Integration of Artificial Intelligence and HEC-RAS for Rapid Flood Inundation Prediction and Risk Assessment | IJET
Mohammad Sayeed, Bhuvan Chandra Bhatt, Mohit Kumar
Abstract: Flooding is one of the most frequent and destructive natural hazards, causing significant damage to human life, infrastructure, agricultural land, and the environment. Accurate and rapid prediction of flood inundation is therefore essential for effective disaster preparedness and risk management. HEC-RAS is widely used for one-dimensional and two-dimensional hydraulic modelling and flood inundation mapping; however, detailed simulations may require considerable computational effort, particularly for large study areas, high-resolution terrain data, and multiple flood scenarios. The integration of Artificial Intelligence (AI) with HEC-RAS provides a promising approach to overcome these limitations by combining physically based hydraulic modelling with the rapid predictive capability of data-driven techniques. This review examines the application of machine learning, artificial neural networks, deep learning, and surrogate modelling techniques in combination with HEC-RAS for rapid prediction of flood depth, flow velocity, inundation extent, and flood hazard. It further discusses the supporting role of Geographic Information Systems (GIS), remote sensing, Digital Elevation Models (DEMs), hydrological observations, and satellite-derived flood information in developing reliable AI-assisted flood assessment frameworks. Particular attention is given to model development, training data generation from HEC-RAS simulations, performance evaluation, validation, computational efficiency, and practical applications in flood risk management. The review indicates that AI–HEC-RAS integration can substantially reduce prediction time while maintaining acceptable accuracy, making it particularly valuable for rapid scenario analysis and near-real-time flood assessment. However, challenges associated with data availability, model generalization, uncertainty, interpretability, and transferability remain important. Overall, an integrated AI–HEC-RAS framework has strong potential to support faster flood forecasting, improved inundation mapping, and more effective risk-informed decision-making for resilient and sustainable flood management. Keywords: Artificial Intelligence, HEC-RAS; Flood Inundation, Machine Learning, Deep Learning, Flood Risk Assessment, Hydraulic Modelling.