Home Skip to main content

Project Details

Title:Exploration of Alternative Spatio-Temporal Methods of Traffic Safety Network Screening
Principal Investigators:Michael D. Pawlovich
University:South Dakota State University
Grant #:69A3551747108 (FAST Act)
Project #:MPC-627
RiP #:01743853
RH Display ID:154690
Keywords:cluster analysis, geographic information systems, network analysis (planning), spatial analysis, traffic crashes, traffic safety


The roadway system represents a major investment and valuable resource that enables mobility and accessibility to users. With rising costs, tight budgets, and limited resources, agencies seek techniques to identify critical mobility and safety concerns, manifested through federal encouragement to increase safety data collection, analysis, and implementation. Network screening methods exist, both historical and recent, with all the methods relying on connection of crashes to the roadway network. Historically, this has been accomplished through assignment of crashes to the network. More recently, methods have been developed that first analyze the spatiotemporal nature of crashes and then connect the resultant clusters to the network. Because the methods either rely on arbitrary network typologies or ignore the network, cohesive screening results may not be provided.

We will explore GIS and spatiotemporal analysis techniques which rely on the crash locations and densities with a coincident network connection. We will examine spatial proximity constrained by the network to form crash distributions and analyze these for distributional clustering. Network characteristics will be included during distribution and cluster development. The goal is to develop a method that produces an efficient and effective means of crash cluster identification. Crash typology distributions and clusters will also be analyzed.

Project Word Files

NDSU Dept 2880P.O. Box 6050Fargo, ND 58108-6050