Skip to contentUnited States Department of Transportation - Federal Highway Administration FHWA Home
Research Home
Public Roads
Featuring developments in Federal highway policies, programs, and research and technology.
This magazine is an archived publication and may contain dated technical, contact, and link information.
Federal Highway Administration > Publications > Public Roads > Vol. 72 · No. 6 > Using CRFs To Improve Highway Safety

May/Jun 2009
Vol. 72 · No. 6

Publication Number: FHWA-HRT-09-004

Using CRFs To Improve Highway Safety

by Frank Gross and Karen Yunk

Crash reduction factors help identify the countermeasures with the most potential to save lives.

Alabama widened the outside shoulder of this horizontal curve in Vinemont, added a guardrail, and installed rumble strips to enhance safety. CRFs can help decisionmakers identify the expected benefits of these countermeasures.
Alabama widened the outside shoulder of this horizontal curve in Vinemont, added a guardrail, and installed rumble strips to enhance safety. CRFs can help decisionmakers identify the expected benefits of these countermeasures.

From 1993 through 2007, vehicle crashes on the Nation's roads claimed more than 40,000 lives annually. Despite efforts by Federal, State, and local transportation agencies, annual roadway fatality numbers — although decreasing gradually — remain stubbornly high.

In 2005, the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (SAFETEA-LU) established the Highway Safety Improvement Program (HSIP) as a core Federal Highway Administration (FHWA) program. The HSIP is structured to help transportation agencies at all levels reduce highway-related fatalities and serious injuries by promoting a strategic, data-driven process to identify, implement, and evaluate cost-effective roadway safety improvements.

For example, if a signalized intersection is experiencing a high number of left-turn crashes, which countermeasures do the data indicate should be considered? What is the likely improvement in safety for stabilizing a roadway shoulder? What change in the total number of crashes could be expected following implementation of various countermeasures to prevent roadway departures? Traffic engineers and other transportation professionals can find answers to these questions and more using crash reduction factors (CRFs).

A CRF is the percentage crash reduction that might be expected after implementing a given countermeasure at a specific site. For example, an agency installing centerline rumble strips on a two-lane road can expect, on average, a 14 percent reduction in all crashes and a 55 percent reduction in head-on crashes, though engineers still need to apply sound judgment and consider site-specific environmental, traffic, geometric, and operational conditions that will affect the safety impact of a countermeasure. CRFs, therefore, support the data-driven process required by SAFETEA-LU by providing quantitative measures for estimating the safety effectiveness of potential roadway improvements.

FHWA and its partners have developed a variety of resources to promote use of CRFs in the transportation community. The resources include a desktop reference for using CRFs; three documents containing countermeasures targeting intersection, pedestrian, and roadway departure crashes; a CRF clearinghouse (in development); decisionmaking tools that use CRFs; and two training courses. These tools are discussed below, followed by an example of how one State is using CRFs to support its HSIP.

Online CRF Resources

With the click of a mouse, transportation professionals can access a comprehensive list of CRFs in the FHWA report Desktop Reference for Crash Reduction Factors (FHWA-SA-08-011). The report documents estimated crash reductions that might be expected when specific countermeasures or groups of countermeasures are implemented to prevent crashes involving intersections, roadway departures, and pedestrians. The Desktop Reference is a compilation of all CRF information available to date. Where applicable, it includes multiple CRFs for the same countermeasure, which enables engineers to review the range of potential effectiveness. The document lists CRFs not only for total crashes but also for specific crash types, such as head-on, rear-end, and angle, and specific crash severities, such as fatal, injury, and property damage only, to help identify the differential effects of countermeasures.

This narrow, two-lane, rural road carries substantial truck traffic, as shown. Installing a shoulder could help enhance safety, and CRFs can help quantify the potential crash reduction benefit.
This narrow, two-lane, rural road carries substantial truck traffic, as shown. Installing a shoulder could help enhance safety, and CRFs can help quantify the potential crash reduction benefit.

FHWA also developed four issue briefs that summarize the best available information on countermeasure effectiveness for intersection, pedestrian, and roadway departure crashes: Traffic Signals, Toolbox of Countermeasures and Their Potential Effectiveness for Intersection Crashes, Toolbox of Countermeasures and Their Potential Effectiveness for Pedestrian Crashes (FHWA-SA-07-014), and Toolbox of Countermeasures and Their Potential Effectiveness for Roadway Departure Crashes (FHWA-SA-07-013). All the documents are available online at http://safety.fhwa.dot.gov/tools/crf.

In addition, FHWA initiated development of an online clearinghouse to serve as a repository for CRFs and their supporting documentation. A search tool will help site users find the appropriate CRF to meet their needs. Transportation professionals will be able to submit their own CRFs to be included in the clearinghouse. FHWA expects the clearinghouse to be available in summer 2009 and updated regularly to reflect the ever-increasing knowledge base of published CRFs.

Decisionmaking Tools

The National Cooperative Highway Research Program (NCHRP) is undertaking a project (NCHRP 17-36) to produce the first edition of the Highway Safety Manual (HSM), expected to be available by the end of 2009, which will prominently feature CRFs. The HSM will contain the latest information and tools to facilitate roadway planning, design, operations, and maintenance decisions based on explicit consideration of the safety consequences. The HSM will feature a synthesis of validated highway research, procedures that are adapted and integrated into practice, and analytical tools for predicting impacts on road safety.

This two-lane, rural road in Alabama incorporates wide shoulders, shoulder rumble strips, a guardrail, and a relatively wide clear zone (the total roadside border area, starting at the edge of the traveled way, that is available for an errant driver to stop or regain control of a vehicle). Designers can use CRFs to help them understand the safety impacts of these design decisions.
This two-lane, rural road in Alabama incorporates wide shoulders, shoulder rumble strips, a guardrail, and a relatively wide clear zone (the total roadside border area, starting at the edge of the traveled way, that is available for an errant driver to stop or regain control of a vehicle). Designers can use CRFs to help them understand the safety impacts of these design decisions.

The HSM will use the phrases "accident modification factors" or "accident modification functions" (AMFs) when referring to CRFs. The main difference between a CRF and an AMF is that CRFs provide an estimate of the percentage reduction in crashes, while AMFs are a multiplicative factor used to compute the expected number of crashes after implementing a given improvement. Also, the AMFs included in the HSM were "filtered" from the available literature to include only information that is deemed reliable based on accuracy, precision, and stability.

Mathematically stated, an AMF = 1 - (CRF/100). Although CRFs and AMFs are simply different conventions for expressing safety effectiveness, CRFs and accident modification factors are constants; accident modification functions allow the factor to vary for different scenarios, such as for different traffic volume scenarios.

Two key safety analysis tools that are already available use CRFs and support implementation of the methodologies and procedures presented in the HSM. The tools help transportation professionals better understand the safety implications of their designs and decisions. The first is SafetyAnalyst, state-of-the-art software designed to help highway agency decisionmakers develop a systemwide program for site-specific safety improvement projects. The other is the Interactive Highway Safety Design Model (IHSDM), a decision-support tool designed to help program managers, engineers, and other safety decisionmakers evaluate the operational effects of geometric design decisions. (See www.safetyanalyst.org and www.fhwa.dot.gov/research/tfhrc/projects/safety/comprehensive/ihsdm/index.cfm for more information.)

Each tool includes some form of CRFs; however, using the tools to their full potential requires an understanding of the fundamental concepts behind the CRFs. Toward that end, FHWA has developed two training courses on understanding and applying CRFs.

Training From NHI

Two Web-based training courses dealing with CRFs are available through the National Highway Institute (NHI): Application of Crash Reduction Factors (FHWA-NHI-380093) and Science of Crash Reduction Factors (FHWA-NHI-380094). Together, they respond to a clear need for formal training and provide basic knowledge on CRFs.

Crews installed chevron signs to enhance warning along this sharp horizontal curve with narrow shoulders. But how effective are the signs? How cost effective would it be to install a shoulder along the outside of the curve? Using CRFs, engineers can quantify the anticipated safety effects of geometric design improvements.
Crews installed chevron signs to enhance warning along this sharp horizontal curve with narrow shoulders. But how effective are the signs? How cost effective would it be to install a shoulder along the outside of the curve? Using CRFs, engineers can quantify the anticipated safety effects of geometric design improvements.

The Application of Crash Reduction Factors course helps safety professionals select the appropriate solutions to specific highway safety problems. Content focuses on CRF background, terminology, and components and how to diagnose safety issues, identify and select potential countermeasures, and compare their anticipated effectiveness. Instructors demonstrate critical points through case studies targeting intersection, pedestrian, and roadway departure crashes.

The course Science of Crash Reduction Factors helps participants learn to critically assess the quality of CRFs by understanding how safety is measured and the statistics and methodologies that form the basis of CRFs. Participants leave the course with a checklist they can use to determine the quality of a CRF and its applicability to their particular situations.

Each course is approximately 2 hours long and blends self-study and online instructor-led portions, which promote frequent interaction among participants and instructors. Case studies and practical exercises help transportation professionals build skills and confidence in applying CRFs, determining the quality and applicability of a CRF, and selecting appropriate safety measures.

Both courses are designed to benefit those who select, plan, design, maintain, and research safety-related highway improvements. Upon successful completion of each course, including a passing score on the final exams, participants receive professional development credits in the form of continuing education units.

Participants can register individually for prescheduled offerings, or a State department of transportation (DOT) might decide to host one of the courses for its own personnel, similar to traditional classroom training sessions. This flexibility through NHI ensures that agencies at different levels of government (local, regional, or metropolitan area) and consultants can participate.

CRFs in Practice

NHI piloted both courses in fall 2008, and participants reported positive feedback on the content and delivery. G. Stuart Thompson, a highway safety engineer with the New Hampshire Department of Transportation (NHDOT), participated in both CRF pilot training sessions. He had become familiar with CRFs when he served as assistant director of the Utah Local Technical Assistance Program (LTAP).

Ohio installed left- and right-turn lanes at this unsignalized rural intersection. The sight distance is limited due to a vertical curve near the intersection, which might have resulted in rear-end crashes. Two new NHI courses explain how to diagnose safety issues like this and apply CRFs to explore the safety effects of potential countermeasures.
Ohio installed left- and right-turn lanes at this unsignalized rural intersection. The sight distance is limited due to a vertical curve near the intersection, which might have resulted in rear-end crashes. Two new NHI courses explain how to diagnose safety issues like this and apply CRFs to explore the safety effects of potential countermeasures.

"The Utah DOT was using CRFs, so we incorporated them into an LTAP training course we were developing," Thompson says. "Since joining NHDOT, I've used CRFs to develop our '5 Percent Reports' [SAFETEA-LU requires States to identify no less than 5 percent of locations exhibiting the most severe safety needs] and in conjunction with some road safety audits we conduct." Thompson says he believes that the more engineers work with CRFs, the more comfortable they become with the process of using them.

"The important thing I learned from the courses is not to piggyback CRFs, something I was unaware of," he says. Thompson underscores an important point: CRFs should not be combined without extreme caution.

If multiple countermeasures are implemented at one location, then common practice is to multiply the CRFs, or more specifically their alternative form, AMFs, to estimate the combined effect of the countermeasures. In fact, there is limited research documenting the combined effect of multiple countermeasures. Although implementing several countermeasures might be more effective than just one, it is unlikely the full effect of each countermeasure would be realized when they are implemented concurrently, particularly if the countermeasures are targeting the same crash type.

Because multiplying several CRFs is likely to overestimate the combined effect, FHWA recommends exercising engineering judgment when estimating the effectiveness of multiple countermeasures. A more appropriate method for estimating the combined effects of multiple countermeasures is to conduct a rigorous before and after evaluation of several locations where the specific combination was implemented.

New Jersey's Experience

The New Jersey Department of Transportation (NJDOT) has long recognized the importance of CRFs in conducting benefit/cost analyses for highway safety improvements. The NJDOT Crash Analysis and Safety Program Development Section, within the Bureau of Safety Programs, applies a data-driven decisionmaking process to select CRFs for proposed safety countermeasures.

Atlantic County, NJ, used CRFs to select a combination of safety improvements to reduce right-angle and samedirection crashes at the intersection of Tilton Road (County Route 563) and Wrangleboro Road (County Route 575)
Atlantic County, NJ, used CRFs to select a combination of safety improvements to reduce right-angle and samedirection crashes at the intersection (above) of Tilton Road (County Route 563) and Wrangleboro Road (County Route 575). The reconstructed intersection (below) includes left-turn lanes on all four approaches, bike lanes, and signal timing modifications.

The reconstructed intersection of Tilton Road (County Route 563) and Wrangleboro Road (County Route 575) in Atlantic County, NJ, shown here, includes left-turn lanes on all four approaches, bike lanes, and signal timing modifications.

For example, a project to improve the safety of an intersection typically involves a detailed review of individual crash types (rear-end, right-angle, left-turn, pedestrian, etc.), including a review of summary data on intersection crashes and individual police reports.

From this review, the analysis and project development team develops a benefit/cost worksheet. The worksheet helps engineers determine the expected cost-effectiveness of selected countermeasures derived from typical quick-fix treatments such as improved signing, striping, signal layout and operation, and minor geometric/curb and sidewalk work.

Engineers use CRFs to calculate the net benefit included in the benefit/cost ratio. (Note that some treatments might reduce specific crash types or severities while increasing others. A classic example is installing a traffic signal, which might reduce right-angle crashes but increase rear-end crashes.) The benefit/cost ratio helps justify and prioritize funding for safety projects as part of New Jersey's HSIP.

According to Kevin Conover, the project engineer who leads the Crash Analysis and Safety Program Development Section, NJDOT engineers research historical data and review available technical briefing sheets, including CRF data from other States. As a result of these efforts, Conover says, NJDOT engineers are gaining confidence in identifying and applying CRFs.

Evaluating completed projects and achieving or exceeding expected reductions in the frequency and severity of targeted crash types also helps boost confidence in the effectiveness of CRFs. However, the problem remains that some crash countermeasures do not have CRF data, or historical evidence for their application in combination with multiple countermeasures is lacking.

"Using CRFs helps us determine whether a small-scale safety project will be cost effective and helps us elevate the priority of larger scale capital projects," Conover says. "Through postproject implementation analysis, we also will find out if the safety factor expected was ultimately realized. That is one of our checks for success, but the bottom line is data. Every successful project begins by studying the data to help us understand what the numbers really mean. We have to distinguish between the real and perceived safety problems and allow for a certain amount of random, incidental events and driver error."

This narrow, two-lane, rural road has centerline pavement markings, but no edgeline markings and a limited shoulder. As explained in the NHI courses, CRFs can help engineers identify the potential benefits of pavement markings and wider shoulders and apply CRFs to compare alternative treatments.
This narrow, two-lane, rural road has centerline pavement markings, but no edgeline markings and a limited shoulder. As explained in the NHI courses, CRFs can help engineers identify the potential benefits of pavement markings and wider shoulders and apply CRFs to compare alternative treatments.

What Was Old Is New Again

As many transportation safety professionals may know, CRFs are not a recent innovation. States have included safety analysis and cost information factors in project decisions for years. CRF resources, such as the Desktop Reference, Toolbox of Countermeasures documents, and the forthcoming clearinghouse, consolidate the latest CRF-related information into one place. Engineers and other project personnel no longer need to spend time searching for CRF data in a variety of publications housed in disparate offices; the information now is readily available in a central location on the Web.

Rudolph Umbs, a senior highway safety engineer at the FHWA Resource Center, is a longtime trainer and advocate for innovative highway safety practices. He participated in both online CRF pilot sessions for the NHI courses. "Whether attendees are experienced FHWA safety professionals or new to engineering, the CRF courses provide a refresher to the concept of crash reduction factors and how to use them," Umbs says. "The courses are a way for highway professionals to stay connected and have a better perspective on how they can use limited safety funds to their highest effect."


Frank Gross, Ph.D., P.E., is a highway safety engineer at Vanasse Hangen Brustlin, Inc. (VHB). He has more than 7 years of experience in transportation research and engineering and as a highway safety researcher. He has experience in highway design, traffic operations, and construction inspection. As a safety researcher, Gross specializes in highway safety evaluations, data analysis, and road safety audits. He was involved with a Transportation Research Board task force to develop core competencies for the highway safety workforce. He earned a Ph.D. in civil engineering from The Pennsylvania State University, where he specialized in transportation safety. He also earned a graduate minor in statistics.

Karen Yunk, P.E., is a transportation specialist with the FHWA Office of Safety. Her primary duties involve promoting decisionmaking tools that support HSIP-related activities. Before joining the Office of Safety, Yunk served as the traffic operations and safety engineer in the FHWA New Jersey Division Office. In this capacity, she managed the Federal safety program and provided technical expertise on a variety of transportation safety-related topics. Previously, she worked as a transportation planning and traffic engineering consultant. Yunk holds both bachelor's and master's degrees in civil engineering from Rutgers, The State University of New Jersey. She is a registered professional engineer in New Jersey.

For course information, or to register for the CRF courses Application of Crash Reduction Factors and Science of Crash Reduction Factors, visit the NHI course catalog at www.nhi.fhwa.dot.gov. For information regarding CRF resources, contact Karen Yunk at 609-637-4207 or karen.yunk@dot.gov.

ResearchFHWA
FHWA
United States Department of Transportation - Federal Highway Administration