Predicting human behavior can better help government agencies decide when and how to issue evacuation orders in the face of significant weather events or other natural disasters. That’s the idea behind a hurricane-reaction simulation being developed by two Arizona State University professors.
Michael Barton and Sean Bergin at ASU collaborated with a team at the University of Colorado Boulder and the National Center for Atmospheric Research.
The program uses census data to create so-called “risk profiles” in the program to emulate how people might make decisions.
Bergin said patterns have been revealed based on certain lifestyles and demographics.
“Do you have a car? No. Okay, well then you're less likely to evacuate. You're gonna take more time to get out. Do you have kids? You probably are going to evacuate much faster," Bergin said.
Tool could also be used for wildfires
Bergin said they also collect anonymous cellphone data to track the number of people who stayed in certain areas following hurricanes.
Barton said this tool can help local governments know when it’s the right time to issue evacuation warnings.
“If you issue the orders too early, then people leave. You know, who are of low risk tolerance. They leave, they clog up the motels, the roads, things like that," Barton said, "When warnings get issued and in what way are vitally important to help people take the right kind of action to, you know, save their lives as well as, maybe even save property."
The team say that the program could be used for other disasters, such as wildfires.