In the chaos of large-scale emergencies, AI software could help direct first responders.

The “Rescue” system developed by researchers at Durham University, UK, comprises up to 4000 individual software agents that represent the public and members of emergency services, programmed with behaviors such as “help an injured person.”

Researchers at the University of Notre Dame have built a detailed simulation of how crowds respond to disaster. The Dynamic Adaptive Disaster Simulation (DADS) uses software agents representing humans programmed to flee from danger and move towards safety.

When used in a real emergency situation, DADS will use location data from thousands of cellphones, triangulated and streamed from the region of the emergency. It can make predictions of how crowds will move by advancing the simulation faster than real-time events.

Running such a simulation in a similar study, a French team found that as they increased a crowd’s density, the model produced waves of people just like those seen in real-life events.