The National Capital Territory of Delhi has experienced a substantial increase in fire incidents over the past decades, with a rising number of fatalities attributed to the low response times in fire emergencies. This study aims to analyze the spatial distribution of fire events in Delhi from 2006 to the present and to estimate fire risk scores based on factors such as population density, building density, building type, and land use and land cover (LULC). Two approaches were employed to determine the risk scores in Delhi: The Random Forest machine learning algorithm and the Analytic Hierarchy Process (AHP). Results from both methods indicated a similar trend in the percentage area calculated for each risk score. A deeper analysis of these risk scores revealed that, according to National Fire Protection Association (NFPA) codes and standards, fire stations should be within a 2 km proximity for urban areas with a population density of 1,000 individuals per square kilometre. However, wards classified as high risk do not meet this criterion, posing a significant threat to both human life and economic assets. Consequently, this study also proposes several mitigation measures to address these risks.