Spatiotemporal characteristics and severity modelling of electric vehicle-pedestrian collision accidents
Article Text (Baidu Translation)
-
摘要: 为了探索电动汽车与行人碰撞事故的时空分布模式，在传统核密度估计法中引入聚类强度和事故严重度指标，建立了交通事故黑点时空核密度估计方法；考虑到事故变量中可能存在的随机变量和异质性，基于均值异质性的随机参数Logit模型，建立了电动汽车与行人碰撞事故严重程度模型，识别影响电动汽车与行人碰撞事故严重程度的关键因素，并通过计算显著变量的边际效用定量确定了关键因素的影响程度。研究结果表明：电动汽车和燃油汽车与行人碰撞事故均主要分布在城市中心区域，电动汽车与行人碰撞事故分布相对更为集中，且工作日和白天时电动汽车的事故数更多，严重程度更高；电动汽车驾驶人年龄在[40，60)岁和事故发生时间段在18:00至次日6:00为随机变量，驾驶人年龄的效用值服从标准差为-0.582、方差为0.9262的正态分布，且在道路功能等级(主干路)和车辆类型(货车)变量中具有均值异质性，事故发生时间段的效用值服从标准差为-0.313、方差为0.5182的正态分布，且在天气(雾霾)变量中具有均值异质性，此外，男性驾驶人、中青年驾驶人和工作日出行等因素会导致电动汽车与行人碰撞事故严重程度的不同增加；考虑均值异质性的随机参数Logit模型的赤池信息准则比随机参数Logit模型减少了26，对数似然值增加了94.272，可见，引入均值异质性后，构建的基于随机参数Logit模型的电动汽车与行人碰撞事故严重程度模型拟合优度有所提升，且能更科学全面地解释电动汽车与行人碰撞事故的致因原理。Abstract: To investigate the spatiotemporal distribution pattern of electric vehicle-pedestrian collision accidents, the indicators of clustering strength and accident severity were incorporated into the traditional kernel density estimation method. A spatiotemporal kernel density estimation approach for traffic accident-prone locations was built. The random parameter Logit model based on mean heterogeneity was developed to model the severity of electric vehicle-pedestrian collision accident by considering the potential random variables and heterogeneity in accident variables. Key factors affecting the severity of electric vehicle-pedestrian collision accidents were identified, and the marginal utility of each significant variable was calculated to quantify its impact on the severity of electric vehicle-pedestrian collision accidents. Research results show that the electric/fuel vehicle-pedestrian collision accidents are predominantly located in urban center areas, whereas the distribution of electric vehicle-pedestrian collision accidents is relatively more concentrated. Moreover, the number and severity of electric vehicle-pedestrian collision accidents are higher on weekdays and daytime. Electric vehicle driver age ranging from 40 to 60 years old and accident occurring time between 18:00 and 6:00 the next day are identified as random variables. The utility value of driver age follows a normal distribution with standard deviation of -0.582 and variance of 0.9262, and has mean heterogeneity in variables of road function level (trunk road) and vehicle type (truck). The utility value of accident occurring time follows a normal distribution with standard deviation of -0.313 and variance of 0.5182, and there is mean heterogeneity in variable of weather (haze). Additionally, the factors such as male drivers, middle-aged and young drivers, and weekday travel are found to contribute to the varying degrees of severity in electric vehicle-pedestrain collision accidents. The Akaike information criterion of the random parameter Logit model considering mean heterogeneity decreases by 26 compared with the random parameter Logit model, and the logarithmic likelihood value increases by 94.272. The introduction of mean heterogeneity improves the fit goodness of severity model for electric vehicle-pedestrian collision accidents based on the random parameter Logit model, providing a more scientific and comprehensive explanation of causal principles behind electric vehicle-pedestrian collision accidents.
-
图 1 行人碰撞事故中影响因素分布特征对比
Figure 1. Comparison of distribution characteristics of influencing factors in pedestrian collision accidents
图 2 汽车与行人碰撞的事故核密度
Figure 2. Accident kerhel densities of vehicle-pedestrian collisions
图 3 不同等级道路上汽车与行人碰撞事故的分布
Figure 3. Distribution of vehicle-pedestrian collision accidents on different grade roads
图 4 汽车与行人碰撞事故的时间分布
Figure 4. Time distributions of vehicle-pedestrian collision accidents
图 5 汽车与行人碰撞事故的时空分布
Figure 5. Spatiotemporal distributions of vehicle-pedestrian collision accidents
表 1 样本描述性统计
Table 1. Sample descriptive statistics
特征分类 参数 电动汽车 燃油汽车 非死亡事故 死亡事故 非死亡事故 死亡事故 起数 百分比/% 起数 百分比/% 起数 百分比/% 起数 百分比/% 驾驶人特征 性别 男 43 97.73 1 2.27 28 59.57 19 40.43 女 14 100.00 4 66.67 2 33.33 年龄/岁 [20，40) 39 97.50 1 2.50 23 65.71 12 34.29 [40，60) 16 100.00 9 52.94 8 47.06 [60，80) 2 100.00 1 100.00 车辆特征 类别 公交 2 100.00 2 66.67 1 33.33 小汽车 50 100.00 25 64.10 14 35.90 货车 5 83.33 1 16.67 5 45.45 6 54.55 道路特征 道路等级分类 快速路 5 100.00 主干路 6 85.71 1 14.29 5 45.45 6 54.55 次干路 6 100.00 10 83.33 2 16.67 支路 45 100.00 17 68.00 8 32.00 环境特征 天气 晴 52 98.11 1 1.89 27 64.29 15 35.71 雾霾 3 100.00 3 42.86 4 57.14 雨、雪 1 50.00 1 50.00 2 50.00 2 50.00 周内分布 工作日 43 97.73 1 2.27 25 59.52 17 40.48 周末 14 100.00 7 63.64 4 36.36 时间段分布 6:00~18:00 42 97.68 1 2.32 12 54.55 10 45.45 18:00~6:00 15 100.00 20 64.52 11 35.48 
下载: 导出CSV
表 2 汽车与行人碰撞事故的分布数值
Table 2. Distribution values of vehicle-pedestrian collision accidents
电动汽车 燃油汽车 时间段 位置代码 聚类强度 时间段 位置代码 聚类强度 14:00~15:00 220 0.744 20:00~21:00 136 0.651 19:00~20:00 314 0.676 19:00~20:00 251 0.526 20:00~21:00 220 0.666 18:00~19:00 152 0.511 16:00~17:00 20 0.617 14:00~15:00 37 0.507 18:00~19:00 659 0.613 16:00~17:00 659 0.493 22:00~23:00 96 0.577 7:00~8:00 72 0.482 20:00~21:00 152 0.536 22:00~23:00 125 0.429 9:00~10:00 95 0.521 20:00~21:00 221 0.411
下载: 导出CSV
表 3 变量的VIF
Table 3. VIF of variables
变量 VIF 变量 VIF 女 1.180 主干路 1.103 [40，60)岁 1.249 次干路 1.143 [60，80)岁 1.194 雾霾 1.260 公交 1.163 雨、雪 1.297 货车 1.131 周末 1.176 快速路 1.380 18:00~6:00 1.309
下载: 导出CSV
表 4 模型估算结果
Table 4. Estimation results of models
变量 随机参数Logit模型 均值异质性随机参数Logit模型 平均边际效应 电动汽车 燃油汽车 电动汽车 燃油汽车 电动汽车 燃油汽车 系数估计 Z值 系数估计 Z值 系数估计 Z值 系数估计 Z值 非死亡事故 死亡事故 非死亡事故 死亡事故 常数 1.577 4.240 0.741 2.010 1.283 3.246 0.752 2.790 女 -1.822 1.420 -0.113 4.303 -1.671 2.752 -0.075 3.714 0.002 -0.002 0.004 -0.004 [40，60)岁 0.649 2.258 2.940 2.911 -0.582 1.255 0.560 1.824 0.003 -0.003 -0.023 0.023 [40，60)岁的标准差 0.254 1.618 0.182 1.785 0.926 2.862 1.425 2.663 公交 0.501 1.738 -0.745 1.793 0.575 3.628 -0.824 3.054 -0.001 0.001 0.002 -0.002 货车 -2.416 2.672 0.391 2.036 -2.367 2.285 0.541 2.793 0.004 -0.004 -0.017 0.017 主干路 -9.284 2.074 1.752 2.825 -9.361 2.349 1.893 2.512 0.006 -0.006 -0.053 0.053 次干路 0.562 3.547 2.437 3.281 0.579 3.277 2.547 3.727 -0.007 0.007 0.076 -0.076 雾霾 -0.283 2.249 0.352 2.461 -0.294 3.162 0.352 2.625 0.001 -0.001 -0.005 0.005 周末 -0.517 2.813 -0.627 1.943 -0.546 2.724 -0.641 1.916 0.003 -0.003 0.006 -0.006 18:00~6:00 -0.677 3.594 1.293 3.416 -0.313 3.012 0.960 3.175 0.006 -0.006 -0.032 0.032 18:00~6:00的标准差 0.467 1.537 0.174 2.723 0.518 1.572 1.186 2.682 [40，60)岁，主干路 -0.170 -3.571 0.136 2.158 [40，60)岁，货车 -0.316 -1.284 0.272 1.682 18:00~6:00，雾霾 -0.146 -2.035 0.229 3.186 AIC 4 311 4 241 4 285 4 109 模型收敛时的对数似然值 -2 300.691 -2 300.451 -2 206.419 -2 041.357 伪R2 0.365 0.326 0.463 0.452
下载: 导出CSV
-
[1] SANGUESA J A, TORRES-SANZ V, GARRIDO P, et al. A review on electric vehicles: technologies and challenges[J]. Smart Cities, 2021, 4(1): 372-404. doi: 10.3390/smartcities4010022 [2] National Highway Traffic Safety Administration. Fatality analysis reporting system[R]. Washington DC: National Highway Traffic Safety Administration, 2018. [3] TATARIS K. Incidence rates of pedestrian and bicyclist crashes by hybrid electric passenger vehicles: an update[J]. Annals of Emergency Medicine, 2014, 64(2): 195-196. doi: 10.1016/j.annemergmed.2014.05.017 [4] WU Jing-shu. Updated analysis of pedestrian and pedalcyclist crashes with hybrid vehicles[R]. Washington DC: National Highway Traffic Safety Administration, 2017. [5] 姜良维, 张树林, 金会庆. 我国新能源汽车交通事故特征及其管控对策[J]. 人类工效学, 2021, 27(6): 59-62.JIANG Liang-wei, ZHANG Shu-lin, JIN Hui-qing. Characteristics and control countermeasures of new energy vehicle traffic accidents in China[J]. Chinese Journal of Ergonomics, 2021, 27(6): 59-62. (in Chinese) [6] 梁新苗, 肖凌云, 王澎, 等. 数据挖掘与现场调查结合的电动汽车事故分析[J]. 中国安全科学学报, 2022, 32(1): 180-187.LIANG Xin-miao, XIAO Ling-yun, WANG Peng, et al. Analysis on electric vehicle accidents based on data mining and site investigation[J]. China Safety Science Journal, 2022, 32(1): 180-187. (in Chinese) [7] DAI D J, JAWORSKI D. Influence of built environment on pedestrian crashes: a network-based GIS analysis[J]. Applied Geography, 2016, 73: 53-61. doi: 10.1016/j.apgeog.2016.06.005 [8] YAO Shen-jun, WANG Jin-zi, FANG Lei, et al. Identification of vehicle-pedestrian collision hotspots at the micro-level using network kernel density estimation and random forests: a case study in Shanghai, China[J]. Sustainability, 2018, 10(12): 4762. doi: 10.3390/su10124762 [9] FAMILI A, SARASUA W A, OGLE J H, et al. GIS based spatial analysis of pedestrian crashes: a case study of South Carolina[C]//ASCE. International Conference on Transportation and Development. Reston: ASCE, 2018: 368-376. [10] TORAN POUR A, MORIDPOUR S, TAY R, et al. Influence of pedestrian age and gender on spatial and temporal distribution of pedestrian crashes[J]. Traffic Injury Prevention, 2018, 19(1): 81-87. doi: 10.1080/15389588.2017.1341630 [11] MA Qing-lu, HUANG Guang-hao, TANG Xiao-yao. GIS-based analysis of spatial-temporal correlations of urban traffic accidents[J]. European Transport Research Review, 2021, 13(1): 1-11. doi: 10.1186/s12544-020-00457-z [12] NASRI M, AGHABAYK K, ESMAILI A, et al. Using ordered and unordered logistic regressions to investigate risk factors associated with pedestrian crash injury severity in Victoria, Australia[J]. Journal of Safety Research, 2022, 81: 78-90. doi: 10.1016/j.jsr.2022.01.008 [13] KONG Chun-yu, YANG Ji-kuang. Logistic regression analysis of pedestrian casualty risk in passenger vehicle collisions in China[J]. Accident Analysisand Prevention, 2010, 42(4): 987-993. doi: 10.1016/j.aap.2009.11.006 [14] OLSZEWSKI P, SZAGAŁA P, WOLA AN'G SKI M, et al. Pedestrian fatality risk in accidents at unsignalized zebra crosswalks in Poland[J]. Accident Analysis and Prevention, 2015, 84: 83-91. doi: 10.1016/j.aap.2015.08.008 [15] HUSSAIN Q, FENG H Q, GRZEBIETA R, et al. The relationship between impact speed and the probability of pedestrian fatality during a vehicle-pedestrian crash: a systematic review and meta-analysis[J]. Accident Analysis and Prevention, 2019, 129: 241-249. doi: 10.1016/j.aap.2019.05.033 [16] ZENG Qiang, GU Wei-hua, ZHANG Xuan, et al. Analyzing freeway crash severity using a Bayesian spatial generalized ordered logit model with conditional autoregressive priors[J]. Accident Analysisand Prevention, 2019, 127: 87-95. doi: 10.1016/j.aap.2019.02.029 [17] HOU Qin-zhong, HUO Xiao-yan, LENG Jun-qiang, et al. Examination of driver injury severity in freeway single-vehicle crashes using a mixed logit model with heterogeneity-in-means[J]. Physica A: Statistical Mechanics and Its Applications, 2019, 531: 121760. doi: 10.1016/j.physa.2019.121760 [18] 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.SONG Dong-dong, YANG Xiao-bao, ZU Xing-shui, 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) [19] LI Y X, ABDEL-ATY M, YUAN J H, et al. Analyzing traffic violation behavior at urban intersections: a spatio-temporal kernel density estimation approach using automated enforcement system data[J]. Accident Analysis and Prevention, 2020, 141: 105509. doi: 10.1016/j.aap.2020.105509 [20] KATICHA S W, FLINTSCH G W. A kernel density empirical Bayes (KDEB) approach to estimate accident risk[J]. Accident Analysisand Prevention, 2023, 186: 107039. doi: 10.1016/j.aap.2023.107039 [21] BRUNSDON C, CORCORAN J, HIGGS G. Visualising space and time in crime patterns: a comparison of methods[J]. Computers, Environment and Urban Systems, 2007, 31(1): 52-75. doi: 10.1016/j.compenvurbsys.2005.07.009 [22] ROMANO B, JIANG Z. Visualizing traffic accident hotspots based on spatial-temporal network kernel density estimation[C]//ACM. Proceedings of the 25th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. New York: ACM, 2017: 7-10. [23] PAL C, HIRAYAMA S, NARAHARI S, et al. An insight of World Health Organization (WHO) accident database by cluster analysis with self-organizing map (SOM)[J]. Traffic Injury Prevention, 2018, 19(S1): S15-S20. http://smartsearch.nstl.gov.cn/paper_detail.html?id=1b1ff5c1ceaf0aaef2bcfb3e0c57f495 [24] 刘强, 严修, 谢谦, 等. 基于事故综合强度的公交事故严重程度分析[J]. 交通运输系统工程与信息, 2022, 22(6): 152-159.LIU Qiang, YAN Xiu, XIE Qian, et al. Bus accident severity analysis based on comprehensive accident intensity[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(6): 152-159. (in Chinese) [25] SUN Zhi-yuan, WANG Duo, GU Xin, et al. A hybrid approach of random forest and random parameters logit model of injury severity modeling of vulnerable road users involved crashes[J]. Accident Analysis and Prevention, 2023, 192: 107235. doi: 10.1016/j.aap.2023.107235 [26] SE C, CHAMPAHOM T, JOMNONKWAO S, et al. The impact of weekday, weekend, and holiday crashes on motorcyclist injury severities: accounting for temporal influence with unobserved effect and insights from out-of-sample prediction[J]. Analytic Methods in Accident Research, 2022, 36: 100240. doi: 10.1016/j.amar.2022.100240 [27] MECHANTE L F S, DE ARGILA LORENTE C M, LOPEZ-VALDES F. A pilot analysis of crash severity of electric passenger cars in Spain (2016-2020)[J]. Traffic Injury Prevention, 2022, 23(S1): S217-S219. doi: 10.1080/15389588.2022.2125230 [28] HARIRFOROUSH H, BELLALITE L. A new integrated GIS-based analysis to detect hotspots: a case study of the city of Sherbrooke[J]. Accident Analysisand Prevention, 2019, 130: 62-74. doi: 10.1016/j.aap.2016.08.015 [29] BÍL M, ANDRÁŠIK R, SEDONÍK J. A detailed spatiotemporal analysis of traffic crash hotspots[J]. Applied Geography, 2019, 107: 82-90. doi: 10.1016/j.apgeog.2019.04.008 [30] 周备, 孙晴, 张生瑞. 考虑时间不稳定性的货车事故严重程度分析[J]. 中国安全科学学报, 2022, 32(11): 160-167.ZHOU Bei, SUN Qing, ZHANG Sheng-rui. Severity analysis of freight car accidents considering time instability[J]. China Safety Science Journal, 2022, 32(11): 160-167. (in Chinese) -
图(8) / 表(4)
计量
- 文章访问数: 1621
- HTML全文浏览量: 534
- PDF下载量: 128
- 被引次数: 0
