United StatesPublic sector2019 – ongoingCity of New York
FDNY ALARM fire-risk inspection prioritization
New York used machine-learning risk scores to prioritize fire inspections across approximately 330,000 buildings
Problem
The Fire Department needed to prioritize inspections across a large population of buildings. Inspection scheduling had to account for both fire probability and potential civilian harm.
Approach
An internal analyst team built ALARM and first used it in November 2019. The system combined machine-learning estimates of structural-fire probability with injury and fatality risk ratios. Its building scores supported scheduling within the Risk-Based Inspection Program.
Results
Takeaway
The city documented an operational scheduling tool. It did not report a measured reduction in fires or casualties.
Categories
- Size
- Large enterprise
- Industry
- Public sector
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
The inventory documented ALARM's use in fire-inspection scheduling and dated first use to November 2019. It described an internally built model serving an inspectable population of approximately 330,000 buildings.
Full details are in the original sources.