«Differences in Motorcycle Conspicuity-related Factors and Motorcycle Crash Severities in Daylight and Dark Conditions Mohammad Saad B. Shaheed ...»
Shaheed, Zhang, Gkritza, Hans
Differences in Motorcycle Conspicuity-related Factors and Motorcycle Crash
Severities in Daylight and Dark Conditions
Mohammad Saad B. Shaheed (Corresponding author)
Graduate Research Assistant, Department of Civil, Construction and Environmental Engineering
Institute for Transportation
Iowa State University, Ames, Iowa 50011
Assistant Professor, Department of Civil, Construction, and Environmental Engineering Center for Transportation Research and Education Institute for Transportation Iowa State University, Ames, Iowa 50011 Phone: 515-294-2343, Fax: 515-294-7424 E-mail: firstname.lastname@example.org Zach Hans Research Engineer Center for Transportation Research and Education Institute for Transportation Iowa State University, Ames, Iowa 50011 Phone: 515-294-2329 E-mail: email@example.com Submitted to the 3rd International Conference on Road Safety and Simulation, September 14-16, 2011, Indianapolis, USA
1 Shaheed, Zhang, Gkritza, Hans This paper further examines the distribution of collision configurations (such as “non- motorcycle” being “at-fault” in rear-end collisions, angle crashes, and sideswipe crashes) and factors (related to “non-motorcycle” vehicle drivers) potentially related to not seeing motorcyclists for different motorcycle crash severity outcomes using contingency table analysis.
Finally, this paper develops a multinomial logit model to investigate the effect of potential motorcycle-conspicuity related factors on motorcycle crash severity outcomes. The results from the model show that angle crashes (“non-motorcycle” turning left), rear end crashes (“nonmotorcycle” hitting “motorcycle”), light conditions, failure to yield ROW (right of way) by nonmotorcycle drivers, light conditions and other variables played significant roles in motorcycle crash-injury outcome. The limitations of examining motorcycle conspicuity by analysis of crash data are also discussed.
Keywords: motorcycle safety, conspicuity, daylight, dark, multinomial logit model.
INTRODUCTIONPrevious studies in the United States and internationally suggest that low motorcycle conspicuity, or the inability of the motorcyclist to be seen by other road users is thought to be an important factor associated with the risk of motorcycle crashes (Cercarelli, 1992; Hurt et al., 1981; Williams and Hoffmann, 1979; Wulf et al., 1989). This may be due to several factors including the size of motorcycle, irregular outline of the vehicle, low luminance or contrast with the background environment, and the ability to maneuver in the traffic stream. Additional measures to enhancing conspicuity include wearing reflective or fluorescent clothing, wearing white or light colored helmets, using headlights during daytime, and using headlamp modulators (Jenkins and Wigan, 1985; Muller, 1984; Olson et al., 1981; Rumar, 1980; Thomson, 1980;
Torrez, 2008; Well et al., 2004). Safety campaigns also advocate the importance of conspicuity to motorcycle drivers. Baer et al. (2010) mentioned that most states in the United States have implemented conspicuity and/or motorist awareness campaigns. In addition, there are recommendations in place encouraging states to initiate public awareness efforts that are focused on the use of high visibility riding gear and daytime running lights; take steps to alert motorists about motorcyclists, using strategies such as incorporating “Share the Road” messages as part of driver’s education classes; and amplify public information and outreach efforts” (Baer and Skemer, 2010).
However, it is not an easy task to motivate all motorcyclists to wear the proper type of reflective gear, especially when there is no law enforcing it. To date, twenty states and the District of Columbia have universal motorcycle helmet laws that require all riders to wear helmet. Twentyseven states require only some riders to wear helmet, and three states (Illinois, Iowa and New Hampshire) do not have a motorcycle helmet law. Baer et al. (2010) stated that while the majority of states promote helmet use, only just over half of the states emphasize the use of eye and face protection.
Motorcycle safety concerns and renewed interest from states in increasing motorcycle conspicuity and motorist awareness make the present moment an opportune time to revisit the problem of motorcycle conspicuity. First, this paper reviews previous studies on motorcycle conspicuity with a focus on the effectiveness of proposed measures for enhancing conspicuity.
2Shaheed, Zhang, Gkritza, Hans
Then, using motorcycle crash data for Iowa from 2001 to 2008, this paper identifies those motorcycle-conspicuity factors that could potentially relate to a collision between a motorcycle and another vehicle. This paper also develops a multinomial logit model to examine the effect of these factors (with a focus on the “non-motorcycle” driver contributing factors) on two vehicle motorcycle crash severity outcomes.
LITERATURE REVIEWClothing and Gear The injury reduction benefits of motorcycle rider clothing have been well established in the literature (de Rome, 2006). Rider clothing and gear have also been associated with the risk of being involved in motorcycle collisions. A case study in New Zealand (Wells et al., 2004) showed that there was a 37-percent lower risk of motorcyclists getting into a severe traffic collision if the driver was wearing any reflective or fluorescent clothing and a 24-percent lower risk if the driver was wearing a white helmet instead of a black one. However, the color of the driver’s frontal clothing and motorcycle was not as important. Seasonal variations in conspicuity of high-visibility garments have been also investigated through a naturalistic, daytime field study of pedestrians (Buonarosa and Sayer, 2007). The results showed that the distances at which pedestrians were first detected depended on the season and the amount of background material (jacket or vest). The researchers did not find any significant differences in daytime conspicuity if pedestrians were wearing either fluorescent yellow-green or fluorescent red-orange clothing.
However, they found that the conspicuity of fluorescent red-orange garments might depend primarily on color contrast, while the conspicuity of fluorescent yellow-green garments might depend primarily on luminance contrast. A similar experimental study on motorcyclists (Hole et al., 1996) confirmed that the brightness contrast between the motorcyclist and the surroundings may be a more important conspicuity-related factor than bright clothing and headlight use alone.
Daytime Running Lights
National and international studies have shown that legislation mandating motorcycle headlightuse during daytime has been effective in reducing the number of multi-vehicle motorcycle collisions (Rumar, 1980; Rudin et al., 1996; Thomson, 1980), and the number of fatal and serious injury accidents (Muller, 1984; Yuan, 2000; Zador, 1985). However, not all previous work has shown positive benefits of motorcycle daytime light laws. For example, the change in the number of minor injury accidents after the implementation of similar legislation in Singapore was insignificant (Yuan, 2000). Interestingly, a study of fatal crashes in the United States from 1975 to 1980 (Muller, 1985) found no statistical significant differences between states with and without daytime headlight laws. The safety benefits of motorcycle daytime headlight laws have also been questioned in (Perlot and Prower, 2003). The discrepancies in the results on the estimated effectiveness of motorcycle headlight laws are likely attributed to the study design and estimation methods used (Muller, 1985; Perlot and Prower, 2003). In addition, enhanced conspicuity might lead to risk perception and risk-compensating (or offsetting) behavior, in which drivers adjust their driving behavior in response to situations that can be perceived as comparatively dangerous or safe. For example, drivers in Norway perceived comparatively safe having the headlights on during daytime (Elvik, 1993). Turning to the specifications of daytime
running lights, two lamps and lamps over 180mm diameter have been shown to offer a greater conspicuity advantage than single or smaller lamps (Donne and Fulton, 1985).
Motorcycle Actions and Driver Awareness Previous research has argued that there is a conspicuity problem associated with motorcycles, as other vehicles in the traffic stream that make a left turn are unable to recognize them in the daylight, in comparison to other vehicles (Olson et al., 1979; Williams and Hoffmann, 1979).
This might be due to the difference in frontal surface area and low visibility of the motorcyclist.
For example, riders who were wearing fluorescent clothing and riding with headlamp lights during daytime were less likely to be involved in this type of collisions. This could be also attributed to the complexity of the situation. For example, making a left turn significantly increases a driver’s head movements, eye movements, and mental workload in comparison to driving straight through an intersection (Wulf et al., 1989). Interestingly, in a comparison of motorcycle-car crashes to car-car crashes (Cercarelli et al., 1992), no statistical significant differences in the day and night distributions of car-car and car-motorcycle crashes were found.
As such, the argument, according to which drivers have more trouble seeing motorcycles in the daylight in comparison to other vehicles, due to the difference in frontal surface area, needs to be revisited.
Public Information and Outreach Efforts
Campaigns or interventions have also been claimed to be successful in reducing conspicuityrelated motorcycle crashes. A report by road safety committee in Victoria, Australia mentioned that public information campaigns in Victoria, Australia to increase the voluntary use of colored and fluorescent clothing and daytime running lights were found to be effective, but their effect tapered off after about nine months. In a survey of state motorcycle safety programs (Baer et al., 2010), 96% of the 44 states that responded to the survey stated that have implemented conspicuity and/or motorist awareness campaigns, and one-half of the states indicated that they have programs at schools to educate students about motorcycle safety. However, voluntary action countermeasures such as motorcycle education and training courses, motorcycle helmet use promotion programs, and education to encourage motorcyclists to increase their conspicuity have not been proven effective yet; and some evidence suggest that these countermeasures are unlikely to be effective (Preusser et al., 2008). Also, public awareness campaigns are increasingly reaching out to both manufacturers of rider’s clothing and motorcyclists in order to encourage the producing and wearing of highly reflective materials to improve the visibility of motorcyclists on the road (Wells, 2004).
DATA AND DESCRIPTIVE STATISTICS
Data on reported motorcycle crashes were collected for the 8–year period from 2001 to 2008 from the crash database maintained by the Iowa Department of Transportation. The data collected included information on two vehicle motorcycle crashes where one vehicle was motorcycle and the other was a non-motorcycle vehicle. Attributes included: year, month, day and time of crash; crash location (urban or rural area); road surface and environmental conditions; manner of crash, crash severity, major cause of the crash, and events contributing to
4Shaheed, Zhang, Gkritza, Hans
the crash; and other information about the motorcycle driver and the driver of the “nonmotorcycle” vehicle involved in the crash (such as helmet use). However, potential conspicuityrelated factors such as rider clothing, color of motorcycle, helmet color, and motorcycle type could not be collected from the crash database. The crashes occurring within one mile of the corporate city limits were defined as urban, while the crashes occurring outside the city boundaries were defined as rural. It should be noted that property damage crashes of less than $1,000 are not included in the crash database maintained by the Iowa Department of Transportation. Table 1 shows the summary statistics for select variables for two-vehicle motorcycle (MC) crashes.
Table 1 Summary statistics for select variables for two-vehicle motorcycle crashes
Manner of crash/collision Head-on/rear-end, non-MC hitting MC/rear-end, MC hitting non-MC/angle, non-MC turning left and MC moving straight/ 3.8/11.7/13.1/10.7/2.3/37.8/6.8/ angle, MC turning left and non-MC moving straight/ 2.1/11.7 broadside/sideswipe, same direction/sideswipe, opposite direction/other
Age of the driver of the “nonmotorcycle” vehicle under 20/ 21 to 30/31 to 40/ 43.3 (24.0) 19.9/20.3/13.9/12.9/10.3/22.7 41 to 50/ 51 to 60/ over 60 years old