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 / Road Safety Audits (RSA) / Road Safety Audits (RSA)

Road Safety Audits (RSA)


Road Safety Audits Logo

< Previous Table of Content Next >

The Effectiveness Of RSA Projects

Background

The RSA project costs and benefits must be understood to determine the effectiveness of an RSA in improving safety. This section discusses basic techniques to compare RSA project costs and benefits to determine a project B/C ratio and provides programmatic guidance when selecting a location to conduct an RSA.

RSA Costs

Three main factors contribute to the cost of an RSA:

The RSA team costs reflect the size of the team and the time required for the RSA, which in turn are dependent on the complexity of the RSA project. In the RSA projects evaluated as part of this project, the team size may have been larger than necessary to expose more professionals to the RSA process.

The design team and owner costs reflect the time required for staff to attend the start-up and preliminary findings meetings and to subsequently read the RSA report and respond to its findings. In addition, staff time is required to compile project or site materials for the RSA team.

The final cost component is implementation costs. These are the costs resulting from the design and construction of measures suggested as part of the RSA. These can range from low-cost measures (e.g., signing and pavement markings) to higher-cost measures that require design and construction. Measures suggested for each of the RSAs reviewed for this project are included in Appendix B.

RSA Benefits

The primary benefit of RSAs is the reduction in crashes as road safety is improved. The U.S. Department of Transportation's (1) estimated costs of automotive crashes are as follows:

The process for determining whether a measure or project is effective at reducing crashes is a B/C analysis that compares the total expected benefits of a measure or project to the total expected costs for a specific measure or project. The output of the analysis, called a B/C ratio, is the cost savings from the reduction in crashes (achieved through the safety measures) divided by the cost of the safety reviews and safety measures implemented (in dollars). A B/C ratio greater than 1.0 represents the project benefits outweighing the project costs. Table 5 illustrates how the cost of various types of safety reviews can be justified for an existing location based on crash history. The table indicates the number of crashes of moderate-(B2), minor-(C), or PDO-(O) severity that need to be reduced as a result of the safety review to achieve a B/C ratio of at least 1.0 based on the average cost of the safety review alone. It is assumed that each type of safety review has a service life of three years. As such, Table 5 refers to the number of crashes in a three-year period. The table highlights that, by addressing a few moderate to property-damage-only crashes, an agency can justify the cost of a simple to complex RSA.

Table 5. Type of Safety Review Based on Number and Severity of Crashes Addressed.

Type of Safety Review
(Existing Roadway)
Average Cost B2 C O
Moderate Minor PDO
Associate Crash Cost
$89,000 $11,600 $6,500
Traditional Safety Review $1,000 1 1 1
RSA – Simple $10,000 1 1 2
RSA – Intermediate $15,000 1 2 3
RSA – Complex $25,000 1 3 4
Detailed Safety Review $50,000 1 5 8
Notes: Any type of safety review can achieve a B/C over 1.0 if there is just one moderate or more severe injury. Each study type is assumed to have a three-year service life. A traditional safety review is typically a quick review of a project conducted by a transportation agency with in-house staff. A detailed safety review refers to a more expansive form of traditional safety review.

In the past, it was difficult to quantify the benefits of design-stage RSAs, especially on new facilities that have no crash record, since they aim to prevent crashes from occurring. Now, through the use of the Highway Safety Manual (HSM) and related tools such as the Interactive Highway Safety Design Model (IHSDM), there are methods to estimate the expected crashes for a new facility based on its design and operational characteristics. The benefits of RSAs on improved facilities and existing roads can be quantified based on observed pre – and post-improvement collision histories, but there are potential biases related to simple before-after comparisons. The HSM identifies methods for incorporating the average crash history at similar sites with the observed crashes at the location of interest to address the potential biases associated with before-after evaluations.

Whether design-stage or existing-stage, when compared to the high cost of automotive crashes, changes implemented from an RSA only need to prevent a few moderate – or high-severity crashes to be cost effective. Table 6 illustrates how different project implementation levels can achieve a favorable B/C ratio given different crash severities. The table indicates the number of crashes (by severity) that need to be reduced over the service life of the project to achieve a B/C ratio of at least 1.0 based on project implementation cost. It is assumed that maintenance projects have a service life of three years and HSIP projects have a service life of 10 years. For example, a low-level maintenance project would prove to be beneficial given at least one PDO crash is eliminated. A medium-level HSIP candidate project would prove to be beneficial if at least one serious injury or three moderate injury crashes were eliminated. Capital projects addressing highway safety issues without the guidance of an RSA may be similarly beneficial, but RSAs provide an opportunity to address multiple issues and coordinate safety-related projects to optimize costs and improve the potential effectiveness in reducing crashes and severity.

There are many other benefits that can be realized from conducting an RSA or safety review other than relying completely on achieving a B/C ratio greater than 1.0. Other benefits of RSAs include reduced life-cycle project costs (i.e., the total cost over the life of a project that includes activities like design and engineering, land acquisition, construction, reconstruction and rehabilitation, and preservation and routine maintenance) (2). This can be attributed to the fact that, as crashes are reduced and there is less need to make changes to the roadway or project, a greater awareness of safety is realized through the development of good safety engineering and design practices within an agency, including consideration of potential multimodal safety issues and integrating human factors into the design, operations, and maintenance of roads.

Table 6. Type of Project Based on Number of Crashes.

Project Type Average Cost K A1 A2 B1 B2 C O
Killed Critical Severe Serious Moderate Minor PDO
Associate Crash Cost
$6,000,000 $4,422,500 $1,087,500 $333,500 $89,900 $11,600 $6,500
Maintenance – Low1 $1,000 1* 1* 1* 1* 1* 1* 1*
Maintenance – Medium1 $10,000 1* 1* 1* 1* 1* 1* 2**
Low-Cost Safety Improvement2 $10,000 1* 1* 1* 1* 1* 1* 2**
Maintenance – High1 $100,000 1* 1* 1* 1* 2** 9^ 16^
HSIP – Medium2 $250,000 1* 1* 1* 1* 3** 22† 39†
HSIP – Med/High2 $500,000 1* 1* 1* 2** 6^ 44† 77†
HSIP – High2 $1,000,000 1* 1* 1* 3** 12^ 87† 154†
* Strong Candidate for RSA and Countermeasures
** Likely Candidate for RSA and Countermeasures
^ Questionable Candidate for RSA and Countermeasures
† Weak Candidate for RSA and Countermeasures
Notes:
1 Assumes a 3-year service life.
2 Assumes a 10-year service life.
Through safety management activities, one should strive for the greatest concentration of severe crashes to target for an effective RSA.
The number of crashes shown is the absolute minimum to achieve a favorable benefit/cost ratio (i.e., greater than 1.0).

RSA Projects Evaluated

Five RSAs were selected for detailed quantitative evaluations as part of this project, based on having implemented safety measures suggested as part of the RSA, as well as the availability of data for similar "comparison" sites within the agency's jurisdiction. The RSA sites selected for detailed evaluation include the following:

The information presented for each RSA was gathered through direct communication with the responsible RSA agency. The specific objective of the quantitative analysis was to estimate the safety effects of engineering improvements that were implemented as a result of conducting RSAs. The expected annual benefits were then compared to the annualized cost of the treatments, including the cost to conduct the RSA, to estimate a B/C ratio. Safety effects were measured with respect to total crashes. Specific crash types were considered whenever possible, based on the individual locations, availability of detailed data, and sample size.

Methodology

Three evaluation methods were employed in this study depending on the availability of data. All three methods were based on an observational before-after study design where a site was selected for study, an RSA was conducted, treatments were implemented based on the RSA findings, and the crash history of the site was compared before and after treatment.

There are several potential biases to account for in before-after studies for which one should account:

If these potential biases are not properly addressed (or at least dismissed), the results could over – or underestimate the safety effects of the treatment.

The following is a brief summary of primary attributes of each evaluation methodology used in the evaluation of RSA outcomes:

A detailed overview of the attributes, advantages, and disadvantages of each evaluation method can be found in Appendix A. More details can be found in FHWA's A Guide to Developing Quality Crash Modification Factors (Gross et al., 2010).

Key Findings

Each of the five RSA sites evaluated exhibited an overall reduction in crashes as a result of the suggestions implemented from the RSAs. Reductions in total crashes ranged from 10 percent to 50 percent, with varying levels of statistical significance. Furthermore, each of the RSA sites for which project cost information was provided showed a B/C ratio greater than 1.0, meaning that the benefits of the treatments implemented as a result of the RSAs exceeded the total RSA and implementation costs. B/C ratios ranged from 1.2:1 to 116:1; the greater the B/C ratio, the greater the benefit compared to the project cost. Areas with high B/C ratios either had a high reduction in crashes (e.g., a reduction in total crashes of 50 percent) or had low project costs (e.g., implementing signing and pavement markings). Areas with lower B/C ratios typically had higher project costs.

Table 7 summarizes the safety improvement measures implemented as a result of the RSAs, the statistical evaluation methods applied, the project costs, and the safety benefits and resulting B/C ratios based on the analysis of the crash types addressed. Appendix A provides further details on the methodologies used to quantify RSA treatment benefits for each case study site. Further details about each RSA project site (i.e., a project description, summary of key findings and suggestions, photographs for each location, the methodology used in the evaluation, data, analysis, and findings) are included in Appendix B. If it was available, additional information is also provided about the agency's RSA procedures and experience outside of the specific RSA project considered for this effort.

Table 7. Quantitative Evaluation of RSA Outcomes.

Location RSA Stage Results
Bullhead Parkway
Bullhead City, Arizona
  • Roadside improvements
  • Guardrail
  • Signing and marking
  • Traffic signals
  • Drainage
Comparison Group: Traffic volume changes, temporal changes
  • Total cost of $869,478 (plus several nominal fee projects)
  • $785,000 of additional projects scheduled
  • 54% reduction in total crashes
  • 50% reduction in fatal/incapacitating injury crashes
  • 30% reduction in intersection-related crashes
  • B/C ratio of 20:1 (total crashes)
  • B/C ratio of 16:1 (injury crashes)
  • B/C ratio of 2:1 (intersection – related crashes)
State Route 101 (Peavine Road)
Cumberland County, Tennessee
  • Signing and marking
  • Sight distance enhancements
  • Vegetation removal
Empirical Bayes: Regression-to-the-mean, traffic volume changes, temporal changes
  • Total cost of $35,000
  • 13.7% reduction in total crashes
  • 31.3% reduction in injury crashes
  • B/C ratio of 51:1 (total crashes)
  • B/C ratio of 116:1 (injury crashes)
Intersection of Collier Boulevard and Golden Gate Parkway
Collier County, Florida
  • Signing and marking
  • Geometric modifications
  • Vegetation removal
  • Pedestrian enhancements
Empirical Bayes: Regression-to-the-mean, traffic volume changes, temporal changes
  • Total cost of $265,000
  • 11% reduction in total crashes, both intersections
  • B/C ratio of 5:1 (total crashes, both intersections)
Immokalee Road
Collier County, Florida
  • Access management
  • Pedestrian signals
  • Median fence
  • Geometric modifications
  • Bicycle facilities
Naïve (with adjustments): Traffic volume changes, temporal changes, other changes (4 – to 6-lane conversion)
  • 10.8% reduction in total crashes
Ninth Street
Ocean City, New Jersey
  • Traffic signal improvements
  • Pedestrian signals
Comparison Group: Traffic volume changes, temporal changes
  • Total cost of $921,000
  • 25.6% reduction in total crashes
  • B/C ratio of 1.2:1 (total crashes)

 

< Previous Table of Content Next >
Page last modified on January 29, 2015
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