Graeme Hackland, CIO of Lotus F1, has been working for the race car team for 16 years and has seen major changes in the way IT is used both on and off the track.
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
He started as network engineer and moved up to an IT infrastructure manager role when the team was taken over by Renault.
The team’s head office is in Oxfordshire where Hackland now spends much of his time.
He says: “Early on, I used to support the guys at the track. But these days there are restrictions.”
Hackland is responsible for the provision of application and infrastructure to support the IT needed for the race team including design, manufacturing, the wind tunnel systems and the supercomputer which runs complex fluid dynamics (CFD) simulations for optimising the cars' aerodynamics. He is also responsible for security and is a member of the senior management team.
He recalls: “When I joined the team, we mainly ran big Unix system for CAD [computer aided design], and we’d take the Unix workstations to the races. We had five laptops on track and green terminals in the factory.”
Leading the adoption of technology
But F1 is highly technical. So while this setup looks primitive by today’s standards, in his experience of the sport: “F1 is always leading the adoption of IT technology.”
He admits he is not a big fan of big data, but success at F1 is all about high speed data analysis.
Data from the car used to fit on a single floppy disk. F1 cars now produce gigabytes of data.
“I don’t like the term big data because it causes confusion. For us the big data challenge is in using multiple data sets, such as data from aerodynamics and CAD, to create performance from the car,” he says.
Such analysis is not restricted to the race track.
“I see a lot of similarities in real time data analysis in financial services. When you are on the track you will have, at most, 90 seconds to make a decision, based on analysing hundreds of parameters,” Hackland says. These cover everything from tyre pressure, the state of the weather, to the position of rival teams, and their tyre strategies.
The team is using Microsoft Dynamics to replace siloed systems to deliver an end-to-end view of the manufacturing process. He uses IT consultancy Avanade for software development and testing. Software is built using a service oriented architecture (SOA) with Microsoft.net.
Lotus F1 does use some off-the-shelf software, like Dynamics, but where there is a competitive advantage, it uses bespoke systems.
“If we see there’s a competitive advantage such as in CFD, the wind tunnel and statistical analysis, we write the software ourselves,” says Hackland.
It is focusing part of its development efforts in a system for improving data analysis.
Avanade is developing the system, called Strategy, which combines engineering and data with Microsoft Dynamics. The Strategy system is a statistical engine that takes race information, engineering and simulation data, to enable the team to make continual improvements to the car and the race strategy.
The system will enable Lotus F1 to get an end-to-end view covering PLM (product lifecycle management), ERP (enterprise resource planning) and the race performance application.
Hackland plans to link CAD and Microsoft Dynamics. This would help Lotus F1 streamline analysis of the car’s performance with engineering data.
“Today it is very manual and it take a few days to bring all the data altogether,” he explains. Linking Dynamics with CAD would speed up data analysis.
Hackland says a race team can win a single race without such data analysis, but winning the F1 championship requires software like Strategy to help the team build a better car over the season.
But the Strategy software will become more important in the 2014 season.
Hackland says: “Strategy software will be critical next season due to changes in regulations, such as the use of 1.6 turbo engines.
“It won’t be so easy to look at the previous season’s data. We’re almost baselining at zero again."
He adds: “We look at hundreds of parameters, including what competitors are doing. A subset of data is over the radio from the car in real time.
"We monitor alarms like a puncture. The strategy system uses this data to run a race simulation every three minutes.”
Given that a F1 car does not carry enough fuel to run at full throttle to the end of the race, Hackland says: “We need to analyse how we will get to the end of race.”
Strategy leads to victory
In Australia, he says the race strategy helped Kimi Räikkönen win the race. “We have been very good at managing tyres. Kimi did two pitstops, rather than three,” says Hackland.
Much of the analysis is run real time at the trackside. However Hackland says Lotus F1 is now building a dedicated room at its head office where people can work with real time data from the factory.
This would have been impossible a few years ago given the state of UK broadband.
He says: “We are in the middle of the English countryside and there is no industry around. In the past, getting communications was difficult.
"But we now a dedicated 10 Mbps link with high availability.”
The reliability of the connectivity means Hackland is giving serious consideration to using cloud services.
“We are started with putting the obvious things in the cloud like email and CRM [customer relationship management] – but we can see huge potential benefits in terms of racing team logistics.
After each race all the IT equipment is packed up and shipped back to Oxfordshire, before being sent to the next race. It is not possible to update software while the equipment is in transit, which means Hackland only has a day to configure the systems before the race preparation starts.
He says: “We cannot do application updates until the Wednesday before the race. It would be much better if we could pull this back to earlier in the week.”
Cloud computing could facilitate this.
That said, he admits that everything trackside needs to run standalone, just in case the global communication link fails.