Analysis of Naturalistic Driving Study Data: Offset Left-Turn Lanes (2014)

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.

4Introduction The research team evaluated the effect of offset left-turn lanes on driver behavior using Naturalistic Driving Study (NDS) data recently collected by the second Strategic Highway Research Program (SHRP 2). The primary data elements were forward- and rear-facing video from left-turn maneuvers made by NDS participants. The video data were used to measure accepted and rejected gaps by study drivers making left turns through intersections with different left-turn offsets. The research was conducted in two phases. The overall objective of Phase 1 was to use a sample of NDS data collected during the first few months of the project to develop, test, and validate a data cap- ture and analysis plan. The objective of Phase 2 of the research was then to apply that analysis plan to the full NDS data set to answer the specified research question. This report provides a brief background of Phase 1 activi- ties but focuses primarily on the Phase 2 research efforts and results. In addition to the discussion of the technical activi- ties conducted to answer the research questions, this report describes the research team’s experience using the NDS data. Additional information can be found on the study’s InSight website (https://insight.shrp2nds.us/). Research Problem Statement The Safety area of the SHRP 2 program focuses on the need to reduce traffic injuries and fatalities by preventing or reduc- ing the severity of collisions. Every 1% reduction in crashes will prevent 330 deaths and about $2.7 billion annually in medical, property damage, legal, emergency medical services (EMS), and congestion costs, as well as lost workplace produc- tivity (Blincoe et al. 2014). Moreover, since crashes are a lead- ing cause of nonrecurring congestion, collision prevention has added benefits in terms of reduced delay, fuel consumption, and emissions. The SHRP 2 Safety program has identified intersection crashes as a priority area. The specific crash type being addressed in this research is angle collisions between left-turning vehicles and opposing through vehicles at intersections. Fatality Analy- sis Reporting System (FARS) data for 2009 show that left-turn collisions constitute 8.5% of all traffic fatalities. Sight-distance blockage by opposing left-turn vehicles, which can be remedied by offset left-turn lanes, is a likely contributing factor in many such crashes. Some substantial proportion of these fatalities occur because of the sight obstruction created by the presence of opposing left-turn vehicles at intersections with negative offset between opposing left-turn lanes. A study by Persaud et al. (2009) suggests (but does not demonstrate with statistical significance) that small corrections in offset may have limited or no effect on crashes, but that provision of substantial posi- tive offset may be effective in reducing left-turn crashes with opposing vehicles. Research regarding the safety effectiveness of offset left-turn lanes is important because current geometric design policy in the AASHTO Green Book recommends offsetting left-turn lanes for divided roadways with medians wider than 18 ft, with- out any quantitative information on how much offset should be provided (AASHTO 2011). Thus, highway designers are told that offset left-turn lanes are desirable but are given no specific or quantitative guidance on when offset left-turn lanes should be used or how much offset is needed. This lack of guidance is reflected in the range of project types evaluated in the Persaud et al. study. Each project began with an intersection in which the opposing left-turn lanes had a negative offset. In some proj- ects, the improved left-turn lanes had a less negative offset, in some projects they had zero offset, and in others they had a positive offset. (See Figure 2.2 for an illustration of positive-, negative-, and zero-offset left-turn lanes.) Thus, it is clear that the highway agencies that designed these projects lack guid- ance on how best to design such locations. The Persaud et al. study is well done, but—because of the inherent limitations of crash studies—does not provide the type of specific design guidance that is needed. C h a P t e R 1

5Research Objectives and Scope The objective of this research was to use the NDS data to address safety issues using analysis methods that were not pos- sible or were not feasible before this data set became available. The results of the studies performed within this project were intended to lead to real-world applications that result in mea- surable safety benefits; to be applicable to a variety of road- ways, vehicles, and drivers in the United States; and to provide insight to safety issues that could not be obtained through use of traditional data sets such as crash records. The scope of this study was to answer the question, Do off- set left-turn lanes (bays) affect turn behavior—for example, gap acceptance and decision to make turn? The research team took the original research question posed in the SHRP 2 Proj- ect S02 report (McGehee et al. 2009) and broke it into more specific research questions as follows: 1. How does the gap-acceptance behavior of left-turning drivers vary as a function of the amount of offset between opposing left-turn lanes for intersections with a broad range of offsets? 2. What range of offsets for left-turn lanes is effective in min- imizing risk-taking for left-turning drivers by reducing the sight obstructions caused by opposing left-turn vehicles? 3. How should guidance on the desirable range of offsets be incorporated into geometric design policy for application by highway agencies? The first two questions can be directly answered with the NDS data. The third question requires interpretation of the results from investigation of the first two questions. The specific hypotheses tested in the study were as follows: 1. Do intersection approaches with negative offsets, zero offsets, and positive offsets for opposing left-turn vehicles differ from one another in the gap-acceptance behavior of left-turning drivers considering the following measures: critical gap (tc or t50), percentage of drivers accepting lags of specific durations less than tc or t50, and rate of occur- rence for erratic maneuvers during left turns? 2. Do intersection approaches with negative offsets, zero off- sets, and positive offsets for opposing left-turn vehicles differ in gap-acceptance behavior (using the same measures as for Hypothesis 1) between times when opposing left-turn vehi- cles are present (and potentially block the view of oncoming through traffic for both left-turning drivers) and times when only one left-turn vehicle is present (so that the driver’s view of oncoming through traffic is not blocked)? Thus, Hypothesis 1 involves a comparison of gap-acceptance behavior between intersections, while Hypothesis 2 involves a comparison of gap-acceptance behavior by drivers under dif- ferent traffic situations at the same intersection. The study data were not only used to test these specific hypotheses but were also used to develop broader relationships between gap-acceptance/ risk-taking measures and amount of offset to serve as a basis for establishing design guidelines (e.g., minimum desirable offset). In addition, human factors issues were explored through this research. The research attempted to determine the influence of the following factors on gap-acceptance behavior: • Length of time spent waiting for a suitable gap; • Presence of following left-turn vehicle(s); • Age and gender of left-turning driver; • Type (size) of turning and opposing through vehicle; • Pavement and weather conditions; and • Light condition. While this study focused on the behavior of left-turning drivers at intersections with varying left-turn lane offset condi- tions, the research team hopes that this research will also provide insight into using the NDS data to address other issues related to intersection left-turn lanes that will assist future research. Organization of the Report The remainder of this report provides background information on the use of offset left-turn lanes and driver gap-acceptance behavior (Chapter 2), a description of the site selection and data collection efforts (Chapter 3), the statistical analysis results (Chapter 4), and recommendations for application of results and future research (Chapter 5). Appendix A presents selected research results in greater detail than Chapter 4.