Researchers at the Indian Institute of Technology (IIT) Bombay have developed an artificial intelligence (AI)-powered flood forecasting system that can predict heavy rainfall and potential flooding in Mumbai up to 90 minutes in advance, offering authorities valuable lead time for emergency response.
The system combines real-time weather radar data, satellite observations and information from more than 60 automatic weather and rainfall monitoring stations across Mumbai. Using AI and machine learning models, it generates hyperlocal rainfall forecasts and flood predictions, updating them every 15 minutes.
Developed by IIT Bombay's Interdisciplinary Programme in Climate Studies in collaboration with the Brihanmumbai Municipal Corporation (BMC), the platform is designed to help civic authorities identify areas at risk of flooding, deploy emergency teams more efficiently and issue timely warnings to residents.
The forecasting system is accessible through the Mumbai Flood website and mobile application, which provide live rainfall data, flood forecasts and water-level updates from monitoring stations across the city. Residents can also contribute by reporting flooding in their neighbourhoods, helping improve the system's real-time accuracy.
Researchers said the AI-based platform is expected to strengthen Mumbai's disaster preparedness during the monsoon by enabling quicker decision-making and reducing the impact of urban flooding through early intervention.