Factors Affecting Red-light Running Behaviors of Takeaway Delivery Riders Considering Heterogeneity in the Means and Variances
-
摘要: 针对外卖骑手闯红灯事件频发、事故风险隐患较高的问题，调查了西安市多个信号交叉口处外卖骑手的闯红灯行为。将闯红灯行为作为因变量，将骑手个人特征、穿越行为特征、交通及环境特征等作为自变量，构建了考虑均值及方差异质性的随机参数Logit模型。采用Halton序列抽样进行参数估计，并结合参数估计结果和平均边际效应值量化分析各自变量对因变量的影响。结果表明：“饿了么”和“UU跑腿”的骑手闯红灯概率较低，到达时段为红灯第2段或第3段、停车线之后等待通行、冲突方向机动车流量较大等变量也会显著降低闯红灯概率，而同向违章人数增加、午高峰、晚高峰等变量会显著增加闯红灯概率。其中，使闯红灯概率提升最大的变量为晚高峰，其平均边际效应为0.278；使闯红灯概率降低最大的变量为停车线后等待，其平均边际效应为-0.222。此外，停车线后等待和晚高峰为随机参数变量，参数均服从正态分布，其均值和标准差分别为-1.379，1.359和2.502，5.360，且都具有显著的均值及方差异质性。对于停车线后等待这个变量，红灯第2段会同时提高其参数均值和方差，即增加闯红灯的概率，并增加该变量对闯红灯行为影响的随机性。对于晚高峰变量，机动车流量较大会同时降低其参数均值和方差，即降低闯红灯的概率，并降低该变量对闯红灯行为影响的随机性。此外，违章人数为1也会降低晚高峰参数的方差。Abstract: To address the frequent occurrences of takeaway delivery riders running red-light and the high risk of crashes associated with this behavior, a filed survey is conducted at multiple signalized intersections in Xi'an, the red-light running (RLR) behaviors of delivery riders are investigated. The RLR behavior is taken as the dependent variable, while independent variables included rider personal characteristics, crossing behavior characteristics, and traffic and environmental features. A random parameter Logit model considering heterogeneity in the means and variances was constructed. Parameter estimation was carried out using Halton sequence sampling, and the impact of each independent variable on the dependent variable was quantitatively analyzed through the estimation results and average marginal effects. The findings indicate that Eleme and UU delivery riders have a lower probability of RLR. Variables such as arriving during the second or third phase of the red light, waiting behind the stop line for the green light, and higher conflicting direction traffic volumes significantly reduce the probability of RLR. Conversely, an increase in the number of violators in the same direction, the noon peak hours and evening peak hours significantly increase the probability of RLR. Among these, the variable that most significantly increases the probability of RLR is the evening peak hour, with an average marginal effect of 0.278; the variable that most significantly decreases the probability of RLR is waiting behind the stop line, with an average marginal effect of －0.222. Besides, the parameters of waiting behind the stop line and evening peak hours are random parameter variables, following a normal distribution with means and standard deviations of －1.379, 1.359 and 2.502, 5.360, respectively. Besides, both random parameters exhibit significant heterogeneity in means and variances. For the variable of waiting behind the stop line, arriving during the second phase of the red light increases both the mean and variance of this variable's parameter, hence increasing the probability of RLR and the randomness of its impact on this behavior. For the evening peak hour, a higher volume of motor vehicle traffic reduces both its parameter's mean and variance, thus lowering the probability of RLR and reducing the randomness of its impact on this behavior. Additionally, having only one violator also reduces the variance of the evening peak hour's parameter.
-
表 1 变量描述性统计
Table 1. Descriptive statistics of variables
变量名称 频数（比例/%） 字段解释 闯红灯行为 无闯红灯行为 245（25.51） 样本未发生闯红灯行为 有闯红灯行为 719（74.59） 样本发生闯红灯行为 性别 男性* 926（96.06） 骑手为男性 女性 38（3.94） 骑手为女性 骑手公司 其他* 163（16.91） 样本为其他公司骑手 美团 319（33.09） 样本为美团骑手 饿了么 183（18.98） 样本为饿了么骑手 UU跑腿 299（31.02） 样本为UU骑手 红灯时段 红灯第1段* 275（28.53） 骑手在红灯前1/3时段内到达 红灯第2段 331（34.34） 骑手在红灯中间1/3时段内到达 红灯第3段 358（37.14） 骑手在红灯后1/3时段内到达 红灯时段 停车线之前* 820（85.06） 骑手在停车线前等待通行 停车线之后 144（14.94） 骑手在停车线后等待通行 等待人数 0* 314（32.57） 骑手到达交叉口时，同方向正在等待通行的行人和非机动车数 1 159（16.49） 2 154（15.98） 3 138（14.32） 4 95（9.85） 5个或更多= 5 104（10.79） 违章人数 0* 735（76.24） 骑手到达交叉口时，同方向正在违规穿越的行人和非机动车数 1 149（15.46） 2 50（5.19） 3个或更多 30（3.11） 行驶方向 直行* 901（93.46） 电动自行车在交叉口直行 左转 63（6.54） 电动自行车在交叉口左转 天气 晴天* 417（43.26） 记录数据时的天气状况 阴天 358（37.14） 雨天 189（19.61） 时间段 平峰* 267（27.70） 记录数据时所处的交通高峰期 午高峰 388（40.28） 晚高峰 309（32.05） 信号灯类型 静态显示* 736（76.35） 交叉口信号灯类型 倒计时显示 228（23.65） 机动车流量 （0，5）* 91（9.44） 1个信号周期内，可能与外卖骑手发生冲突的每分钟每车道的标准机动车流量，单位是PCU/车道/min。 注：*为该自变量的基准变量。 
下载: 导出CSV
表 2 外卖骑手闯红灯行为的Logit模型估计结果
Table 2. Estimation results of Logit model for RLR behavior of takeaway delivery riders
变量名称 二项Logit模型参数值 随机参数Logit模型参数值 考虑均值及方差异质性的随机参数Logit模型参数值 常数项 2.425 1.882 2.113 骑手公司 饿了么 -0.651 -0.493 -0.587 UU跑腿 -0.802 -0.638 -0.763 红灯时段 红灯第2段 -1.085 -0.900 -1.188 红灯第3段 -1.641 -1.305 -1.423 等待位置 停车线之后等待 -1.369 -1.075 -1.379 停车线之后等待(标准差) 一 0.840 1.359 1 1.173 0.948 0.863 违章行为人数 2 1.910 1.541 1.392 3个或以上 2.964 2.294 2.068 行驶方向 左转 2.022 1.529 一 午高峰 0.492 0.364 0.368 时间段 晚高峰 0.609 0.728 2.502 晚高峰(标准差) 一 1.586 5.360 机动车流量 $ \geqslant 8 \mathrm{PCU} / $车道$ / \mathrm{min} $ -0.751 -0.583 -0.472 晚高峰$ ( $机动车流量$ \geqslant 8 \mathrm{PCU} / $车道$ / \mathrm{min}) $均值异质性 一 一 -2.030 停车线之后等待(红灯第2段)均值异质性 一 一 0.873 晚高峰(机动车流量$ \geqslant 8 $ PCU/车道/min)方差异质性 一 一 -1.883 晚高峰(违章人数为1)方差异质性 一 - -0.497 停车线之后等待(红灯第2段)方差异质性 一 一 2.229 模型收敛时的对数似然函数值 -441.681 -436.560 -430.968 AIC 911.362 905.120 903.936
下载: 导出CSV
表 3 各显著变量的平均边际效应
Table 3. Average marginal effects of significant variables
变量类型 变量名称 平均边际效应 个人特征 骑手公司 饿了么 -0.085 UU跑腿 -0.110 穿越行为特征 红灯时段 红灯第2段 -0.170 红灯第3段 -0.203 等待位置 停车线之后等待 -0.222 违章行为人数 1 0.108 2 0.150 3个或以上 0.190 交通及环境特征 时间段 午高峰 0.051 晚高峰 0.278 机动车流量 $ \geqslant 8 \mathrm{PCU} / $ (车道$ / \mathrm{min} $) -0.068
下载: 导出CSV
-
[1] 王维莉, 卢晓磊, 张旺, 等. 基于改进社会力的单向流电动自行车行为建模研究[J]. 交通运输系统工程与信, 2022, 22(5): 223-232. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202205023.htmWANG W L, LU X L, ZHANG W, et al. Modeling of electric bicycle behavior in unidirectional flow based on improved social forces[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(5): 223-232. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202205023.htm [2] 王薇. 认知偏差、风险感知对电动车驾驶行为的影响[D]. 银川: 宁夏大学, 2022.WANG W. The Impact of drivers' cognitive bias and risk perception on electric vehicle driving behavior[D]. Yinchuan: Ningxia University, 2022. (in Chinese) [3] BAI L, LIU P, CHEN Y G, et al. Comparative analysis of the safety effects of electric bikes at signalized intersections[J]. Transportation Research Part D: Transport and Environment, 2013, 20: 48-54. doi: 10.1016/j.trd.2013.02.001 [4] TANG T P, WANG H, MA J, et al. Analysis of crossing behavior and violations of electric bikers at signalized intersections[J]. Journal of Advanced Transportation, 2020(1): 1-14. [5] ZHANG F, KUAI C C, LYU H T, et al. Investigating different types of red-light running behaviors among urban e-bike rider mixed groups[J]. Journal of Advanced Transportation, 2021(1): 1-9. [6] 张凡, 吕卉焘, 沈小燕, 等. 计划行为理论下外卖配送员闯红灯行为研究[J]. 中国安全科学学报, 2019, 29(5): 1-6. https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201905001.htmZHANG F, LYU H T, SHEN X Y, et al. Study on takeaway deliverers' red light running behavior based on planned behavior theory[J]. China Safety Science Journal, 2019, 29(5): 1-6. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201905001.htm [7] YANG H T, LIU X H, SU F, et al. Predicting e-bike users' intention to run the red light: an application and extension of the theory of planned behavior[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2018, 58: 282-291. doi: 10.1016/j.trf.2018.05.027 [8] 王雅坤, 卢昕玮, 陈巧丽, 等. 外卖配送公共交通违法行为影响因素分析[J]. 中国安全生产科学技术, 2019, 15(12): 169-174. https://www.cnki.com.cn/Article/CJFDTOTAL-LDBK201912033.htmWANG Y K, LU X W, CHEN Q L, et al. Analysis on influencing factors of public traffic violation behavior in take out delivery[J]. Journal of Safety Science and Technology, 2019, 15(12): 169-174. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-LDBK201912033.htm [9] ZHANG F, JI Y J, LYU H T, et al. Analysis of factors influencing delivery e-bikes' red-light running behavior: a correlated mixed binary Logit approach[J]. Accident Analysis & Prevention, 2021, 152(1): 1-12. [10] LYU H T, LI H J, SZE N N, et al. The impacts of non-motorized traffic enforcement cameras on red light violations of cyclists at signalized intersections[J]. Journal of Safety Research, 2022, 83: 310-322. doi: 10.1016/j.jsr.2022.09.005 [11] 周继彪, 王群燕, 张敏捷, 等. 7种因素对电动自行车忍耐时间的实证研究[J]. 交通运输系统工程与信息, 2017, 17(5): 242-249. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201705035.htmZHOU J B, WANG Q Y, ZHANG M J, et al. An empirical study on seven factors influencing waiting endurance time of e-bike[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(5): 242-249. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201705035.htm [12] GUO Y Y, ZHOU J B, WU Y, et al. Evaluation of factors affecting e-bike involved crash and e-bike license plate use in China using a bivariate probit model [J]. Journal of Advanced Transportation, 2017(10): 1-12. [13] ZHOU J B, ZHENG T, DONG S, et al. Impact of helmet-wearing policy on e-bike safety riding behavior: a bivariate ordered probit analysis in Ningbo, China [J]. International Journal of Environmental Research and Public Health, 2022, 19(5): 1-21. [14] 宋栋栋, 杨小宝, 祖兴水, 等. 基于均值异质性随机参数Logit模型的城市道路事故驾驶员受伤严重程度研究[J]. 交通运输系统工程与信息, 2021, 21(3): 214-220. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202103027.htmSONG D D, YANG X B, ZU X S, et al. Examination of driver injury severity in urban crashes: a random parameters Logit model with heterogeneity in means approach[J]. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(3): 214-220. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202103027.htm [15] 赵伟宁. 基于改进Logit模型的高速公路交通事故严重程度分析方法[D]. 哈尔滨: 哈尔滨工业大学, 2020.ZHAO W N. Analysis of traffic accident injury severity on freeway based on improved Logit model[D]. Harbin: Harbin Institute of Technology, 2020. (in Chinese) [16] LEE J, LI X, MAO S Y, et al. Investigation of contributing factors to traffic crashes and violations: a random parameter multinomial Logit approach[J]. Journal of Advanced Transportation, 2021(1): 1-11. [17] 郝小妮, 石文瀚, 刘建荣, 等. 基于随机系数Logit模型的城市群城际出行方式选择行为研究[J]. 交通信息与安全, 2022, 40(5): 139-146. doi: 10.3963/j.jssn.1674-4861.2022.05.015HAO X N, SHI W H, LIU J R, et al. An analysis of mode choice behavior of inter-city travel in urban agglomeration areas using a random-parameter Logit model[J]. Journal of Transport Information and Safety, 2022, 40(5): 139-146. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2022.05.015 [18] 焦朋朋, 李汝鉴, 王健宇, 等. 考虑潜在类别的老年行人交通事故严重程度致因分析[J]. 交通运输系统工程与信息, 2022, 22(5): 328-336. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202205034.htmJIAO P P, LI R J, WANG J Y, et al. Causes analysis on severity of elderly pedestrian crashes considering latent classes[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(5): 328-336. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202205034.htm [19] 施颖, 潘义勇, 吴静婷. 基于随机参数Logit模型的校车事故伤害严重程度分析[J]. 交通信息与安全, 2021, 39(5): 43-49. doi: 10.3963/j.jssn.1674-4861.2021.05.006SHI Y, PAN Y Y, WU J T. An analysis of injury severities in school bus accidents based on random parameter Logit models[J]. Journal of Transport Information and Safety, 2021, 39(5): 43-49. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2021.05.006 [20] HUAN M, YANG X B, JIA B. Crossing reliability of electric bike riders at urban intersections[J]. Mathematical Problems in Engineering, 2013(1): 1-8. [21] GAO X, ZHAO J, GAO H. Red-light running behavior of delivery-service E-cyclists based on survival analysis[J]. Traffic Injury Prevention, 2020, 21(8): 558-562. [22] TANG T P, WANG H, MA J, et al. Analysis of crossing behavior and violations of electric bikers at signalized intersections[J]. Journal of Advanced Transportation, 2020(1): 1-14. [23] YANG X B, HUAN M, ABDEL-ATY M, et al. A hazard-based duration model for analyzing crossing behavior of cyclists and electric bike riders at signalized intersections[J]. Accident Analysis & Prevention, 2015, 74: 33-41. -
点击查看大图
图(1) / 表(3)
计量
- 文章访问数: 54
- HTML全文浏览量: 29
- PDF下载量: 16
- 被引次数: 0