Can analytics help firefighters save lives?
- By Mark Pomerleau
- Apr 08, 2015
The New Orleans Fire Department is using analytics as the basis of its smoke alarm outreach campaign, in which firefighters install alarms in neighborhoods where data has shown they are most needed.
The fire department and the city's Office of Performance and Accountability (OPA) created a predictive model that identifies those who are least likely to have a smoke alarm and most likely to die in structure fires.
Using data from the American Housing Survey, the American Community Survey, the latest census and NOFD administrative data, OPA analysts created a model that maps the risk of fire fatalities by block group. The model factors in the age of the structure, previous fires and the residents’ age, poverty level and length of time in residence.
Armed with that information, firefighters were able to target specific neighborhoods for free smoke alarm installation.
New York City employed a similar data-based analytical model a few years ago to predict and target buildings to inspect. With 300,000 buildings in the Fire Department’s inspection universe and the resources to only inspect 50,000 of those buildings a year, the city needed a way to prioritize inspections. A data warehouse was built to store all inspection information and a risk-based inspection system scores and tracks buildings that the fire department should prioritize for inspection.
Mark Pomerleau is a former editorial fellow with GCN and Defense Systems.