Vehicle Speed and Risk Modeling of Horizontal Curves at a Network Level

Managing the safety of road users traversing horizontal curves is a major issue confronted by road agencies globally. Statistics from the United States and New Zealand show that around 25 percent of fatal and serious crashes occur on horizontal curves. Traditional methods of detecting safety issues tend to involve geospatial analysis of crash data to highlight blackspots and reveal crash trends. Whilst these approaches enable horizontal curves with an established safety problem to be identified, they miss curves with an inherent high level of risk where few crashes have occurred in the past. This paper presents a geospatial risk prediction methodology that models vehicle speeds along high-speed road corridors and assesses the safety risk of horizontal curves based on curve approach speed and curve radii at a network level. Results show that injury crash rates on horizontal curves classified as high-risk using the methodology are approximately 95% higher than other horizontal curves and 450% higher than straight road segments. These findings demonstrate that the horizontal curve risk assessment methodology is a strong indicator of underlying safety risk. Building off this methodology, a prioritization process was developed to identify corridors with the highest risk of curve crashes. This process established that the highest ranked 10% of corridors by length had a curve-related injury crash rate that was 97% higher than the next highest ranked 15% of corridors. This proactive approach of identifying high-risk corridors is helping road agencies across Australasia target their efforts at a comparatively small proportion of the network where a disproportionately large amount of risk exists.

• Supplemental Notes:
  • This paper was sponsored by TRB committee ANB20 Standing Committee on Safety Data, Analysis and Evaluation.
• Authors:
  • Abeysekera, Raja
  • Brodie, Colin
  • Durdin, Paul
  • Gardener, Robyn
  • Harris, Dale
  • O'Neil, Carl
• Conference:
  • Transportation Research Board 97th Annual Meeting
  • Location: Washington DC, United States
  • Date: 2018-1-7 to 2018-1-11
• Date: 2018

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; Maps; References;
• Pagination: 18p

Subject/Index Terms

• TRT Terms: Crash injuries; Crash rates; Highway curves; Mathematical models; Mathematical prediction; Risk assessment; Speed
• Geographic Terms: New Zealand; United States
• Subject Areas: Highways; Safety and Human Factors;

Filing Info

• Accession Number: 01657017
• Record Type: Publication
• Report/Paper Numbers: 18-01591
• Files: TRIS, TRB, ATRI
• Created Date: Jan 23 2018 9:28AM