The Chicago Police Department has used predictive analytics to determine where to position officers to prevent retaliatory shootings.
Why not do the same to decide which of Chicago’s 15,000 restaurants and food establishments to inspect first, based on how likely they are to face health code violations?
Mayor Rahm Emanuel’s administration is doing just that, thanks to a $1 million prize from a philanthropic organization formed by retired New York Mayor Michael Bloomberg to help cities devise new ways, through savvy use of loads of data, to deliver old services.
The grant helped bankroll research conducted by the city’s Department of Innovation and Technology in collaboration with the Civic Consulting Alliance and Allstate Insurance.
The researchers identified a host of risk factors that could trigger health code violations. They include 311 requests, sanitation complaints at establishments in the area, and information on previous inspections and permits.
All of those factors were thrown into the mix to create a “predictive analytics model” used to determine which restaurants needed to be inspected first to get ahead of critical violations that posed the greatest threat of food-borne illness.
With a $3.1 million Food Sanitation Division comprised of just 42 employees charged with overseeing 15,000 food establishments, it doesn’t take a rocket scientist to figure out that Chicago needs to find a way to do more with less.
That’s where predictive analytics comes in handy.
“This new tool helps us better direct resources to service those restaurants most in need,” Health Commissioner Dr. Julie Morita said in a news release.
Gerrin Cheek Butler, the department’s director of food protection services, said the earlier the inspection, the better it is for restaurant owners, employees and patrons.
“This innovation allows us to give guidance to establishments earlier, so they can make the necessary correction and get back to serving their customers. … This also helps ensure possible problems do not become worse over time,” she said in a news release.
Two years ago, the Chicago Sun-Times disclosed that the city was using a predictive analytics model also bankrolled by the Bloomberg Philanthropies’ Mayors Challenge to curb Chicago’s burgeoning population of rats.
The Department of Innovation and Technology identified 31 city service requests not directly related to rat sightings that trigger rodent control complaints. They ranged from stray animal calls to vacant and abandoned buildings, missing or overflowing garbage carts, and restaurant violations made within seven days in the same general area.
By scouring 311 service requests for those patterns on a daily basis, the city no longer had to wait until residents “smell a rat” or see one.
Streets and Sanitation Commissioner Charles Williams could dispatch one of his roughly dozen rodent control crews to start baiting before rats establish a colony.
Predictive analytics has also been used before in the area of food safety through a 2013 program known as FoodBorne Chicago. That project scoured Twitter for food poisoning complaints from Chicago residents, employees and tourists. Authors of those tweets were encouraged to file formal complaints, triggering 150 additional inspections.
