Data Tool Created By CMU Student Helps Predict Fire-Prone Buildings In Pittsburgh

Follow KDKA-TV: Facebook | Twitter

PITTSBURGH (KDKA) -- A data tool created by a Carnegie Mellon University doctoral student is helping the City of Pittsburgh with fire prevention and safety.

Michael Madaio is the student behind the predictive model that determines the fire risk for the city's commercial buildings, helping prioritize inspections.  There are roughly 22,000 commercial properties in the city

"It's one more thing we have in the tool box," said Pittsburgh Bureau of Fire Chief Darryl Jones. "It allows us to take a look from a bird's-eye view of where potential fires will occur. And we will use this information to concentrate on inspection and fire prevention efforts in those areas."

The project started in July 2017, and in the first six months, it identified 57 properties with a high-risk of fire. Of those, 50 properties or 88 percent, did experience fire incidents.

The model uses artificial intelligence as well as past fire incidents, property assessments, and other data to calculate the risk.

"We will continue to push and change our culture so that we will be a leader not only in the nation but in the world when it comes to fire prevention and fire safety," Chief Jones said.