Analysis of Naturalistic Driving Study Data: Roadway Departures on Rural Two-Lane Curves (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.

8C h a p t e r 2 Background The Federal Highway Administration (2009) estimates that 58% of roadway fatalities are roadway departures, while 40% of fatalities are single-vehicle run-off-road (SVROR) crashes. Addressing roadway departure crashes is therefore a priority for national, state, and local roadway agencies; horizontal curves are of particular interest because they have been cor- related with overall increased crash occurrence. Glennon et al. (1985) reported that curves have approximately three times the crash rate of tangent sections, and Preston (2009) reported that 25% to 50% of severe road departure crashes in Minnesota occurred on curves, even though they account for only 10% of the system mileage. Farmer and Lund (2002) found that the odds of having a rollover on a curved section were 1.42 to 2.15 times greater than the odds of having a roll- over on a straight section. The majority of crashes on curves involve roadway departures. A total of 76% of curve-related fatal crashes are single vehicles leaving the roadway and strik- ing a fixed object or overturning. Another 11% of curve- related crashes are head-on collisions (AASHTO 2008). Curve-related crashes have a number of causes, including roadway and driver factors. Degree of curve or radius of curve is the roadway factor most cited in the literature as having an impact on crash risk (Luediger et al. 1988; Miaou and Lum 1993). Other factors that have been correlated to the frequency and severity of curve-related crashes include length of curve, type of curve transition, lane and shoulder widths (Zegeer et al. 1991), preceding tangent length (Milton and Mannering 1998), presence of spirals (Council 1998), grade (Fink and Krammes 1995), and required speed reduction between the tangent and curve. Driver error on horizontal curves is often due to inappro- priate speed selection, which results in an inability to main- tain lane position. FHWA estimates that approximately 56% of run-off-road (ROR) fatal crashes on curves are speed related. Distracting tasks such as radio tuning or cell phone conversations can draw a driver’s attention away from speed monitoring, changes in roadway direction, lane keeping, and detection of potential hazards (Charlton 2007). Other factors include sight distance issues, fatigue, or complexity of the driving situation (Charlton and DePont 2007; Charlton 2007). McLaughlin et al. (2009) evaluated ROR events in VTTI’s 100- car naturalistic driving study and found that distraction was the most frequently identified contributing factor, along with fatigue, impairment, and maneuvering errors. Environmental factors, such as the roadway surface condi- tion, and vehicle factors, such as the center of gravity, also have an impact on a driver’s ability to safely negotiate a curve. McLaughlin et al. (2009) found that ROR events were 1.8 times more likely on wet roads than dry, 7.0 times more likely on roads with snow or ice than dry roads, and 2.5 times more likely in nighttime than daytime conditions. Studies of roadway factors—such as degree of curve, pres- ence of spirals, or shoulder width and type—have provided some information regarding which curve characteristics are the most relevant, but information is still lacking. In addition, little information is available that identifies driver behaviors that contribute to curve crashes. As a result, a better under- standing of how drivers interact with various roadway fea- tures and countermeasures will provide valuable information to highway agencies in determining how resources can best be allocated to maximize driver performance and reduce crashes. rationale for research Although some studies have assessed the relationship between crash risk and driver and roadway characteristics, the con- tributory factors have not been well established. This is pri- marily because crash data typically have only a limited number of roadway variables, and driver behavior leading up to a crash can only be inferred by the reporting police officer. This lack of understanding makes it difficult to design, select, and apply the appropriate countermeasures, given that Introduction

9 safety professionals do not really understand how drivers are failing to interact with the roadway. Most studies that have evaluated countermeasures are based on crash or speed reductions. The results of these studies only demonstrate that a countermeasure works or does not work. They do not explain why the countermeasure works. For instance, several studies have examined the use of wider pavement markings or raised pavement markings (Donnell et al. 2006) to reduce crashes in curves. Although the treatments have shown some promise, why they are effective is not well understood. The treatment may be successful for several reasons: (1) the treat- ment provides better delineation so that a driver is better able to judge the sharpness of the curve; (2) the treatment may simply get the driver’s attention; or (3) the treatment may cause the lane to appear narrower, causing the driver to feel more constrained and resulting in lower speeds. Because it is unclear why the treatment is effective, it is unclear whether applying a particular countermeasure at a different curve will also be effective, and the understanding of the problem is insufficient to support the design and selection of alternative treatments. A better understanding of the interaction between driver characteristics and curve negotiation needs can lead to better design and application of countermeasures. For instance, if older drivers have the hardest time with curve negotiation because they are less likely to see visual cues, the best solution might be larger chevrons. Conversely, a solution geared toward younger drivers might include more closely spaced chevrons to help drivers gauge the sharpness of the curve. Distracted drivers might require another solution, such as a tactile cue from transverse rumble strips. Objectives Rural curves pose a significant safety problem, and the inter- action between the driver and roadway environment is not well understood. To address this knowledge gap, this research aimed to assess the relationship between driver characteris- tics and behavior, roadway factors, environmental factors, and the likelihood of roadway departures on rural two-lane paved curves. To accomplish this objective, the second Strategic Highway Research Program (SHRP 2) Naturalistic Driving Study (NDS) and Roadway Information Database (RID) were used to develop models that explore how drivers interact with the roadway environment and that identify the conditions pres- ent both when a driver does not successfully negotiate a rural curve and when successful negotiation occurs. These condi- tions include driver, roadway, and, to limited extent, environ- mental conditions. Most highway agencies are proactive in implementing a range of countermeasures to reduce roadway departures on curves and in other areas. However, agencies have only limited information about the effectiveness of different counter- measures. The results of this research will provide more infor- mation about the specific roadway features that are correlated to increased risk of roadway departure. The results will also provide valuable information about how drivers interact with roadway features and the impact that that interaction has on the effectiveness of countermeasures. This information will allow agencies to make better decisions about countermeasure selection. This research focused on two-lane rural curves. Addressing roadway departures on all curves is important; however, the SHRP 2 NDS encompassed almost 5 million trips, and—given time and resources constraints—researchers could not con- sider all curve-related roadway departure scenarios. Rural two-lane roads were prioritized and selected because of the disproportionate number of roadway departure crashes expe- rienced on these roads (Garder 2006; Fitzpatrick et al. 2002). Only paved roadways were included because the machine vision application does not function well when lane lines or obvious discontinuities between the lane and shoulder sur- face are not present. Rural was defined as one or more miles outside an urban area. Additionally, only roadways posted at 64 km/h to 97 km/h (40 mph to 60 mph) were included. research Questions addressed This main research question addressed was the following: What is the relationship between driver distraction, other driver characteristics, roadway characteristics, environmental characteristics, and risk of roadway departure? The data that were obtained and reduced allowed research- ers to explore driver behavior on curves in several additional ways. Each way offers a different perspective, and each was posed as an individual research question. As a result, a differ- ent methodology was proposed for each, as described in the corresponding sections. This research was tailored to address four fundamental research questions: 1. What defines the curve area of influence? 2. What defines normal behavior on curves? 3. What is the relationship between driver distraction; other driver, roadway, and environmental characteristics; and risk of roadway departure? 4. Can lane position at a particular state be predicted as a function of position in a prior state? As described in Chapter 4, distraction for the purposes of this research was defined as engagement in a non-driving-related activity while the driver was glancing in a location other than the forward roadway, or “eyes-off-roadway.”