CTRE is an Iowa State University center, administered by the Institute for Transportation.

Address: 2711 S. Loop Drive, Suite 4700, Ames, IA 50010-8664

Phone: 515-294-8103
FAX: 515-294-0467

Website: www.ctre.iastate.edu/

Iowa State University--Becoming the Best

Risk Factor Identification


Principal investigators:

Co-principal investigators:

Project status


Start date: 07/01/15
End date: 12/31/16


Report: Risk Factor Identification (4.47 mb pdf) December 2016

Tech transfer summary: Risk Factor Identification (97.56 kb pdf) Dec 2016


Sponsor(s): Iowa Department of Transportation

About the research


The aim of this study was to provide assistance in the identification of risk factors for traffic crashes on two facility types in Iowa: intersections and horizontal curves. The risk factors were identified through the analysis of a robust database, which combined information from various sources and included traffic volumes, roadway geometry, and other characteristics.

For both intersections and horizontal curves, the researchers developed crash trees and regression models, and conducted exploratory visual analytics of Iowa’s crash data. The researchers further investigated the effects of skew angle and other factors associated with safety at rural intersections in Iowa through the estimation of safety performance functions (SPFs). The scope on this part of the study was limited to intersections on high-speed (speed limit of 45 mph or higher), rural, two-lane roadways. This analysis provides important results that reinforce the extant research literature as to the relationship between intersection skew angle and crash frequency.

The researchers also conducted a more in-depth investigation into safety risk factors for horizontal curves as a part of this study. Crash frequency data for horizontal curves were analyzed using a negative binomial modeling framework, while the crash severity data were analyzed using an ordered probit model. The results demonstrate the relationships between crash frequency/severity and various curve characteristics. Ultimately, the results of this research will allow for more effective network surveillance and identification of high-risk locations.