McLaren Racing pursues car build efficiency with Alteryx

0
303
Oracle enhances customer experience platform with a B2B refresh

Source is ComputerWeekly.com

Ed Green, head of commercial technology at McLaren Racing, says the team’s technology partnerships go way beyond logos on the cars and on the drivers’ kit. “Our car won’t go on the track on a Sunday afternoon if our IT systems aren’t running,” he says. “So, those relationships go pretty deep and are meaningful.”

Green was speaking of McLaren’s most recent partnership, with data analytics software supplier Alteryx, an engagement that includes fan outreach as well as making the racing cars more performant. That relationship will become publicly more visible this coming weekend with the Belgian Grand Prix on Sunday 29 August.

“And you’re taking some of that IT to the extreme,” he says. “You’re putting it in 23 different locations a year, and you’re deploying in a hostile garage. In my first week at McLaren, I found an IT colleague using a vacuum to clean out the back of the garage. I’ve never seen in my entire IT career seen someone hoovering a server.

“When you are pushing hardware and software right to the very edge to be mission-critical – including driver safety, and analysing information about critical systems on the car from multiple datasets – you’ve got to make sure that it’s performant. And so, yes, you want to make sure your suppliers are deep in the team.”

That can extend to calling on technology partners to assist mid-weekend when the race is on. Dell supplies the team’s high-performance compute, and DarkTrace is its cyber security partner.

“We want to find partners that are passionate,” says Green. “It’s really important to find people who are just as excited and passionate about making the cars go quicker on a Sunday afternoon as the rest of the team is.”

He is complimentary about the specific relationship with the Alteryx team, which began in March 2021. “It was very hands-on, working with our datasets quickly,” he says. “There was less conceptualisation of the workflow or of bringing multiple datasets together. We didn’t sit around doing endless workshops. We started with one or two small ideas about making the car build more efficient, and how we could do something to better engage with, or better understand, the [McLaren] fans.”

The McLaren Racing IT team of “around a dozen” people and its digital transformation and marketing teams are among the first users of Alteryx. But they are looking forward to seeding the technology’s use with McLaren’s aerodynamicists and other engineers, says Green.

The cost cap introduced into Formula 1 this year, and announced in October 2020, was the trigger to find something like Alteryx to streamline the manufacturing costs of the car, said Green. The cost cap for 2021 for the 10 F1 teams is $145m.

Green refers to the car build as a major area where costs can be controlled better. There are many disparate datasets in play there, and while they do have the existing ability to see where the costs lie, there is a pressing need to do so “in the moment”, around the races themselves.

“There are 80,000 components on the car, and they change at a rapid rate,” he says. “In any given season, about 90% of the car will change. And there is small carryover into next season’s car, even less next year due to regulation changes.

“The car is always best described as a prototype. Depending on the characteristics of the track, that will change features of the car. We are roughly producing a new part every 17 minutes throughout the season.” And that is at the track, using 3D printing, as well as at the factory, he says.

Green says Alteryx is helping to bring together over a dozen different datasets to automate the analysis and processing of information to help design and engineering teams avoid excess material and manufacturing wastage, to be more efficient in terms of what is manufactured and delivered to track.

McLaren’s use of Alteryx will also help what the high-performance car industry calls “aero model correlation”. The aerodynamic modelling of a car will traverse a multitude of analysis tools, each producing data points – from purely virtual computer simulation, wind tunnelling, and then “real, real life” of the races themselves, says Green.

And it is the data associated with each part that is critical. Alteryx will provide correlation and analysis of each part’s “journey” from design through testing and deployment in the races. The essential idea is to align better and better how the part performs, from virtual simulation through wind tunnelling to the car as it is driven under race conditions.

Green adds: “We’ll be looking to see ‘did that part perform as we thought it would, based on our modelling and simulation?’ So, you have all those different datasets and data coming from the car. And, ultimately, we want to bring that all together into one view.

“What we are trying to see is: ‘did that part perform from the virtual world to the physical world when that part went into the wind tunnel as scaled down model?’ And if we think those two sets of information do correlate, then there’s a strong likelihood that we will say yes, we think that’s a good performance advantage for the part. We would then put that into manufacturing. Then we get to the car. You then want to see if the part you produced worked at scale with all the other aerodynamic properties around the car.”

There are about 300 sensors on the car, which, along with the on-board video and radio feeds in its “brain”, its electronic control unit (ECU), produce one and a half terabytes of data over a typical race weekend.

“Our challenge as a team,” says Green, “is to get all that data back and combine it to spot anomalies, and so on. A lot of strategy work takes place in simulating the races. We do about 300 million simulations of the race before we see the green light on Sunday afternoons.”

For the drivers, such as Lando Norris and Daniel Ricciardo, that means taking in information such as: if it starts raining on the first lap, this is what we recommend you do. “Those decisions have to make in as little as three seconds,” says Green.

For the team members working with the race data, Alteryx’s value is that “it gives people time to focus on the output and the results, and working and collaborating with other people rather than manually munging the data”, he says.

An additional, and differentiating, benefit of Alteryx, says Green, is the “unique” capability that the supplier delivers with geo-data. This means they can connect McLaren fans with their commercial partners. “So, if one of our lifestyle partners has a location in a city and we can see that a lot of our fans are based in that area, we might go and do a fan engagement at their headquarters, regional offices, or one of their stores, or equivalent,” he says.

By this Green means Alteryx datasets, using geo-data to help McLaren understand what race venues are closest to their fans and how to build experiences close to them. Alteryx Data Sets are available to their users through their install platform. An Alteryx spokesperson says the datasets are either “location intelligence or consumer intelligence packages and consist of  data from providers such as mapbox, Tom Tom, Experian and Digital Globe”.

Source is ComputerWeekly.com

Vorig artikelPushing the Limits Paid Off for Didi, Until China Cracked Down
Volgend artikelVery Personal Computing: In Artist’s New Work, A.I. Meets Fatherhood