Flood disaster management—particularly in urban and rural areas that suffer the greatest impacts—holds critical importance. In Iran, recent floods such as the large-scale events of 2018–2019 in the provinces of Golestan, Lorestan, and Khuzestan affected more than 10 million people, resulting in multi-billion-dollar economic losses, infrastructure destruction, and loss of life (Seddighi and Seddighi, 2020). These crises not only disrupted the lives of millions but also led to long-term challenges such as forced migration, reduced food security, and ecosystem degradation. In this context, the use of numerical flood models, combined with advanced technologies such as machine learning, can significantly enhance the accuracy of flood prediction and risk assessment. The aim of this postdoctoral project is to conduct high-precision numerical flood modeling and prepare flood hazard and vulnerability maps using machine learning algorithms, providing an effective decision-support tool for urban, environmental, and crisis management. It aligns with the United Nations Sustainable Development Goals (SDGs), particularly Goal 13 (Climate Action) and Goal 11 (Sustainable Cities and Communities). The project emphasizes the use of open data and cloud-based tools to develop scalable and transferable solutions for crisis management in data-scarce regions of Iran, aiming to reduce economic and social losses and to contribute to progress under the Sendai Framework for Disaster Risk Reduction.