Evaluation of Safety Strategies at Signalized Intersections (2011)

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.

5Background In 1997, the American Association of State and Highway Transportation Officials (AASHTO) Standing Committee for Highway Safety along with Federal Highway Administration (FHWA), National Highway Traffic Safety Administration (NHTSA), and the Transportation Research Board (TRB) Committee on Transportation Safety Management convened a meeting of national experts in the highway safety area to develop a Strategic Highway Safety Plan. This plan focuses on 22 highway safety challenges or emphasis areas that have an impact on highway safety. To advance the implementation of countermeasures to reduce accidents and injuries, NCHRP Project 17-18(3) began the development of a series of imple- mentation guides which were subsequently published in the form of several volumes of NCHRP Report 500. Each guide addresses one of the 22 emphasis areas, and includes an intro- duction to the problem, a list of objectives for improving safety in that emphasis area, and strategies for each objective. Each strategy is designated as: proven, tried, or experimental. Many of the strategies discussed in these guides have not been rigorously evaluated. FHWA has initiated a Low Cost Safety Improvements Pooled Funds study to evaluate some of these strategies. The first two phases have been completed and included the evaluation of the following strategies: (1) cen- ter TWLTLs for two-lane roads, (2) higher retro-reflectivity sheeting for STOP signs, (3) pavement markings noting ‘stop ahead,’ (4) flashing beacons at stop controlled intersections, (5) the trade-off between lane and shoulder width given a pavement width, (6) advance street name signs, (7) curve delineation, and (8) offset left-turn lanes at signalized inter- sections. These evaluations were conducted using before-after data from locations where the safety improvements have been made and a reference group of untreated locations. As a follow-up to these FHWA evaluations, NCHRP initiated NCHRP Project 17-35. The focus of this project is to select and evaluate strategies from NCHRP Report 500, Volume 12: A Guide for Reducing Collisions at Signalized Intersections (Antonucci et al., 2004). The desired result would be a set of Crash Modification Factors (CMFs) that specify the ratio of the expected crash frequency after and before the implemen- tation of a treatment. It is expected that the CMFs developed would augment the CMFs currently in the Highway Safety Manual. Study Objectives and Overview The objective of this project was to evaluate the safety effec- tiveness for selected strategies identified in NCHRP Report 500, Volume 12. The intent is to develop reliable CMFs. At a min- imum, for CMFs to be reliable they must meet the following criteria: 1. CMFs should be methodologically and statistically valid. Many existing CMFs are derived from before-after analysis of actual countermeasure implementation. Indeed, such before-after studies, as opposed to cross-sectional/ regression-type analysis, will produce the best CMF esti- mates, but only if conducted properly to account for the regression to the mean effects at sites selected for treatment because of unusually high accident frequencies. Unfortu- nately, much of the available knowledge on CMFs may be tainted by this problem because this selection bias is quite prevalent. Other methodological problems that have affected the reliability of currently available CMFs include: – Failure to properly separate out the safety effects of other changes (e.g., traffic volumes, the impacts of other mea- sures, crash reporting, underlying trends across time). – Sample sizes that are too small—large numbers of sites with the same combination of applied counter- measures are needed for a valid analysis. For some treat- ments that are expected to affect a low proportion of the total crashes at a site (e.g., pedestrian treatments), hundreds of locations may be necessary along with many years of crash data. C H A P T E R 1 Introduction

– Use of comparison groups that are unsuitable for a variety of reasons, including the fact that sites may have been affected by the treatment. – Incorrect interpretation of accuracy of estimates or pre- senting results without statements of accuracy. – For many treatments, there may be different effects at different sites, so a single CMF that is typically estimated is often not applicable. 2. The CMFs should represent the different crash categories that reflect the impact of the improvement. Crash cate- gories might include total crashes, severe injury crashes, property damage only crashes, and specific crash types (such as rear end and angle). 3. The variability in CMFs should be stated. The best estimate of the CMFs, along with some technique that reflects their variability (such as ranges, confidence intervals, standard deviation, or some other technique) should be presented. This will facilitate not only the application of the CMF but also the amalgamation with CMF results from other evaluation studies. 4. The CMF should reflect the savings in “total harm” that the treatment provides. Many treatments affect both crash frequency and crash severity, some just severity, and some tradeoff crashes of different severities (e.g., traffic signaliza- tion can decrease more-severe angle crashes but increase less-severe rear-end crashes). CMFs must capture changes in severity as well as frequency in order to measure “harm savings.” The identification and development of CMFs that meet most of these requirements involved a study effort with the following tasks: • Task 1—Literature Review • Task 2—Conduct a Survey of State and Local Agencies • Task 3—Develop a Work Plan • Task 4—Meet with the Panel • Task 5—Collect Data and Conduct Evaluation • Task 6—Develop a Final Report In Task 1, based on a critical review of published studies for each treatment/strategy, the research team assigned a level of predictive certainty for each available CMF. Four levels of predictive certainty were defined: High, Medium-High, Low-Medium, and Low. The literature review also covered knowledge about ongoing or planned research, which was based on a review of several research-in-progress databases, discussions with other highway safety researchers, and con- versations with research sponsors such as FHWA and the Insurance Institute for Highway Safety (IIHS). Details about the literature review are presented in Chapter 2. In Task 2, web-based surveys were conducted using a tool called Zoomerang. The survey was sent to two listservs, contacts in all 50 States, and several local agencies. The re- spondents were asked to indicate if they had implemented a particular strategy/treatment and the approximate number of installations for each strategy/treatment. In addition, they were also asked to rate the importance of knowing the CMF for each treatment. These ratings along with the information about the installations were used to develop a short list of treatments for further consideration. To get further informa- tion about the availability and suitability of the data for the short-listed strategies, selected agencies that responded to the original web-based survey were contacted by telephone. Further description of Task 2 can be found in Chapter 3. In Task 3, the research team developed a recommended and prioritized list of strategies to be evaluated to provide the best use of available funds, and a work plan for evaluating these strategies. Task 4 involved a meeting with the NCHRP panel to discuss the work plan and develop the list of strategies to be evaluated in Task 5. Further description about Tasks 3 and 4 can be found in Chapter 4. Task 5 involved an evaluation of the strategies. The results from this evaluation are provided in Chapter 5. Chapter 6 provides a summary page showing the recommended CMFs for each treatment that was evaluated in this study. Chapter 7 provides some general conclusions and directions for further research. 6