This repository contains the code and data analysis behind my Computer Science thesis: Optimization of Race Strategy in Formula 1 using Machine Learning. The goal of this project was to explore how historical race data can be used to evaluate the effectiveness of race strategies — focusing on pit-stop timing, tire compounds, and lap performance.
This project was accepted for presentation at the International Conference on Informatics and Information Technologies (CIIT 2025).
Analyze historical F1 data to understand which strategic decisions correlate with better race outcomes. Focus areas included:
- Tire compound usage
- Pit stop strategies
- Lap time patterns
- Position changes
Lap-by-lap data from the 2024 Formula 1 season up to and including the Dutch Grand Prix. The datasets include weather, track conditions, pit stop logs, and driver positions.
- Random Forest Regressor
- Gradient Boosting Regressor
- Mean Squared Error (MSE)
- R² Score
- Language: Python
- Libraries: Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn