2010Transportation Research Board 89th Annual MeetingTransportation Research BoardRequires access

Examination of Horizontal Curve Collision Characteristics andCorresponding Countermeasures

Joseph E. Hummer, William Rasdorf, Daniel J. Findley, Charles V. Zegeer, Carl Sundström

Open publisher page 0 citations

Abstract

Horizontal curves are relatively dangerous portions of the roadway network. Agencies optimizing the use of their safety funds on curves should have an idea of the characteristics of the collisions that occur on those segments. However, few previous published papers had attempted to characterize the crashes on horizontal curves. This paper describes an effort that characterized crashes reported to be on curves in North Carolina using the Highway Safety Information System. The research compared crashes on two-lane road curves to crashes on all two-lane roads and to crashes on all roads. Since North Carolina has wide varieties of terrain, climate, and development, the findings should be generally representative of the US. The analysis revealed that over-represented two-lane curve crash types included: • Crashes on grades;Rural;Severe injury or fatal; Fixed object (particularly tree, ditch, and embankment); Overturn; Off peak hours (particularly during darkness on unlighted roads); Weekend; Holiday periods; and Wet, icy or snowy pavement. In addition, the analysis showed that there were few short roadway segments (of 0.1- to 1-mile length) with more than 10 reported curve crashes in 3 years. The paper provides a summary of recommendations from the national literature to treat the over-represented crash types. Based on the findings, agencies should target countermeasures for the most common and over-represented crash types. Agencies should also analyze longer segments to provide curvy segments with concentrations of crashes a chance to compete for funding with other types of segments.

About this research paper

What this paper is about

Horizontal curves are relatively dangerous portions of the roadway network. Agencies optimizing the use of their safety funds on curves should have an idea of the characteristics of the collisions that occur on those segments. However, few previous published papers had attempted to characterize the crashes on horizontal curves. This paper describes an effort that characterized crashes reported to be on curves in North Carolina using the Highway Safety Information System. The research compared crashes on two-lane road curves to crashes on all two-lane roads and to crashes on all roads. Since North Carolina has wide varieties of terrain, climate, and development, the findings should be generally representative of the US. The analysis revealed that over-represented two-lane curve crash types included: • Crashes on grades;Rural;Severe injury or fatal; Fixed object (particularly tree, ditch, and embankment); Overturn; Off peak hours (particularly during darkness on unlighted roads); Weekend; Holiday periods; and Wet, icy or snowy pavement. In addition, the analysis showed that there were few short roadway segments (of 0.1- to 1-mile length) with more than 10 reported curve crashes in 3 years. The paper provides a summary of recommendations from the national literature to treat the over-represented crash types. Based on the findings, agencies should target countermeasures for the most common and over-represented crash types. Agencies should also analyze longer segments to provide curvy segments with concentrations of crashes a chance to compete for funding with other types of segments.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Horizontal curves are relatively dangerous portions of the roadway network. Agencies optimizing the use of their safety funds on curves should have an idea of the characteristics of the collisions that occur on those segments. However, few previous published papers had attempted to characterize the crashes on horizontal curves. This paper describes an effort that characterized crashes reported to be on curves in North Carolina using the Highway Safety Information System. The research compared crashes on two-lane road curves to crashes on all two-lane roads and to crashes on all roads. Since North Carolina has wide varieties of terrain, climate, and development, the findings should be generally representative of the US. The analysis revealed that over-represented two-lane curve crash types included: • Crashes on grades;Rural;Severe injury or fatal; Fixed object (particularly tree, ditch, and embankment); Overturn; Off peak hours (particularly during darkness on unlighted roads); Weekend; Holiday periods; and Wet, icy or snowy pavement. In addition, the analysis showed that there were few short roadway segments (of 0.1- to 1-mile length) with more than 10 reported curve crashes in 3 years. The paper provides a summary of recommendations from the national literature to treat the over-represented crash types. Based on the findings, agencies should target countermeasures for the most common and over-represented crash types. Agencies should also analyze longer segments to provide curvy segments with concentrations of crashes a chance to compete for funding with other types of segments.

Key concepts: Crash, Terrain, Collision, Transport engineering, Ditch, Geography, Forensic engineering, Computer science

Related papers

Back to paper searchBrowse research topicsOriginal source
Examination of Horizontal Curve Collision Characteristics andCorresponding Countermeasures — Research Paper | ScholarLens