Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{210357,
        author = {Dr MK Jayanthi Kannan and JAI ARORA and Yesh Sharma and Uday Pratap Singh and Satyam and Nandish Dave and VAIBHAV SHARMA},
        title = {AI-Based Flood Risk Prediction and Early Warning System for Disaster Management},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {6},
        pages = {822-833},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=210357},
        abstract = {Floods remain one of the most frequent and destructive natural disasters affecting India, with states such as Assam experiencing recurring, severe flooding almost every monsoon season. India has recorded a 40% rise in flash flood events over the past three years, and Assam alone saw 4.29 million people affected across 35 districts in 2024. Existing flood management systems are largely reactive, region-specific, and lack the ability to generalize predictions to new or data-scarce areas. This paper proposes an AI-based flood risk prediction and early warning system trained on historical rainfall, river-level, and geographic data from multiple Indian states. A leave-one-region-out validation strategy is employed, in which the model is trained on data from all major flood-prone states except Assam and then tested specifically on Assam to evaluate its ability to generalize to a high-risk region not seen during training. The system further includes a dashboard for visualizing flood risk and an alert mechanism designed to reduce response delay and potential loss of life. By combining machine learning-based prediction with a practical early warning interface, this work aims to move flood disaster management from a reactive to a proactive model, addressing a gap left open by existing conceptual, descriptive, and narrowly-scoped technical approaches in the literature.},
        keywords = {Flood Prediction, Machine Learning, Disaster Management, Early Warning System, Assam, Generalization, Decision Support System, Flood Risk Mapping.},
        month = {November},
        }