Private Relay Service Leaks Users’ Real IP Addresses
Facebook CTO Mike Schroepfer to Step Down, Will Become Company’s First Senior Fellow in 2022

Leave a Reply

Your email address will not be published. Required fields are marked *