Crash Modification Factors in the Highway Safety Manual: A Review (2023)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1   Crash Modification Factors in the Highway Safety Manual: A Review Background and Objectives Highway safety practitioners were given a significant new tool in 2010 with the publi­ cation of the AASHTO Highway Safety Manual (HSM) (AASHTO 2010). In Part D of the HSM, crash modification factors (CMFs) are provided to estimate the safety effects for a variety of treatments or countermeasures. Some of these CMFs are also presented in Part C to com plement the predictive methodology in assessing the safety effects of design decisions. By the time the second edition of the HSM (HSM2) is published, many more important CMFs will have been developed and additional CMFs will be needed to support enhancements to the predictive methodologies—for example, to better predict crashes by type and severity. The objectives of this research were to • Assess the current criteria and existing process for evaluating the quality of CMFs and identifying CMFs for appropriate use with the HSM; • Develop proposed revisions to the criteria and process, including how existing and new CMFs may be incorporated in the HSM, and provide guidance for practitioner application of the revised process; • Apply the evaluation criteria to identify and assess CMFs and develop a list of appropriate CMFs for the HSM; • Conduct a CMF gap analysis to identify treatments for which quality CMFs are needed; • Estimate crash modification functions (CMFunctions) for select treatments for which detailed data on site characteristics were available from the original studies; • Estimate adjustment factors for safety performance functions (SPFs) that were estimated in NCHRP Project 17­62, “Improved Prediction Models for Crash Types and Crash Severities” (Ivan et al. 2021); • Calibrate the prediction models planned for inclusion in the HSM2 (also called single state calibration); • Validate the combination of SPFs estimated in NCHRP 17­62 (Ivan et al. 2021) with the SPF adjustment factors estimated in this effort; and • Develop CMFs for roadside crashes, building on results from NCHRP Project 17­54, “Consideration of Roadside Features in the Highway Safety Manual” (Carrigan and Ray 2018). Outcomes of the Research Effort The first part of NCHRP Project 17­72, “Update of Crash Modification Factors for the Highway Safety Manual,” involved conducting a review of procedures for assessing CMF quality. This included a review of factors used to describe CMF quality and a review of S U M M A R Y

2 Crash Modification Factors in the Highway Safety Manual: A Review techniques used to improve CMF quality. (See Appendices A and A.1 for the results from this review.) While conducting the review, the project team distributed a questionnaire to practitioners and conducted a focus group of experienced users. The questionnaire and the focus group addressed different topics, including selecting and applying CMFs; CMF quality; presentation of CMFs; and the content of the HSM2, including Part D (the summary of the results from the questionnaire and focus group is given in Appendix B). On the basis of the outcomes of the reviews, questionnaires, and focus group, the project team developed a revised rating system for CMFs that are estimated from before–after and cross­sectional studies. In the revised rating system, each CMF could receive a maximum of 150 points. This information was presented as part of the interim report and discussed at the interim meeting along with guidance on a threshold for HSM approval of CMFs. The project team then developed a rating system for meta­analysis and meta­regression studies. The new CMF rating system was discussed in different meetings and conferences. In 2020, the CMF rating system was reviewed and critiqued by a task force that included practitioners and researchers from AASHTO and the TRB Highway Safety Performance Committee. Following that, the project team developed a revised CMF star rating system for the CMF Clearinghouse that was based on the original CMF rating system developed for NCHRP 17­72. With this development, the task force also questioned the need for HSM inclusion criteria. Appendix C describes the CMF rating system. Appendix D provides a CMF rating summary and describes the contents of Appendix D.1, which provides the list of CMFs that have been rated by the NCHRP 17­72 rating system, and Appendix D.2, which provides the legacy CMF Clearinghouse star ratings along with the NCHRP 17­72 ratings for all CMFs. The next activity was to estimate CMFunctions. This was done to address the limited information on the variability in CMFs with application circumstance. CMFunctions were estimated for select treatments with the intent of choosing those treatments for which the original data were available. In the case of empirical Bayes (EB) before–after studies, the project team was looking for information about individual treatment sites that were used in the evaluation, including the observed number of crashes in the after period, the EB expected number of crashes in the after period had the treatment not been implemented, and the variance of the expected number of crashes in the after period. For cross­sectional studies, the project team was looking for the data that were used in estimating the CMFs. CMFunctions were estimated for improving curve delineation, modifying shoulder and median width on rural multilane roads, introducing a two­way left­turn lane, and conversion of two­way stop signs to four­way stop signs. Appendices E through H provide the details of the CMFunction development. NCHRP 17­72 was tasked with identifying adjustment factors for SPFs that were estimated in NCHRP 17­62. Highly rated CMFs from the CMF Clearinghouse were used to identity the SPF adjustment factors. Appendices I and J provide the adjustment factors that were identified. Before the prediction models were included in Part C of the HSM, Highway Safety Information System (HSIS) data from California were used to calibrate the models predict­ ing intersection crashes, and data from Washington State were used to calibrate the models predicting segment crashes. It was argued that the end user could employ these recalibrated models to directly compare the expected safety performance of different facility types. NCHRP 17­72 was tasked with recalibrating models that would be included in the HSM2. This exercise is also referred to as “common state calibration” or “single state calibration.” Data from Ohio were used for calibrating the prediction models for segments in rural roads. Data from North Carolina were used for calibrating the prediction models for seg­ ments in urban roads and intersections in rural and urban roads. Appendices K through M Appendices The appendices for this report are provided in NCHRP Web-Only Document 352: Crash Modification Factors in the Highway Safety Manual: Resources for Evaluation, which can be accessed at nap.nationalacademies.org by searching on “NCHRP Web-Only Document 352.”

Summary 3   provide the calibration factors and examine the validation of the SPF adjustment factors that were estimated to be used with the NCHRP 17­62 prediction models. NCHRP 17­72 included the task of developing a procedure for estimating a CMF for roadside design. Appendix N describes in detail the development of a procedure for esti­ mating a CMF that describes the safety effect of roadside design elements associated with a rural two­lane two­way highway. Appendix O describes the procedural steps for computing the proposed roadside design CMF and using this CMF within the predictive model frame­ work in HSM Chapter 10 to evaluate a road segment. A CMF gap analysis was also conducted under NCHRP 17­72 to identify the research needed to guide future CMF development efforts. The gap analysis, described in Appendix P, was based on a detailed review of the CMF Clearinghouse and an assessment of ongoing research efforts to develop CMFs. There is a need for explicit guidance on the proper selection and application of the CMFs to be included in Part D of the HSM. The intent is for this guidance to provide an overview of CMF selection, application, and development, with a focus on countermeasure­specific CMFs that are a key part of the processes and methods described in Part B of the HSM.