Industrial Generative AI for
next-generation race analytics, powered by Orquestra

⋅⋅⋅⋅⋅⋅⋅⋅     ANDRETTI GLOBAL x ZAPATA AI SEASON 3

Our second season with Andretti Global is behind us. Take a look at how Zapata AI is helping Andretti Global with race time analytics and strategy – all powered by Industrial Generative AI.

The Challenge

Races generate a lot of data — about 1 terabyte per car. Andretti Global is always looking for better ways to analyze that data to gain an edge over the competition. Their goal? Upgrade their existing data analytics infrastructure with proprietary Generative AI techniques to drive their race strategy and win more races.

Our Approach

In 2022, Zapata AI began upgrading Andretti’s analytics infrastructure and piloting advanced Industrial Generative AI applications and techniques. In the future, as Generative AI technology improves and advanced hardware (such as quantum computers) matures, the team can test the performance of new devices and algorithms on Orquestra, and easily swap in whichever backend delivers the best performance.

The Race Analytics Command Center

In early 2022, we deployed the Orquestra^® platform within the Andretti Global | Zapata AI Race Analytics Command Center (RACC). The RACC is a mobile engineering environment where engineers from Zapata AI and Andretti work side-by-side in pursuit of a real-time advantage on race day. The hybrid infrastructure combines data lake integration, cloud and dashboards to drive decision-making — all managed with Orquestra. Engineers working from the RACC are testing various use cases in machine learning and optimization.

Not all the action happens on the track—there’s an entire engineering world inside this truck.

⋅⋅⋅⋅⋅⋅⋅⋅     EXPLORING USE CASES IN MACHINE LEARNING AND OPTIMIZATION

Tire Degradation Analysis

Tires wear out quickly when they are going over 200MPH, requiring time-consuming pit stops to change them. Zapata AI is working with Andretti to create a machine learning model to guide strategic decisions around tire changes, such as when a car should swap out tires, which tires should be used, and how often they should be changed based on current conditions. This use case translates to predictive maintenance problems across industries.

Fuel Savings Optimization

The fewer times a car has to refuel, the more time it can save in the race. Zapata AI and Andretti are applying machine learning and advanced analytics to help their drivers optimize fuel consumption as well as determine the best time for refueling. Similar fuel savings solutions have wide applicability in any industry looking to shrink its carbon footprint or time to delivery.

Predictive Modeling: Yellow Flag

When there’s an accident or debris on the track, drivers are required to reduce their speed and are prohibited from passing other cars. This is a yellow flag, and since cars aren’t going full speed, it’s often a good time for a pit stop. Zapata AI and Andretti are creating a model to predict when a yellow flag is likely to occur based on track conditions, the status of various cars, the drivers in those cars, and other factors. This ability to predict and preemptively respond to disruptive events has wide applicability beyond racing.