Project: Identification of traffic patterns leading to crashes

Principle Investigator: Dr. Nicholas Garber, University of Virginia, 804-924-6366, njg@virginia.edu

Project Objective: To identify the patterns of interactions of the traffic characteristics that lead to all crashes, and that led to specific types of crashes. (Full project title: Identification of traffic patterns leading to different types of crashes on interstate highways in Virginia.)

Project Abstract: This project is a follow-on to the author's just-completed ITS Center project, "Modeling of vehicle crash causes", which found that, using negative binomial and zero inflated Poisson models, correlations exist between crashes and both volume of traffic and standard deviation in the speed of traffic. The proposed study hypothesizes that, when road geometries are similar, crashes are due primarily to the interaction of traffic characteristics existing immediately before the crash. That is to say, there is no single obvious cause of crashes, but, rather, crashes are due to a combination of factors. The research will use UVA Smart Travel Lab data to develop an understanding of these multiple causal factors.

Tasks: 1) Literature review. 2) Retrieval of crash data from the database at the Smart Travel Laboratory. 3) Retrieval of corresponding traffic data. 4) Analysis of crash data. 5) Analysis of traffic data. 6) Identification of interaction patterns. 7) Prepare report.

Milestones: Completion by June 30, 2006. (Assumes no-cost grant extension.) .

Student Involvement: 2 students

Budget

FY04

Faculty 40,000

FY05

Faculty 160,000
Students 20,000

Project total 220,000

Relation to other research: This follows directly from the Center's "Modeling of vehicle crash causes", by the same scholar.

Technology Transfer: Presentations at conferences, publication of journal papers, final report

Potential Benefits: Increased attention to safety factors in the planning and implementation of traffic management.

TRB Keywords: ITS, safety, traffic management