以京港澳高速公路（G4）粤境北段3年发生的1 354起交通事故为研究对象，将基础数据根据路段长度一致、曲线半径一致和坡度一致划分路段单元对基础数据进行处理，从道路线形和环境条件2个方面选取13个自变量，分别采用负二项（Negative Binomial，NB）回归模型和非线性负二项（Nonlinear Negative Binomial，NNB）回归模型建立交通事故起数预测模型，根据模型的拟合优度和预测准确性对比分析负二项回归和非线性负二项回归模型的优劣，并找出影响交通事故起数的显著自变量，分析显著自变量对交通事故起数的影响程度。研究结果表明：无论采用上述何种路段划分方法，非线性负二项回归模型构建的交通事故起数预测模型均优于负二项回归模型；采用坡度一致划分方法明显优于路段长度一致和曲线半径一致划分方法，更适合应用于山区高速公路交通事故数预测研究；从显著变量相关性来看，路段长度、相邻路段坡度变化值、弯坡组合、曲率、是否存在隧道路段以及是否为易结冰和起雾路段均是非线性模型的显著影响因素。
Abstract
In this paper, the significant influencing factors of expressway traffic accidents were identified, and the relationship between traffic accident frequency and influencing factors such as road geometry and environmental conditions on expressways were analyzed. Accordingly, 1354 accidents that occurred on the Beijing-Hong Kong-Macau Expressway within the northern part of the Guangdong Province in a three-year period were evaluated. Three types of segmentation methods were used to divide the study section:fixed-length, homogeneous horizontal radius, and homogeneous longitudinal grade. Thirteen independent variables were selected from the road geometry and environmental conditions. Further, an accident frequency prediction-model was established using negative binomial (NB) and nonlinear negative binomial (NNB) regression models to explore the significant influencing factors. Elastic analysis was used to determine the degree of influence of the independent variables. The results indicate that the accident frequency prediction model based on the NNB regression model is better than that of the NB regression model for the fixed-length, homogeneous horizontal radius, and homogeneous longitudinal grade segmentation methods. Additionally, homogeneous longitudinal grade segmentation methods are better than the fixed-length and homogeneous horizontal radius methods for studying mountain expressways. From the correlation of significant variables, it was found that the length of section, change value of slope of adjacent sections, combination of curves and slopes, curvature, existence of tunnel sections, and whether the sections are susceptible to freezing and fog are the significant factors of the nonlinear model.
关键词
交通工程 /
交通事故预测 /
负二项回归模型 /
高速公路 /
非线性负二项回归模型
{{custom_keyword}} /
Key words
traffic engineering /
traffic accident frequency prediction /
negative binomial regression model /
expressway /
nonlinear negative binomial regression model
{{custom_keyword}} /
中图分类号：
U491.31
{{custom_clc.code}}
({{custom_clc.text}})
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 公安部交通管理局.中华人民共和国道路交通事故统计年报(2016年度)[R].北京:公安部交通管理局,2017.Traffic Management Bureau of the Ministry of Public Security. Road Traffic Accident Statistics' Annual Report of the People's Republic of China (2016)[R]. Beijing:Traffic Management Bureau of the Ministry of Public Security, 2017.
[2] KRAUS J F, ANDERSON C L, ARZEMANIAN S, et al. Epidemiological Aspects of Fatal and Severe Injury Urban Freeway Crashes[J]. Accident Analysis & Prevention, 1993, 25(3):229-239.
[3] 高建平.成渝高速公路重庆段安全事故分析研究[J].重庆交通学院学报,2003,22(3):74-79.GAO Jian-ping. The Study of Traffic Crashes of Cheng-yu Expressway in Chongqing Area[J]. Journal of Chongqing Jiaotong University, 2003, 22(3):74-79.
[4] 赵新勇.基于多源异构数据的高速公路交通安全评估方法[D].哈尔滨:哈尔滨工业大学,2013.ZHAO Xin-yong. Freeway Traffic Safety Evaluation Method Based on Multi-source Heterogeneous Data[D]. Harbin:Harbin Institute of Technology, 2013.
[5] MIAOU S P. The Relationship Between Truck Accidents and Geometric Design of Road Sections:Poisson Versus Negative Binomial Regressions[J]. Accident Analysis & Prevention, 1994, 26(4):471-482.
[6] CAFISO S, DI GRAZIANO A, DI SILVESTRO G, et al. Development of Comprehensive Accident Models for Two-lane Rural Highways Using Exposure, Geometry, Consistency and Context Variables[J]. Accident Analysis & Prevention, 2010, 42(4):1072-1079.
[7] MAHER M J, SUMMERSGILL I. A Comprehensive Methodology for the Fitting of Predictive Accident Models[J]. Accident Analysis & Prevention, 1996,28(3):281-296.
[8] JOVANIS P P, CHANG H L. Modeling the Relationship of Accidents to Miles Traveled[J]. Transportation Research Record, 1986(1068):42-51.
[9] MIAOU S P, LUM H. Modeling Vehicle Accidents and Highway Geometric Design Relationships[J]. Accident Analysis & Prevention, 1993, 25(6):689-709.
[10] DANIEL J, CHIEN S I J. Truck Safety Factors on Urban Arterials[J]. Journal of Transportation Engineering, 2004, 130(6):742-752.
[11] GEEDIPALLY S R, LORD D, DHAVALA S S. The Negative-binomial-lindley Generalized Linear Model:Characteristics and Application Using Crash Data[J]. Accident Analysis & Prevention, 2012, 45(2):258-265.
[12] LAO Y, ZHANG G, WANG Y, et al. Generalized Nonlinear Models for Rear-end Crash Risk Analysis[J]. Accident Analysis & Prevention, 2014, 62(5):9-16.
[13] WONG S C, SZE N N, LI Y C. Contributory Factors to Traffic Crashes at Signalized Intersections in Hong Kong[J]. Accident Analysis & Prevention, 2007, 39(6):1107-1113.
[14] ABDEL-ATY M, HALEEM K. Analyzing Angle Crashes at Unsignalized Intersections Using Machine Learning Techniques[J]. Accident Analysis & Prevention, 2011, 43(1):461-470.
[15] OKAMOTO H, KOSHI M.A Method to Cope With the Random Errors of Observed Accident Rates In Regression Analysis[J]. Accident Analysis & Prevention, 1989, 21(4):317-332.
[16] 罗石贵,周伟.道路交通事故多发点鉴定方法探讨[J].西安公路交通大学学报,1999,19(1):30-33.LUO Shi-gui, ZHOU Wei. Research on the Identification of Road Accident Blackspots[J]. Journal of Xi'an Highway University, 1999, 19(1):30-33.
[17] HAUER E, KONONOV J, ALLERY B, et al. Screening the Road Network for Sites with Promise[J]. Transportation Research Record, 2002(1784):27-32.
[18] ZHANG C, IVAN J. Effects of Geometric Characteristics on Head-on Crash Incidence on Two-lane Roads in Connecticut[J]. Transportation Research Record, 2005(1908):159-164.
[19] MAYROA J M P. Relevant Variables for Crash-rate Prediction on Spain's Two-lane Rural Roads[C]//TRB. Proceedings of the Transportation Research Board 82nd Annual Meeting. Washington DC:TRB, 2003:120-132.
[20] 《中国公路学报》编辑部. 中国交通工程学术研究综述·2016[J]. 中国公路学报, 2016, 29(6):1-161.Editorial Department of China Journal of Highway and Transport. Review on China's Traffic Engineering Research Progress:2016[J]. China Journal of Highway and Transport, 2016, 29(6):1-161.
[21] 王雪松,宋洋,黄合来,等.基于分层负二项模型的城郊公路安全影响因素研究[J].中国公路学报,2014,27(1):100-106.WANG Xue-song, SONG Yang, HUANG He-lai, et al. Analysis of Risk Factors for Suburban Highways Using Hierarchical Negative Binomial Model[J]. China Journal of Highway and Transport, 2014, 27(1):100-106.
[22] 郭璘,周继彪,董升,等.基于改进K-means算法的城市道路交通事故分析[J].中国公路学报,2018,31(4):270-279.GUO Lin, ZHOU Ji-biao, DONG Sheng, et al. Analysis of Urban Road Traffic Accidents Based on Improved K-means Algorithm[J]. China Journal of Highway and Transport, 2018, 31(4):270-279.
[23] 高珍,高屹,余荣杰,等.连续数据环境下的道路交通事故风险预测模型[J].中国公路学报,2014,31(4):280-287.GAO Zhen, GAO Yi, YU Rong-jie, et al. Road Crash Risk Prediction Model for Continuous Streaming Data Environment[J]. China Journal of Highway and Transport, 2014, 31(4):280-287.
[24] ABDEL-ATY M A, RADWAN A E. Modeling Traffic Accident Occurrence and Involvement[J]. Accident Analysis & Prevention, 2000, 32(5):633-642.
[25] 张祎祎.基于改进负二项回归模型的高速公路交通事故起数预测方法研究[D]. 西安:长安大学,2017.ZHANG Yi-yi. Research on Expressway Traffic Accident Frequency Prediction Method Based on Improved Negative Binomial Model[D]. Xi'an:Chang'an University, 2017.
[26] AKAIKE H. Information Theory and an Extension of the Maximum Likelihood Principle[M]. Berlin:Springer, 1998.
[27] SCHWARZ G E. Estimating the Dimension of a Model[J]. The Annals of Statistic, 1978, 6(2):461-464.
[28] HIBE J M. Negative Binomial Regression[M]. Cambridge:Cambridge University Press, 2011.
[29] HAUER E, COUNCIL F, MOHAMMEDSHAH Y. Safety Models for Urban Four-lane Undivided Road Segments[J]. Transportation Research Record, 2004(1897):96-105.
[30] MA Z, ZHANG H, CHIEN S, et al. Predicting Expressway Crash Frequency Using a Random Effect Negative Binomial Model:A Case Study in China[J]. Accident Analysis & Prevention, 2017, 98:214-222.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金项目（51208052）；陕西省自然科学基础研究计划项目（2017JM5084）；教育部人文社会科学研究青年基金项目（18YJCZH130）；中央高校基本科研业务费专项资金项目（300102228202）
{{custom_fund}}
{{custom_fund}}
PDF全文下载(1105 KB)
