This project aims to develop an AI-supported crisis management and evacuation decision-support framework that delivers real-time situational intelligence during emergencies in complex buildings. The approach integrates IoT sensing, indoor positioning, environmental monitoring, and human-centered decision interfaces to help emergency teams interpret and respond to rapidly changing conditions. The system provides continuous insight into hazards, help requests, occupancy levels, ongoing activities, and anomalies, while tracking the location of both evacuation teams and occupants through sensor infrastructure and a mobile application. Timely coordination and informed decision-making enable safer, more efficient evacuations. In this project, we integrate smartphone sensors, cameras, radar, and other environmental remote sensors. The current ongoing sub-projects are: - Real-time Indoor positioning/Tracking the evacuation leaders/ occupants during evacuation - Prediction of Overcrowding at Exits by monitoring the Building Evacuation Flow - Estimation of total Escape Time of the occupants (evacuation time) by monitoring and tracking the flow - Physical Obstacle/Hazard Detection during evacuation - Real-time People Counting and Occupancy level detection - Activity / Intention Recognition during evacuation - Anomaly Behavior Recognition during evacuation - Guidance to an Optimal Evacuation Route - Enhancing Inclusive Emergency Communication for Evacuation Leaders and Occupants