IMPROVING FLOOD MANAGEMENT WITH ARTIFICIAL INTELLIGENCE: APPLICATIONS IN CRISIS MANAGEMENT

Floods are worldwide among the most significant natural hazards. They can result from various causes, including severe storms, intense or prolonged rainfall, and landslides. The damage they cause is substantial, both economically and in terms of human life and wellbeing. Modern technologies, particularly artificial intelligence (AI), offer wide-ranging opportunities in this domain. AI can help forecast the weather, rainfall amounts and intensity, and other variables influencing flood occurrence. When integrated into early warning systems, these outputs enable location-specific, probabilistic alerts that trigger timely preparedness and response actions. By processing and analysing large volumes of data, AI supports more effective prevention, preparedness, and response within crisis management. This paper examines the applications of AI in crisis management in the context of floods and introduces existing prevention and response systems from selected countries. It proposes a conceptual model that shows how AI components can be integrated with existing crisis management infrastructures and early warning systems along the whole flood risk management cycle. In doing so, it underscores the need to integrate modern technologies, especially AI, into decision support for crisis management, including in regions with limited resources, where modular and scalable solutions are particularly important. Several countries have already implemented, or are working to implement, such technologies in crisis management. These examples provide valuable inspiration for developing future systems that build on best practice and lessons learnt and point to future research needs in areas such as interoperability, real-time data integration, and the operational deployment of AI in flood management.