AI model predicts building fire spread, redirecting evacuees to safer exits

A fire alarm goes off. You head for the nearest exit. But what if that exit is already engulfed in flames?

Researchers at the National Institute of Standards and Technology have developed an AI model called Safe Step, designed to predict how fires spread and redirect building occupants toward safer exits in real time.

Unlike traditional evacuation algorithms that only assess current conditions, Safe Step uses reinforcement learning to anticipate how a fire will evolve, factoring in toxic gas exposure along every possible route.

"Fires can grow and spread," said NIST research associate Hongqiang Fang. "Our model forecasts how the fire is evolving and can help update emergency exit displays to direct people toward the safest exit."

The system could eventually coordinate evacuees across multiple floors while guiding firefighters inside simultaneously. Researchers estimate it could appear in buildings within five to ten years, pending regulatory approval and reliability testing.