U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Safety

FHWA Home / Safety / Roadway Departure / Low-Cost Treatments for Horizontal Curve Safety 2016

Low-Cost Treatments for Horizontal Curve Safety 2016

CHAPTER 2. THE TWO COMPONENTS OF SAFETY IMPROVEMENT: SITE ANALYIS AND THE SYSTEMIC APPROACH

< Previous Table of Contents Next >

The most effective safety improvement process has two components:

  1. A site analysis component. Data analysis is used to identify locations where a clear safety problem exists. Treatment of these locations may include higher-cost strategies. This can be thought of as the reactive component of a safety program.
  2. A systemic component. Data analysis is used to identify risk factors associated with a particular type of severe crash and to identify locations at higher risk. Normally, lower-cost strategies are then deployed at a larger number of these high-risk locations. This can be thought of as the proactive component of a safety program.
Systemic Safety Improvement: An improvement widely implemented based on high-risk roadway features that are correlated with high severity crash types.
Systemic Safety Project Selection Tool, FHWA

The priority of a safety improvement program should be preventing fatal and serious injury crashes. In fact, the purpose of the Federal Highway Safety Improvement Program (HSIP) is stated in law as follows:

The purpose of the highway safety improvement program shall be to achieve a significant reduction in traffic fatalities and serious injuries on all public roads, including non-State-owned public roads and roads on tribal land.

How a safety program and data analyses are focused-severe crashes versus total crashes-influences the degree to which a particular safety problem is addressed with the systemic approach versus the more traditional site analysis approach. With a program where all crashes are used as the performance measure, high crash locations will be more prevalent and treatment strategies will tilt more heavily toward addressing high crash locations. In contrast, a program that uses severe crashes as the performance measure will use a stronger systemic component as severe crash locations are not as concentrated.

This is particularly true for severe roadway departure crashes, which tend to be highly scattered across the rural and local roads system (see Figure 3). This does not mean that severe crashes are random. They tend to be overrepresented at locations with high risk characteristics, horizontal curvature being one of those.

A safety improvement process should include both components: treating high severe crash locations where they exist as well as systemically addressing locations or segments at higher risk. Both components will provide optimal results with good data and data analysis. FHWA's Systemic Safety Project Selection Tool (2013) provides analytical techniques for determining a reasonable balance between the implementation of spot safety improvements and systemic safety improvements

Screenshot of the Minnesota Crash Mapping Analysis Tool. The screenshot shows a GIS map of Minnesota's southeastern District with I-35 and I-90 present on the map. Red dots on the map represent severe (K and A) roadway departure crashes at curves over a five year period.

Figure 3. Map. Severe (K and A) roadway departure crashes at curves over a five year period in Minnesota's southeastern District. The wide dispersion of crashes indicates that a traditional site analysis/spot location approach will not sufficiently address this type of severe crash. Source: Minnesota Crash Mapping Analysis Tool, MnDOT.

All Public Roads-Addressing the Local System

In addition to a focus on prevention of fatal and serious injury crashes, the most effective safety programs also consider all public roads, including those under local jurisdiction.

In most States, an examination of crash data demonstrates that focusing safety investment only on higher-level facilities such as the Interstate System and State highways will not sufficiently address the severe roadway departure crashes most prevalent on horizontal curves.

Importance of the local system:
From 2010–2012, approximately 39 percent of fatalities in the United States occurred on the local system of roads.
—Fatality Analysis Reporting System

The Fatality Analysis Reporting System (FARS) is a national database of fatal highway crashes in the United States. FARS does not break down highway fatalities by State versus local jurisdiction, but the "Route Signing" field can provide a useful approximation of the magnitude of the fatal crash problem that occurs on the local system of roads. Using this method, FARS data suggests that from 2010-2012 approximately 39 percent of fatalities in the United States occurred on the local system. There were 39 States with 30 percent or higher fatalities on local roads and streets.

To most effectively improve safety at horizontal curves, it is important to analyze data on both the State and local systems. Spot locations where severe crashes are concentrated are even less common on the local system, and data analysis of the complete roadway network will add further support for including a strong systemic component.

The Pennsylvania Department of Transportation (PennDOT) experienced a higher percentage of fatal curve-related crashes on rural roads due to the predominance of horizontal curves. To combat this, PennDOT identified priority curves by examining crash frequency, crash rate, and crash severity. PennDOT then systemically implemented curve improvements, such as oversized fluorescent yellow advanced curve warning signs, advanced curve pavement markings, correction of any shoulder drop offs within the curve, chevron delineation, and curve widening. A three-year before/after analysis of locations where a combination of these countermeasures were implemented between 2000 and 2008 resulted in the following:

  • 17-percent reduction in overall crashes.
  • 44-percent reduction in fatal crashes.
  • 40-percent reduction in major injury crashes.

See Appendix A for more information.

A Change in Mindset

For some agencies, improving safety at spot locations on State highways has been the traditional approach, making up the bulk of safety improvement projects. Shifting to a systemic approach to prevent severe roadway departure crashes at curves along all public roads may require a change in mindset.

Determining answers to the following questions through data analysis is an effective first step in the process:

  1. Are there a large number of severe (fatal and serious injury) crashes scattered widely across the system? If possible, plot them spatially as shown in Figure 3.
  2. Are the types of safety improvement projects funded in relative alignment with the findings of question 1? If mostly high-cost projects at high-crash locations are being implemented, does this approach align with how severe crashes are located and dispersed?
  3. What is the distribution of severe crashes on the State system as compared to local system? If there are severe crashes occurring on the local system, are safety funds made available to local agencies so that severe crashes are reduced on all public roads?
In 2010, the Minnesota Department of Transportation (MnDOT) began incorporating a systemic approach in all of Minnesota's counties after many years of exclusively using the traditional reactive approach to safety. MnDOT performed a network screening on their horizontal curves and found five risk-factors associated with high-crash curves. MnDOT addressed these curves with countermeasures such as center line rumble strips, advanced signing, 2-foot shoulder paving with Safety EdgeSM, and edge line rumble strips, 6-inch edge line, and most commonly, chevrons. MnDOT has seen a drop in roadway departure crashes because of these efforts. See Appendix B for more information.

Figures 4 through 7 are examples of low-cost systemic treatments that can be applied to a large number of high-risk curve locations.

Photograph showing a two-lane roadway in a rural area with three chevron signs installed along the outside of the curve.

Figure 4. Map. Severe (K and A) roadway departure crashes at curves over a five year period in Minnesota's southeastern District. The wide dispersion of crashes indicates that a traditional site analysis/spot location approach will not sufficiently address this type of severe crash. Source: Minnesota Crash Mapping Analysis Tool, MnDOT.

Close-up photograph of centerline rumble strips. It is a two-lane roadway located in a rural area. The rumble strips are milled into the surface underneath the centerline.

Figure 5. Map. Severe (K and A) roadway departure crashes at curves over a five year period in Minnesota's southeastern District. The wide dispersion of crashes indicates that a traditional site analysis/spot location approach will not sufficiently address this type of severe crash. Source: Minnesota Crash Mapping Analysis Tool, MnDOT.

Close-up photograph of a narrow paved shoulder (2 feet). It is a two-lane roadway in a rural area. On the shoulder of the curve, rumble strips are installed adjacent to the edgeline and safety edge is installed along the edge of the roadway surface.

Figure 6. Map. Severe (K and A) roadway departure crashes at curves over a five year period in Minnesota's southeastern District. The wide dispersion of crashes indicates that a traditional site analysis/spot location approach will not sufficiently address this type of severe crash. Source: Minnesota Crash Mapping Analysis Tool, MnDOT.

Close-up photograph of a 6-inch edge line along a curve separating the travel lane and the paved shoulder.

Figure 7. Photo. Delineation with enhanced (6-inch) edge lines.

Additional Resources

More information on the systemic approach to safety is available at the following resources:

< Previous Table of Contents Next >
Page last modified on April 12, 2016
Safe Roads for a Safer Future - Investment in roadway safety saves lives
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000