How Zapata and Andretti Motorsport will use quantum computing to gain an advantage at the Indianapolis 500

You might think car racing wouldn’t be a good application for quantum computing, since teams are made up of greasy monkeys who might know car mechanics but wouldn’t know how to take advantage of advanced computing. But you would be wrong.

Motor racing is a big business where there can be a very thin line between success and failure. To give you an idea of ​​how small things can make a big difference, you can check out the results of the 2015 Indianapolis 500. In this race, the time difference between first Juan Pablo Montoya and second Will Power was of 104.6. milliseconds. And those 104.6 milliseconds were the difference between winning a top prize of $2.44 million or not.

It turns out that a car race generates a lot of data, around 1 terabyte per car in a typical race, which, if analyzed and used wisely, can help give a race team a critical advantage. To that end, Zapata Computing and Andretti Motorsports formed a partnership earlier this year to work together on race analytics and see how they could use Zapata’s advanced analytics, quantum techniques, and workflow and classical/quantum hybrid data from Orquestra to win more races.

While this work between the two companies is just beginning, a big event for both companies will take place this weekend with the 2022 Indianapolis 500 race. We spoke with Chris Savoie, CEO of Zapata Computing, and he outlined three early use cases where they believe advanced analytics, machine learning, and quantum computing can potentially make a difference.

Tire degradation analysis

When you have a car that goes over 200 MPH, the tires wear out very quickly. In a typical Indianapolis 500 race, tires can be changed 5 or more times and require time-wasting pit stops. Also, tires have different characteristics when they are just fitted and when they have been used for a while. So the Race Director has a lot of varying strategic juggling. When should the car be called for a pit stop to change tyres, what set of tires should it put on the car, and how many tire changes should it have, and what are the weather and track conditions current? For a data analyst, this is an important optimization problem and will be one of the first areas Zapata will work on with Andretti to create an ML model that can help guide these decisions using data collected during previous race sessions as well as data collected in real time during the race.

Fuel Savings Opportunities

The cars must be refueled during the race. In addition, the driver has some control over fuel consumption by the way he drives. If a race team can find a way to minimize the number of pit stops and avoid a pit stop, it can save a lot of time. Also, you don’t want to cross the finish line with a full tank because that would be a waste. In the 2016 race, driver Alexander Rossi took a gamble and decided against making a final pit stop to refuel with 33 laps to go. Turns out he ran out of gas at the very end and crossed the finish line. But he won the race because the second place guy decided to fuel up and the extra downtime cost him the race. So, finding ways to improve fuel efficiency and determine the best time to refuel also turns out to be an optimization problem that can be an opportunity to use machine learning and advanced analytics to find the best solution and improve performance in the race.

Yellow Flag Predictive Modeling

A yellow flag during the race occurs when an accident occurs or there is debris on the track. Drivers are required to reduce their speed and overtaking another car is prohibited. One of the impacts of this is that the relative lead of one car over another is reduced. But it can also be a good time to make a pit stop since the cars are not going full speed while the flag is on. If a race team had a crystal ball and could predict when a yellow flag would occur, it could help them determine their best pit stop strategy. It might sound a bit far-fetched, but the Zapata/Andretti team will try to create a model for this that will be based on the conditions on the track, the status of the different cars in the cars, what particular drivers are in those cars , and other factors collected during the race. It will be interesting for us to see if they can actually build a useful model for when yellow flags may occur from this data.

From an operations perspective, working in this environment can present unique challenges. But it also provides learning opportunities for the Zapata team as they face real-world challenges and find ways to solve them that can be used for future product improvements and customer engagements in other places. other areas. One of the first things to understand is that the racing environment requires real-time decisions and you don’t want to be using a quantum computer somewhere in the cloud on race day. The latencies will be too slow and you don’t want to struggle with spotty Wi-Fi connections. So, Zapata and Andretti set up an onsite Race Analytics command center as shown in the image below.

Zapata and Andretti won’t be installing a quantum computer in this trailer, but it will have great classical computing capability to help the team make real-time decisions on race day. Machine learning applications are usually divided into a training session that develops the optimal coefficients for a model and a runtime part that simply runs the model and provides an output based on the previously configured coefficients. The training part is the most computationally intensive part of an ML model, they don’t have to run in real time and are a good opportunity to take advantage of quantum computing. Running a model once it’s created is less computationally intensive and can be done on a regular CPU. The team can feed data from previous races and trials, create an ML model over several days or weeks, but then run the ML model in real time on conventional computers installed in this trailer.

The collaboration between Zapata and Andretti goes well beyond the exploitation of quantum computing. The overall program will involve working with multiple databases that could reside with cloud providers, cutting-edge computing data from various sensors, and managing workflows of both classical and quantum nature. Zapata will use its Orquestra product to help manage all of this.

It will be a long-term collaboration. Because available quantum computers are not yet powerful enough to provide an advantage, early implementations of this work will use quantum-inspired algorithms. However, the intention is that as quantum processors become more powerful, these algorithms will eventually be moved to full quantum computers and allow companies to build larger, more complex, and more accurate models to boost their advantage. Andretti participates in many types of motor racing and has many different teams. So both companies will have plenty of opportunities to try and develop this capability. We also expect enterprises to find additional use cases for leveraging advanced computing capabilities when working together.

For more information on this collaboration, a press release posted on Zapata’s website can be viewed here.

May 26, 2022

Sherry J. Basler