SWIFTAID: An AI-Driven Ambulance Dispatch and Route Optimization Platform for Dense Urban Environments

Urban ambulance response in dense Indian cities such as Bengaluru is frequently delayed by heavy traffic, reac- tive dispatch protocols, poor coordination and limited real-time visibility. SWIFTAID is a platform that integrates a mobile-first SOS interface, driver application, cloud backend and machine- learning modules for predictive dispatch and traffic-aware route optimization. The system captures accurate location and min- imal medical metadata from patients, applies ML to allocate ambulances and rank route alternatives using historical and live traffic signals, and provides continuous tracking for patients, drivers and hospital staff. The core innovation is a complete machine learning pipeline that collects real ambulance trip data in Firebase, trains gradient-boosted regression models (XGBoost)on historical traffic patterns, and performs real-time route predic- tion with R2 = 0.84 accuracy. Evaluation under Bengaluru traffic conditions demonstrates average response time improvements of 18–22% during peak hours compared to conventional GPS-based routing.