Chicago developed a predictive model that forecasts which restaurants are the most high-risk for critical violations, and then allocates food inspectors accordingly, reducing the public’s exposure to potential food poisoning in the process. Following a pilot launch that demonstrated a 26 percent increase in efficiency from inspection teams using the model, the Chicago Department of Public Health (CDPH) began incorporating the forecasting model into their operations. Chicago’s data science team also built an application for the model that provides a complete list of predictions, with the riskiest establishments listed first. For DPH’s ease-of-use, the application is also easily navigable, allowing for specific restaurants to be referenced by the food inspection manager to triage complaints as needed. Today, the analytical model is updated daily, and the food inspection team now continually allocates its workers based on the suggestions made through the application. The model itself is built using Chicago’s open data portal and it is entirely available for replication in other cities, or for collaboration with researchers.