Our statistical models accurately predict expected lap times depending on the state of critical explanatory variables, such as fuel load, tyre compound, tyre age, relative car performance and others. After the model is trained to estimate lap times for any combination of fuel load, tyre compound, tyre age and other parameters, it can be used in strategy optimisation, race simulations, driver performance analysis and other tasks.

We develop computationally efficient optimisation algorithms to minimise race time and formulate optimal strategies. The outputs include many parameters and valuable data for a strategy engineer: expected race times, number of pit stops, tyre compounds and stint lengths, expected tyre wear per lap for every wheel and compound, fuel load, optimal refuelling schedule and many others.

Our race simulation software helps engineers test optimal strategies for a race against random events and rational opposition. Mathematical models we develop for race simulations consider sporting and technical regulations, expected tyre wear, fuel effects, relative car performance, value of track position, probability of safety car deployments, pit stop time loss, lapped traffic and other parameters.

With in-depth knowledge and experience in motorsport, we create proprietary driver performance rankings based on big data analytics. Some examples of our work in this field include driver net performance index, junior driver rankings, overtaking abilities, etc. This data helps professional teams make informed decisions about driver development and scouting.