The city of New Orleans used data science to formulate a preventive approach to firefighting. As part of its Targeted Smoke Alarm Outreach Program, the city developed a predictive model to identify areas at the highest risk of fires and fire fatalities. The data fed into the model came from open sources such as the Census American Housing Survey and American Community Survey, as well as the fire department’s own data. Taking into account factors such as poverty, building age, location, previous fire history, and the likelihood of dwellings having fire alarms, the project turned once-siloed data into actionable insights.
Officials created a heat map of the city to pinpoint areas for a door-to-door campaign. For instance, since the analysis revealed that those under 5 and over 60 were most susceptible to fire fatalities, authorities distributed and installed fire alarms in areas with concentrations of these age groups. New Orleans distributed more than 7,500 alarms by the end of 2015. Analytics, cross-agency collaboration, and data integration helped the city optimize its resources to protect its most vulnerable residents