Modelling Fire Service Response Times in West Midlands, UK

This study examines the prediction of median fire response time of grid cells in West Midlands County, using supervised machine learning, including Lasso regression and tree-based models (Random Forest, XGBoost). Results show that random forest achieved the highest accuracy (R² = 0.80), with fire station distance and neighbouring fire frequency being the most important factors. This research has the potential to inform fire response strategies and fire station siting, in order to improve emergency preparedness.