School of Civil & Environmental Engineering, Georgia Institute of Technology, 788 Atlantic Drive NE, Atlanta, GA 30332, United States.
School of Civil & Environmental Engineering, Georgia Institute of Technology, 788 Atlantic Drive NE, Atlanta, GA 30332, United States.
Accid Anal Prev. 2021 Oct;161:106351. doi: 10.1016/j.aap.2021.106351. Epub 2021 Aug 27.
Cyclists and pedestrians account for a disproportionate amount of the world's 1.3 million road deaths every year. This is a growing problem in the United Sates where bicyclist and pedestrian fatalities have increased steadily since 2009. A large body of research suggests vehicle speeds are a key contributing factor for crashes. However, few studies of bicycle or pedestrian crash probability incorporate detailed vehicle speed data. This study uses probe vehicle speed data to examine the impact of vehicle speeds on bicycle and pedestrian crashes on the state of Georgia's network of major arterial roadways. The analysis examines 7000 road segments throughout the state in 2017. A Negative Binomial model relates annual crash and speed data on each segment. Models using speed percentiles (85th, 50th and 15th) are contrasted with models using speed differences (85th-50th and 50th-15th percentile). A small set of covariates are included: segment length, number of lanes, Average Annual Daily Traffic, and urbanicity. Results indicate that larger differences in high-end speed percentiles are positively associated with bicycle and pedestrian crash frequency on Georgia arterials. Furthermore, the coefficients on the high end of the speed distribution, measured by the difference in 85th and 50th percentile speeds, have greater magnitude and statistical significance than the low end of the distribution. This research shows a negative relationship between speed and crashes may be flawed, as it does not account for the distributions of speed. The findings in this study suggest that planners and engineers should identify areas with large speed distributions, especially at the high vehicle speeds, and work to reduce the fastest speeds on these roadways. To do so, differences in speed percentiles measured using probe vehicle speeds can be used to determine where high risk areas are located.
骑自行车的人和行人在全球每年 130 万人的道路死亡中占不成比例的比例。这是美国一个日益严重的问题，自 2009 年以来，骑自行车和行人的死亡人数稳步上升。大量研究表明，车辆速度是导致撞车事故的一个关键因素。然而，很少有研究将自行车或行人碰撞概率与详细的车辆速度数据结合起来。本研究使用探针车辆速度数据来研究车辆速度对佐治亚州主要动脉道路网络上自行车和行人碰撞的影响。该分析检查了 2017 年全州的 7000 个路段。负二项式模型将每个路段的年度碰撞和速度数据联系起来。使用速度百分位数（第 85 个、第 50 个和第 15 个）的模型与使用速度差异（第 85 个-第 50 个和第 50 个-第 15 个百分位数）的模型形成对比。包含一小部分协变量：路段长度、车道数量、平均年日交通量和城市化程度。结果表明，高端速度百分位数的较大差异与佐治亚州动脉自行车和行人碰撞频率呈正相关。此外，速度分布高端的系数（通过第 85 个和第 50 个百分位数速度之间的差异来衡量）的幅度和统计学意义大于分布低端的系数。这项研究表明，速度与碰撞之间的负相关关系可能存在缺陷，因为它没有考虑到速度的分布。本研究的结果表明，规划者和工程师应该确定速度分布较大的区域，特别是在车辆速度较高的区域，并努力降低这些道路上的最快速度。为此，可以使用探针车辆速度测量的速度百分位数差异来确定高风险区域的位置。
