January 20, 2017

Chicago’s Predictive Analytics System Recognized by Harvard as a 2016 Top Innovation in Government

Erica Duncan    Erica.Duncan@cityofchicago.org

The City of Chicago’s predictive analytics system for food inspections has been honored as a Top 25 Bright Idea by Harvard University. The Ash Center for Democratic and Innovation at the John F. Kennedy School of Government, Harvard University recognized the new system implemented by the Chicago Department of Public Health (CDPH). The system leverages public data to identify Chicago restaurants most likely to face health code challenges, allowing CDPH inspectors to prioritize inspections for those restaurants, helping them resolve any issues as quickly as possible and prevent food borne illnesses before they can affect residents. This innovation has allowed CDPH inspectors to give proactive guidance to food serving establishments so that they can sustain operations and serve patrons in a safe and healthy manner.

“This recognition is an important reminder of Chicago’s leadership in collaboration and innovative use of technology to keep residents safe,” said CDPH Commissioner Julie Morita, MD. “CDPH is committed to maintaining the safety of food bought, sold or prepared for public consumption in Chicago. Our use of this predictive analytics system is a testament to our ability, to maximize efficiencies and to more effectively oversee conditions of more than 15,000 food establishments city wide and to give our residents confidence that their local eateries are safe.”

Through predictive analytics CDPH’s sanitarians can be more pointed in prioritizing retail food establishments such as restaurants, grocery stores, bakeries, convenience stores, day care facilities, schools and temporary food service events. Inspections focus on food handling practices, product temperatures, personal hygiene, facility maintenance and pest control.

The tool uses data from various City of Chicago data systems along with sophisticated research techniques to identify factors most likely to result in a restaurant facing health code violations—including established risk categories, a range of 311 data, nearby sanitation complaints, previous inspection data and permitting data. An evaluation of the model found that the violations most likely to lead to food poisoning were found 25 percent earlier using predictive analytics. Over a 60-day period, critical violations were found over one week earlier. The research was conducted in collaboration with the Department of Innovation and Technology (DoIT), Civic Consulting Alliance, and pro bono assistance from Allstate Insurance.
The project was released as a freely available, open source tool that can be adopted by other governments and can be evaluated by other researchers. In 2015, Open Data Nation—a startup company—worked with Montgomery County, Maryland to adopt the food inspection model. Since releasing the project, dozens of researchers, students, and data scientists have contributed to the project.
The new system was created by the City in 2012 as part of Bloomberg Philanthropies’ Mayors Challenge—an ideas competition to generate innovative ideas that solve major challenges and improve city life—and have the potential to spread to other cities. The City of Chicago was awarded $1 million in 2013 to develop a data-informed approach to decision-making. The FoodBorne predictive analytic app utilizes open code to forecast, an approach often leveraged by business industry, but rarely used in public health. Since its launch, it has been consistently modeled after by other leading organizations and municipalities.

“Chicago’s predictive analytics system allows us to make smarter, quicker decisions to address issues before they arise,” said Brenna Berman, Commissioner and CIO of DoIT. “This platform has allowed us to develop nimble responses that ensure efficient and effective use of City resources.”

“Chicago’s development and implementation of predictive analytics illustrates a real example on how governments can use technology, analytics, and civic engagement to make an impact on everyday life,” said Steve Goldsmith, the Daniel Paul Professor of Practice of Government at Harvard’s Kennedy School of Government. “We are encouraged by Chicago’s ability to bring together diverse stakeholders to drive greater efficiency.”

The Innovations in American Government Awards was created by the Ford Foundation in 1985 in response to widespread pessimism and distrust in government’s effectiveness. Since its inception, more than 500 government innovations across all jurisdiction levels have been recognized and have collectively received more than $22 million in grants to support dissemination efforts. Such models of good governance also inform research and academic study around key policy areas both at Harvard Kennedy School and academic institutions worldwide. Past winners have served as the basis of case studies in more than 450 Harvard courses and over 2,250 courses worldwide.

A full list of the Top 25 programs is available here.


 News Release Facts

 I Want To