AegisNet: An Explainable Multi-Hazard Digital Twin Framework for Disaster Risk Forecasting and Emergency Resource Optimization.

Objective: This study investigates the hypothesis that integrating Digital Twin simulation with Explainable Artificial Intelligence (XAI) can improve the transparency and operational usefulness of disaster risk forecasting systems without significantly increasing computational latency Methodology: The AegisNet framework is comprised of five interconnected layers: Data Acquisition, Digital Twin, Forecasting, Explainability, and Resource Optimization Key Formulation: The framework defines the overall disaster risk score (R_t) as R_{t}=\alpha E_{t}+\beta H_{t}+\gamma D_{t}+\delta V_{t}, where \alpha + \beta + \gamma + \delta = 1 and the score is normalized to 0–100 Impact: The framework enables continuous state synchronization, which improves situational awareness and supports scenario analysis for emergency resource prioritization