SafeCurves: Identify - Primer

Horizontal curves

What makes them so high risk and how can we improve outcomes?

The purpose of this SafeCurves Primer series is to answer questions about horizontal curves and their importance in road safety.

What is the problem?

The USA has a big road safety problem, with more than 42,000 people killed on US roads each year (National Center for Statistics and Analysis, 2023). At 12.61 deaths per 100,000 people, the US is significantly underperforming when compared to other developed countries (Institute for Health Metrics and Evaluation, 2019).

Managing the safety of road users traversing horizontal curves is a major issue for road agencies globally. In the United States, over 25% of all fatal crashes occur on horizontal curves. The average crash rate on horizontal curves is approximately three times higher compared to straight sections of road (FHWA, 2023).

What makes horizontal curves high risk?

Horizontal curves are one of the most challenging features that drivers must negotiate on high-speed rural roads.
Drivers can be surprised by curves that are inconsistent with what they are expecting, elevating the risk of roadway departures and head-on collisions. These types of curves are regarded as ‘out-of-context’ – the degree of curvature is out-of-context with the driver’s experience prior to reaching the curve. Examples of out-of-context curves include:

• an isolated curve on an otherwise long and straight road

• the first curve entering a series of curves

• an unexpectedly sharp curve among a sequence of larger-radii curves.

Research shows that the crash risk on horizontal curves is strongly correlated with the curve radii and the difference between the approach speed and curve negotiation speed. For example, a curve that requires a driver to reduce their speed by 30 km/h (approximately 20 mph), from an approach speed of 100 km/h (approximately 60 mph), elevates the risk of run-off-road casualty by 5.1 times (Austroads, 2015; Cardoso, 2005).

What is done now to identify risk: site analysis versus a systemic approach

Traditionally, safety issues with horizontal curves are detected by looking at the crash data to identify blackspots and reveal crash trends. This is known as site analysis. However, relying on crash data to identify high-risk sites is a reactive approach to managing safety, as road managers must wait for crashes to occur before deficiencies can be addressed. This approach can sometimes amount to chasing crash hotspots while failing to identify where serious crashes could occur in the future. This is particularly problematic on lower volume roads where crashes tend to be sporadic and more difficult to predict.

Increasingly aware of the limitations of crash data, road managers are using proactive methods to identify high-risk parts of their road network. These methods consider the underlying attributes of a road network and use established research to highlight sites where high-severity crashes should be expected. This is known as a systemic approach.

A systemic approach aims to identify the comparatively small proportion of the road network where there is a disproportionately large amount of risk to road users. In doing so the greatest reward can be achieved, for potentially the least investment.

Currently, practitioners assess horizontal curves in a fragmented way that is often very time-consuming, can include significant analytical variability and does not provide any prioritization method. US practitioners are advocating for change, including the rapid adoption of the Safe System approach.
This requires accurate, appropriate and timely data.

How many horizontal curves are there in the US?

We have categorised horizontal curve context across all high-speed roads (greater than 40mph) in the United States. The analysis covers approximately 1,443,000 lane miles (or more simply, 847,364 centreline miles) and identified 1,702,049 horizontal curves.

These horizontal curves were categorised into the following performance classes:

• Class 1 (unacceptable) (10%),

• Class 2 (undesirable) (12%), and

• Class 3 (desirable) (78%).

Class 1 and Class 2 curves are collectively referred to as ‘out-of-context curves’.

What demonstrates out-of-context curves need treatment?

On average, there are 1378 fatalities per year on out-of-context (Class 1 and Class 2) curves within the United States. The length of these curves represents 4% of the total analyzed network length, as shown in Figure 1. While out-of-context curves represent only 4% of the total analyzed network length, out-of-context curves represent 8% of all fatalities¹ as shown in Figure 2.

That is because curves require more cognitive effort to be successfully negotiated, so there is a greater probability a driver will make a mistake leading to a crash, and that the crash will result in a fatality. This underscores the significance of investigating roadway departures (RwD) and lane departure (or loss of control) crashes² on curves.

Figure 1 Length by segment type

Figure 2 Fatalities by segment type

Unsurprisingly, the fatal crash rate per length of road is higher for out-of-context curves than other road segments. This is shown in Figure 3, which presents the average fatal crashes (per annum) per 1,000 miles (excluding intersections) by straight and curve class. For comparison, for the entire length of the I95 (Maine to Florida) for 2020 was 2 fatal crashes per 1,000 miles.³

What is done to address safety issues at horizontal curves? 

Having consistent geometric design along routes is important for rural road safety outcomes. Where consistent geometry cannot be provided, treatments should be implemented to alert drivers to the presence of out-of-context curves and reduce the likelihood and severity of potential crashes on those curves.  

Common low-cost treatments include:  

• Speed advisory and warning signs;  

• Improving skid resistance;  

• Widening shoulders; and   

• Installing barriers.  

Reducing speed limits along routes can also improve safety, as this lowers the speed of vehicles entering out-of-context curves. 

Installing low-cost countermeasures at prioritized horizontal curves can be a very cost-effective approach to address safety issues and reducing roadway departure crashes. 

How are countermeasures funded? 

The Highway Safety Improvement Program (HSIP) is a core Federal-aid program with the purpose to achieve a significant reduction in traffic fatalities and serious injuries on all public roads. HSIP requires an evidence-based, data-driven, and strategic approach.  

Funding is enabled through the Infrastructure Investment and Jobs Act (IIJA) which provides USD $1.2 trillion funding.  The Highway Safety Improvement Program (HSIP) is part of the IIJA and provides USD $3.1bn p.a.; 2022-2026 (AASHTO, 2021).  States can receive funding for improvement projects from HSIP, where the improvements are specified in a State’s Strategic Highway Safety Plan (SHSP).  SHSPs must be evidence based and data driven. 

What is Abley SafeCurves? 

Abley SafeCurves^TM is a suite of products designed to help road managers and engineers quickly and authoritatively Identify, Prioritize and deliver Interventions on horizontal curves across highway and rural road networks to save lives.    

SafeCurves: Identify: 

• assesses the degree to which each curve is ‘out-of-context’ considering the geometric road information coupled with the operating speeds of vehicles when approaching the curve.  

• is aligned with the Safe System approach promoted by FHWA. 

• has validated outputs measured against real world crash data demonstrating the proactive, systemic and data led methodology to identify road safety risk.     

• is provided as a data as a service (DaaS) application making it flexible, accessible, and immediately available. 

How does Abley SafeCurves: Identify work? 

The methods underpinning SafeCurves: Identify are outlined in the Abley white paper. 

References

AASHTO (2021). Analysis of Infrastructure Investment and Jobs Act. American Association of State Highway and Transportation Officials, Washington, DC, USA. Available at: https://transportation.org/policy/wp-content/uploads/sites/56/2023/04/2021-09-15-AASHTO-Comprehensive-Analysis-of-IIJA-FINAL-v51.pdf Accessed 30 January 2024.  

Austroads (2015). Road geometry study for improved rural safety, by Jurewicz, C., Aumann, P., Bradshaw, C., Beesley, R., Lim, A. Austroads, Sydney, NSW. 

Austroads (2021). Guide to Road Design Part 3: Geometric Design. Sydney, NSW, Australia.  

Cardoso, J. L. (2005). Safety assessment for design and redesign of horizontal curves. International symposium on highway geometric design, 3rd, Chicago, Illinois, Transportation Research Board, Washington, DC, USA, 20 pp. 

Federal Highway Administration (2023). Horizontal Curve Safety. Available at: https://highways.dot.gov/safety/rwd/keep-vehicles-road/horizontal-curve-safety Accessed 21 December 2023.  

Institute for Health Metrics and Evaluation, Global Burden of Disease (2019) – processed by Our World in Data. Available at: https://ourworldindata.org/grapher/death-rates-road-incidents Accessed 7 February, 2024.   

National Center for Statistics and Analysis. (2023). Early estimate of motor vehicle traffic fatalities in 2022 (Crash Stats Brief Statistical Summary. Report No. DOT HS 813 428). National Highway Traffic Safety Administration. Washington, DC. 

U.S. Department of Transportation Federal Highway Administration (2011). Procedures for Setting Advisory Speeds on Curves. Washington, DC.  

Footnotes

¹Extracted from the US Department of Transport National Highway Traffic Safety Administration Fatality Analysis Reporting System (FARS) 2010 - 2021, non-intersection crashes, excludes functional Class 1 roads https://www.nhtsa.gov/research-data/fatality-analysis-reporting-system-fars

²“FHWA defines a roadway departure (RwD) crash as a crash which occurs after a vehicle crosses an edge line or a center line, or otherwise leaves the traveled way. Another term our partners often use is lane departure, which is synonymous with RwD, since both include head-on collisions when a vehicle enters an opposing lane of traffic.” https://highways.dot.gov/safety/RwD

³Brian Tegtmeyer, USDOT, NHTSA National 911 Program, TRB 2^nd TRB International Conference on Roadside Safety, June 2024