An Analysis of Severity of Traffic Accidents on Urban Roadways Based on Binary Logistic Models
-
摘要: 为从多维度精准剖析影响城市道路交通事故严重程度的因素，选取了我国某城市2018—2020年交通事故数据库中的4 587条数据作为研究对象，基于二元Logistic模型，从人、车、路、环境这4个方面，分别针对财产损失事故、伤人事故、死亡事故建立了模型。深入分析了道路物理隔离位置、路侧防护设施类型等因素对事故严重程度的影响，并利用Hosmer-Lemeshow检验和一致性检验对模型有效性进行验证。结果表明：①道路物理隔离的空间位置对事故严重程度有显著影响，仅布设中心隔离设施发生死亡事故的概率是同时布设中央和机非隔离的2.304倍。在有中心隔离设施的高等级道路中，增设机非隔离设施能有效降低事故发生的概率。②路侧防护设施类型为行道树、绿化带时，发生死亡事故的概率分别是金属护栏的1.982倍、1.648倍。与金属护栏相比，行道树更容易引发严重事故。③夜间无路灯照明发生死亡事故的概率是夜间有路灯照明的1.808倍，夜间无路灯照明是导致死亡事故的重要因素之一。④受过高等教育的驾驶人发生财产损失事故和伤人事故的概率较高，受过中等教育的驾驶员发生死亡事故的概率较高；受过中等教育驾驶员发生死亡事故的概率是高等教育驾驶员的2.049倍。研究深入分析了影响城市道路交通事故的显著因素及其对事故的影响，为事故严重程度的精细化分析提供了理论支持，为交通规划与管理部门提供了决策依据。
-
关键词:
- 交通安全 /
- 事故影响因素 /
- 事故严重程度 /
- 二元Logistic回归分析
Abstract: In order to accurately identify the factors affecting the severity of traffic accidents on urban roadways, 4 587 records from the traffic accident database of a city in China from Year 2018 to 2020 are used. A series of binary logistic models are developed for property damage accidents, personal-injury accidents, and fatal accidents based on the data regarding the following four aspects, including participants, vehicles, roads, and environment. The impacts of location of dividing strips and the types of roadside protection facilities on the severity of accidents are analyzed. Further, the Hosmer-Lemeshow tests and the consistency tests are used to check the soundness of the models. The results from the binary Logistic models show that: ① locations of dividing strips have a significant effect on the severity of the accidents. The probability of fatal accidents of placing central dividing strips only is 2.304 times higher than placing both central and motor/non-motor dividing strips. On high-grade roads with central dividing strips, adding motorized/non-motorized dividing strips can effectively reduce the probability of accidents. ② The probability of fatal accidents for the roadside protection facilities being street trees and green belts is 1.982 times and 1.648 times higher than the roadside protection facilities being metal guardrails, respectively. Street trees as the type of roadside protection facilities are likely to lead to more serious accidents than metal guardrails. ③The probability of fatal accidents at night "with no streetlight" is 1.808 times higher than that "with streetlights" at night. No streetlight at night is a significant risk factor leading to the fatal accidents. ④ Drivers with a high-level of education are more likely to have property-damage accidents and personal-injury accidents. Fatal accidents are more likely to occur among drivers with a medium-level of education, and their probability of resulting in the fatal accidents is 2.049 times higher than drivers with a high-level of education. In conclusion, this study presents an analysis of the factors and their impacts on traffic accidents at urban roadways. Furthermore, it can serve as a theoretical support for the refined analysis of accident severity on urban roadways, as well as a reference for traffic safety planning and management authorities. -
图 2 不同严重程度事故发生的验证概率
Figure 2. Probability of the occurrence of accidents of different severity
表 1 因变量描述性统计与赋值
Table 1. Descriptive statistics and assignment of dependent variables
变量类型 变量名称 变量频数 变量赋值 事故严重程度 财产损失事故 666 发生财产损失事故：Y=1;不发生财产损失事故：Y=0 伤人事故 3 375 发生伤人事故：Y=1;不发生伤人事故：Y=0 死亡事故 545 发生死亡事故：Y=1;不发生死亡事故：Y=0 
下载: 导出CSV
表 2 自变量描述性统计与赋值
Table 2. Descriptive statistics and assignment of independent variables
变量类型 变量名称与类别 变量统计 赋值 道路条件 道路物理隔离 无隔离 2 499 1 机非隔离 297 2 中心隔离加机非隔离 677 3 中心隔离 1 113 4 路侧防护设施类型 无防护 1 916 1 绿化带 1 574 2 行道树 839 3 金属护栏 257 4 路口路段类型 路段 2 676 1 交叉口 1 910 2 交通信号方式 无控制 1 616 1 有控制 2 970 2 环境 天气 晴 3 830 1 阴 469 2 雨 251 3 雪 21 4 雾 15 5 照明条件 白天 2 919 1 夜间有路灯照明 1 093 2 夜间无路灯照明 574 3 能见度/m <50 300 1 50~100 726 2 101~200 1 009 3 >200 2 549 4 驾驶人 性别 女 4 034 1 男 552 2 受教育程度 高等教育 1 696 1 中等教育 2 890 2 年龄/岁 ≤25 566 1 >25~35 1 581 2 >35~45 1 326 3 >45~55 960 4 >55 154 5 驾龄/年 ≤2 425 1 >2~5 887 2 >5~10 1 433 3 >10~15 880 4 >15 962 5 车辆 行驶状态 直行 3 731 1 非直行 856 2
下载: 导出CSV
表 3 共线性检验结果
Table 3. Collinearity test result
变量名称 容忍度 VIF 道路物理隔离 0.87 1.15 路侧防护设施类型 0.89 1.12 路口路段类型 0.97 1.04 交通信号方式 0.95 1.05 天气 0.99 1.02 照明条件 0.8 1.25 能见度 0.81 1.24 性别 0.96 1.05 驾龄 0.7 1.44 年龄 0.70 1.45 受教育程度 0.94 1.07 行驶状态 0.96 1.04
下载: 导出CSV
表 4 模型标定结果
Table 4. Model calibration results
财产损失事故 伤人事故 死亡事故 影响因素 B 影响因素 B 影响因素 B 路口路段类型(1=路段) 0.252 道路物理隔离 道路物理隔离 交通信号方式(1=无控制) 0.329 道路物理隔离(1=无隔离) 0.235 道路物理隔离(1=无隔离) -0.406 照明条件 道路物理隔离(2=机非隔离) 0.193 道路物理隔离(2=机非隔离) -0.566 照明条件(1=白天) -0.308 道路物理隔离(3=中心隔离加机非隔离) 0.288 道路物理隔离(3=中心隔离加机非隔离) -0.835 照明条件(2=夜间有路灯照明) 1.078 路口路段类型(1=路段) -0.288 路侧防护设施类型 年龄/岁 照明条件 路侧防护设施类型(1=无防护) 0.128 年龄(1= <25) 0.509 照明条件(1=白天) 0.623 路侧防护设施类型(2=绿化带) 0.499 年龄(2= >25~35) 0.798 照明条件(2=夜间有路灯照明) -0.394 路侧防护设施类型(3=行道树) 0.684 年龄(3= >35~45) 0.438 受教育程度(1=高等教育) 0.178 路口路段类型(1=路段) 0.245 年龄(4= >45~55) 0.439 行驶状态(1=直行) -0.446 交通信号方式(1=无控制) -0.356 受教育程度(1=高等教育) 0.269 常量 1.065 照明条件 行驶状态(1=直行) 0.451 照明条件(1=白天) -0.716 常量 -3.306 照明条件(2=夜间有路灯照明) -0.592 受教育程度(1=高等教育) -0.718 常量 -1.246
下载: 导出CSV
表 5 财产损失事故与影响因素的二元Logistic回归分析结果
Table 5. Results of binary Logistic regression analysis of property loss accidents and influencing factors
影响因素 B 标准误差 瓦尔德 自由度 显著性 exp(B) 路口路段类型1=路段 0.252 0.097 6.727 1 0.009 1.286 交通信号方式1=无控制 0.329 0.098 11.277 1 0.001 1.39 193.205 2 <0.001 照明条件 1=白天 -0.308 0.151 4.182 1 0.041 0.735 2=夜间有路灯照明 1.078 0.154 48.758 1 <0.001 2.939 16.749 4 0.002 1=≤25 0.509 0.345 2.171 1 0.141 1.663 年龄/岁 2= >25~35 0.798 0.328 5.908 1 0.015 2.222 3= >35~45 0.438 0.333 1.729 1 0.189 1.55 4= >45~55 0.439 0.339 1.68 1 0.195 1.551 受教育程度 1=高等教育 0.269 0.096 7.808 1 0.005 1.308 行驶状态 1=直行 0.451 0.137 10.864 1 0.001 1.569 常量 -3.306 0.375 77.583 1 <0.001 0.037
下载: 导出CSV
表 6 伤人事故与影响因素的二元Logistic回归分析结果
Table 6. Results of binary Logistic regression analysis of injury accidents and influencing factors
影响因素 B 标准误差 瓦尔德 自由度 显著性 exp (B) 8.832 3 0.032 1=无隔离 0.235 0.087 7.285 1 0.007 1.265 道路物理隔离 2=机非隔离 0.193 0.165 1.368 1 0.242 1.213 3=中心隔离加机非隔离 0.288 0.122 5.593 1 0.018 1.334 路口路段类型(1=路段) -0.288 0.076 14.395 1 <0.001 0.75 155.617 2 <0.001 照明条件 1=白天 0.623 0.107 33.755 1 <0.001 1.865 2=夜间有路灯照明 -0.394 0.119 10.973 1 0.001 0.674 受教育程度 1=高等教育 0.178 0.078 5.229 1 0.022 1.195 行驶状态 1=直行 -0.446 0.104 18.481 1 <0.001 0.64 常量 1.065 0.152 49.227 1 <0.001 2.901
下载: 导出CSV
表 7 死亡事故与影响因素的二元Logistic回归分析结果
Table 7. Results of binary Logistic regression analysis of fatal accident and influencing factors
影响因素 B 标准误差 瓦尔德 自由度 显著性 exp(B) 26.983 3 <0.001 1=无隔离 -0.406 0.121 11.289 1 0.001 0.667 道路物理隔离 2=机非隔离 -0.566 0.231 5.987 1 0.014 0.568 3=中心隔离加机非隔离 -0.835 0.183 20.867 1 <0.001 0.434 22.393 3 <0.001 路侧防护设施类型 1=无防护 0.128 0.259 0.243 1 0.622 1.136 2=绿化带 0.499 0.249 4.031 1 0.045 1.648 3=行道树 0.684 0.26 6.92 1 0.009 1.982 路口路段类型 1=路段 0.245 0.102 5.777 1 0.016 1.278 交通信号方式 1=无控制 -0.356 0.108 10.78 1 0.001 0.7 30.707 2 <0.001 照明条件 1=白天 -0.716 0.129 30.621 1 <0.001 0.489 2=夜间有路灯照明 -0.592 0.157 14.229 1 <0.001 0.553 受教育程度 1=高等教育 -0.718 0.116 38.392 1 <0.001 0.488 常量 -1.246 0.265 22.159 1 <0.001 0.288
下载: 导出CSV
表 9 肯德尔(W)检验
Table 9. Kendall(W)test
事故类型 肯德尔(W) 渐进显著性 财产损失事故 0.797 0.144 伤人事故 0.900 0.126 死亡事故 0.910 0.122
下载: 导出CSV
-
[1] 付锐, 郭应时, 袁伟, 等. 连续下坡道路事故率与纵面参数关系研究[J]. 中国公路学报, 2009, 22(3): 101-106. doi: 10.3321/j.issn:1001-7372.2009.03.018FU R, GUO Y S, YUAN W, et al. Study on the relationship between accident rate and longitudinal parameters of continuous downhill road[J]. China Journal of Highway and Transport, 2009, 22(3): 101-106. (in Chinese) doi: 10.3321/j.issn:1001-7372.2009.03.018 [2] HU W, DONNELL E T. Median barrier crash severity: Some new insights[J]. Accident Analysis & Prevention, 2010, 42 (6): 1697-1704. [3] 白玉, 魏毅. 重大道路交通事故严重程度影响因素分析[J]. 交通与运输, 2022, 38(3): 22-26. https://www.cnki.com.cn/Article/CJFDTOTAL-YSJT202203005.htmBAI Y, WEI Y. Influencing factors on severity of serious road tra ffi c accidents[J]. Traffic & Transportation, 2022, 38(3): 22-26. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSJT202203005.htm [4] 马符铭, 刘浩. 降雪天气对快速路交通特征的影响及对策[J]. 交通科技与经济, 2016, 18(4): 53-56. https://www.cnki.com.cn/Article/CJFDTOTAL-KJJJ201604016.htmMA F M, LIU H. Influence of snowfall on traffic characteristics of expressway and countermeasures[J]. Technology & Economy in Areas of Communications, 2016, 18(4): 53-56. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KJJJ201604016.htm [5] 冯忠祥, 雷叶维, 张卫华, 等. 道路环境对绕城高速公路交通事故严重程度影响分析[J]. 中国公路学报, 2016, 29(5): 116-123. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201605016.htmFENG Z X, LEI Y W, ZHANG W H, et al. Analysis of the influence of road environment on the severity of traffic accidents on ring expressway[J]. China Journal of Highway and Transport, 2016, 29(5): 116-123. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201605016.htm [6] 赵跃峰, 张生瑞, 马壮林. 基于部分优势比的公路隧道交通事故严重程度分析模型[J]. 中国公路学报, 2018, 31(9): 159-166. doi: 10.3969/j.issn.1001-7372.2018.09.018ZHAO Y F, ZHANG S R, MA Z L. Analysis model of traffic accident severity in highway tunnel based on partial advantage ratio[J]. China Journal of Highway and Transport, 2018, 31(9): 159-166. (in Chinese) doi: 10.3969/j.issn.1001-7372.2018.09.018 [7] 王磊, 吕璞, 林永杰. 高速公路交通事故影响因素分析及伤害估计[J]. 中国安全科学学报, 2016, 26(3): 86-90. https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201603016.htmWANG L, LYU P, LIN Y J. Influencing factors analysis and injury estimation of expressway traffic accidents[J]. China Safety Science Journal, 2016, 26(3): 86-90. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201603016.htm [8] 张圆, 邓院昌, 林庆丰, 等. 不同性别驾驶员小轿车交通事故的影响因素分析[J]. 安全与环境工程, 2019, 26(3): 166-170. https://www.cnki.com.cn/Article/CJFDTOTAL-KTAQ201903025.htmZHANG Y, DENG Y C, LIN Q F, et al. Influencing factors of car traffic accidents among different sex drivers[J]. Safety and Environmental Engineering, 2019, 26(3): 166-170. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KTAQ201903025.htm [9] 由冰玉, 廉福绵, 孟祥海. 基于故障树贝叶斯网的山区高速公路事故成因分析[J]. 交通信息与安全, 2019, 37(4): 44-51. doi: 10.3963/j.issn.1674-4861.2019.04.006YOU B Y, LIAN F M, MENG X H. Cause analysis of mountain expressway accidents based on fault tree bayesian network[J]. Journal of Transport Information and Safety, 2019, 37(4): 44-51. (in Chinese) doi: 10.3963/j.issn.1674-4861.2019.04.006 [10] BENLAGHA N, CHARFEDDINE L. Risk factors of road accident severity and the development of a new system for prevention: New insights from China[J]. Accident Analysis & Prevention, 2020(136): 105411. [11] 孙轶轩, 邵春福, 岳昊, 等. 基于SVM灵敏度的城市交通事故严重程度影响因素分析[J]. 吉林大学学报(工学版), 2014, 44(5): 1315-1320. https://www.cnki.com.cn/Article/CJFDTOTAL-JLGY201405015.htmSUN Y X, SHAO C F, YUE H, et al. Analysis of influencing factors of urban traffic accident severity based on SVM sensitivity[J]. Journal of Jilin University(Engineering and Technology Edition), 2014, 44(5): 1315-1320. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JLGY201405015.htm [12] 吕通通, 张湛, 陆林军, 等. 基于互信息贝叶斯网络的交通事故严重程度分析[J]. 交通信息与安全, 2021, 39(6): 36-43. doi: 10.3963/j.jssn.1674-4861.2021.06.005LYU T T, ZHANG Z, LU L J, et al. The analysis of traffic accident severity based on mutual information bayesian network[J]. Journal of Transport Information and Safety, 2021, 39(6): 36-43. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2021.06.005 [13] 胡立伟, 薛刚, 李林育, 等. 高原地质及气象环境下公路交通风险致因耦合分析[J]. 中国公路学报, 2018, 31(1): 110-119. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201801014.htmHU L W, XUE G, LI L Y, et al. Coupling analysis of highway traffic risk causes in plateau and meteorological environment[J]. China Journal of Highway and Transport, 2018, 31 (1): 110-119. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201801014.htm [14] SAVOLAINEN P T, MANNERING F L, LORD D, et al. The statistical analysis of highway crash-injury severities: A review and assessment of methodological alternatives[J]. Accident Analysis & Prevention, 2011, 43(5): 1666-1676. [15] 马壮林, 邵春福, 董春娇, 等. 基于累积Logistic模型的交通事故严重程度时空分析[J]. 中国安全科学学报, 2011, 21 (9): 94-99. https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201109017.htmMA Z L, SHAO C F, DONG C J, et al. Temporal and spatial analysis of traffic accident severity based on cumulative Logistic model[J]. China Safety Science Journal, 2011, 21(9): 94-99. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201109017.htm [16] 马壮林, 张祎祎, 杨杨, 等. 公路隧道交通事故严重程度预测模型研究[J]. 中国安全科学学报, 2015, 25(5): 75-79. https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201505013.htmMA Z L, ZHANG Y Y, YANG Y, et al. Research on prediction model of severity of highway traffic accidents[J]. China Safety Science Journal, 2015, 25(5): 75-79. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201505013.htm [17] STIPANCIC J, ZANGENEHPOUR S, MIRANDA-MORENO L, et al. Investigating the gender differences on bicycle-vehicle conflicts at urban intersections using an ordered logit methodology[J]. Accident Analysis & Prevention, 2016 (97): 19-27. [18] OLSZEWSKI P, SZAGALA P, WOLANSKI M, et al. Pedestrian fatality risk in accidents at unsignalized zebra crosswalks in Poland[J]. Accident Analysis & Prevention, 2015 (84): 83-91. [19] FENG S M, LI Z N, CI Y S, et al. Risk factors affecting fatal bus accident severity: Their impact on different types of bus drivers[J]. AccidentAnalysis & Prevention, 2016, 86(1): 29-39. [20] JALAYER M, SHABANPOUR R, POUR-ROUHOLAMIN M, et al. Wrong-way driving crashes: A random-parameters ordered probit analysis of injury severity[J]. Accident Analysis & Prevention, 2018(117): 128-135. [21] 林庆丰, 邓院昌, 胡继华. 机非交通事故中驾驶员过错和事故严重程度影响因素的Logistic回归分析[J]. 安全与环境工程, 2019, 26(5): 187-193. https://www.cnki.com.cn/Article/CJFDTOTAL-KTAQ201905028.htmLIN Q F, DENG Y C, HU J H. Logistic regression analysis of driver's fault and accident severity in motor-bicycle accidents[J]. Safety and Environmental Engineering, 2019, 26 (5): 187-193. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-KTAQ201905028.htm -
点击查看大图
图(2) / 表(9)
计量
- 文章访问数: 991
- HTML全文浏览量: 360
- PDF下载量: 80
- 被引次数: 0