Third Year Candidacy Presentation: Spatial Variations in Police Response Time in the City of New Orleans

This talk analyzes data from the New Orleans Police Department on all calls for service in 2018. The primary variable of interest is response time, defined as the duration between the placement of a 911 call and an officer’s arrival at the scene. Differential response systems classify these calls by priority, resulting in varied response times, with non-critical calls potentially facing delays. Other key factors include neighborhood effects—which serve as a proxy for queuing assignment, patrol unit distance, and real-time traffic conditions—along with call volume and time of day. For the literature review, we examine accelerated failure time (AFT) models within survival analysis. We then propose a two-component mixture of AFT models to account for immediate and non-immediate officer dispatch. Additionally, a logistic regression model is used to analyze the selection of immediate dispatch. Findings from both response time modeling and dispatch selection analysis are presented. Finally, we discuss hierarchical spatial modeling and computational approaches to further enhance the analysis.

Advisor: Debashis Mondal