Model predicts broken fire alarms

New Orleans' city analytics team wanted a computer model that predicts the likelihood any given city block will lack alarm coverage. Enter New York city data experts Enigma, who used public data sets to generate simple predictive maps for cities across the U.S.

25,000 U.S. residents are killed or injured each year in 1 million fires, reports Enigma, which has appealed for more local governments to upload their data.

Flowing Data summarizes the methodology:

To develop the model, they used data from the American Housing Survey and American Community Survey, both conducted by the Census Bureau, as the seed of their solution. The former asks residents if they have a working smoke alarm, which makes it seem straightforward to build a model, but it's only at the city level. Not useful for on-the-ground workers. The task required block-level information.

As you can see above, Pittsburgh is basically one inaccurately-flicked cigarette away from going completely up in flames.