CMU program helps Pittsburgh predict where fires will break out

CMU research program helps Pittsburgh predict where fires will break out

Pittsburgh says it has the technology to predict fires before they happen, thanks to work being done at Carnegie Mellon University.

It's called predictive analytics, and Mayor Bill Peduto says it’s leading Pittsburgh to becoming one of the safest cities in the world for fire prevention.

“What if you could use technology to determine the next building to catch fire?” he said.

Content Continues Below

The city has joined forces with CMU to predict which commercial buildings are at risk of catching fire.

“We will be able to stop that risk before a tragedy occurs,” said Wendell Hissrich, the city’s public safety director.

Using reports from city building and fire inspectors, including violations, CMU researchers using a computer program compare and contrast the data and determine which buildings are at risk.


“The features that are most important are density, size … property value is a factor,” said Michael Madaio of CMU.

And they've had alarming success. They identified 57 of the highest risk buildings in the city, and 50 of them had some sort of fire incident, ranging from smoke detector alarms to actual fires.

"We are able to drill down to an 80th percentile of understanding where the next fire is going to occur," Peduto said. "It is at a level that is much higher than anywhere else in the country."

The fire department will then use the data to alert property owners of potential fire hazards and suggest improvements to reduce the risk.

“We make sure that systems that are integral are functioning, like sprinkler systems, standpipe systems, alarm systems,” said Fire Chief Darryl Jones. “We make sure that the residents of those buildings understand and are educated in fire safety.”

Peduto said the program, funded by several foundations, also can lead to lower property insurance rates.