FURI | Spring 2025

Predicting & Analyzing DUI with Statistical Modeling

Driving under the influence (DUI) remains an issue of public safety in the U.S., causing thousands of accidents and fatalities annually. This research endeavor entails the investigation of historical DUI data and building a predictive model for the forecast of future DUI occurrences. The model will factor in many different variables such as time location, demographic information, and police patrol procedures. Through statistical modeling and machine learning techniques, this study aims to provide decision-relevant information to assist in reducing the threats of DUI. Through anticipative deployment of resources by policymakers and law enforcement agencies based on high-risk locations and times, DUI accidents can be reduced, creating a safer network of roads and communities.