Risk Factor Identification, December 2016

(2016) Risk Factor Identification, December 2016. Transportation, Department of

Preview	PDF Final Report_risk_factor_identification_w_cvr.pdf File Size:4MB
Preview	PDF Tech Brief_risk_factor_identification_t2.pdf File Size:99kB

Abstract

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 throughout the estimation of safety performance functions. 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 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.