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{210501,
        author = {MUKILAN K and MUKESH M and SAI NITHICK ROSHAAN S and VAITHYANADHAN S G and Mrs. Ganga Naidu},
        title = {Cognitive GeoAI Digital Twin for Disaster Prognostication},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {6},
        pages = {1098-1102},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=210501},
        abstract = {A link between inundation, prediction, road accessibility and an interpretable spatial interface is required for urban flood decision support. This paper aims at proposing a cognitive geospatial artificial intelligence (GeoAI) digital twin for disaster prognostication and its initial realisation in the intended testing area of the Adyar River basin in Chennai. The proposed architecture integrates terrain and road data with a NumPy simulation, Random Forest and long short-term memory forecasting, a FastAPI/WebSocket service to feed a CesiumJS interface, and flood-aware routing. River-connected, elevation-based inundation is implemented in prototype code and dijkstra routing with depth-based road closures are implemented. The demonstrations are not a validated implementation of Adyar, but rather are attached to the discussion of Srirangam. Evaluation will be conducted using a planned approach, which will involve comparing the forecasting models with persistence and linear regression using event-separated data, and evaluate route validity, detour distance, and delivery latency. The trained-model comparison and observed-event validation and latency is pending. The contribution consists of an integrated system design, documentation of a scenario prototype and an evaluation plan for testing. Operational flood prediction is future work.},
        keywords = {Adyar basin, digital twin, evacuation routing, flood simulation, GeoAI, long short-term memory, Random Forest.},
        month = {November},
        }