吉林大学学报(工学版) ›› 2015, Vol. 45 ›› Issue (6): 1771-1778.doi: 10.13229/j.cnki.jdxbgxb201606006

• • 上一篇    下一篇

面板数据模型分析及交通事故预测

孙璐1，2， 徐建1，3， 崔相民4   

  1. 1.东南大学 交通学院,南京 210096;
    2.美国天主教大学 土木工程系,华盛顿 20064;
    3.美国德克萨斯州大学奥斯汀分校 交通研究中心,奥斯汀78712;
    4.山东省公路检测中心,山东 济宁 272000
  • 收稿日期:2013-10-08 出版日期:2015-11-01 发布日期:2015-11-01
  • 作者简介:孙璐(1972-),男,教授,博士生导师.研究方向:交通工程,道路工程.E-mail:sunl@cua.edu
  • 基金资助:
    国家自然科学基金重点项目(U1134206); 国家自然科学基金外青学者项目(51250110075); 国家道路安全科技行动计划项目(2009BAG13A02); 交通运输部西部项目(0901005C); 美国国家科学基金总统奖项目(CMMI-0408390); 美国国家科学基金项目(CMMI-0644552)

Panel data models for analysis and prediction of crash count

SUN Lu1,2, XU Jian1,3, CUI Xiang-min4   

  1. 1.School of Transportation, Southeast University, Nanjing 210096, China;
    2.Department of Civil Engineering, Catholic University of America, Washington DC 20064, USA;
    3.Center for Transportation Research, University of Texas at Austin, Austin 78712, USA;
    4.Shandong Provincial Highway Detection Center, Jining 272000, China
  • Received:2013-10-08 Online:2015-11-01 Published:2015-11-01

RICH HTML

0   

PDF (PC)

899

摘要： 构建了一种基于面板数据的空间非集计模型——负二项面板模型,包括混合效应、固定效应和随机效应3种类型,以同性质路段为研究单元,选取道路线形、交通特性、土地利用和降雨量等影响因素,利用事故率比例指标IRR,分析和预测未受伤事故、受伤事故、死亡事故和事故总数等4种类型事故,并通过F检验和Hausman检验以及对数似然值和离差信息准则DIC,对比分析3种类型模型的拟合效果。发现协变量对各类型交通事故的影响作用和统计显著性不尽相同,如限速每增加1.609 km/h(1 mile/h),未受伤事故、受伤事故、事故总数分别减少3.89%、2.24%和2.79%,而死亡事故增加6.38%。研究结果表明:负二项面板固定效应模型比混合模型和随机效应模型更优,另外越严重的事故,模型拟合效果越好。

关键词: 交通运输安全工程, 负二项面板模型, 固定效应模型, 土地利用, 降雨量

Abstract: Spatially disaggregated panel data models using negative binomial process, including pooled model, fixed-effects model and random-effect model, were developed. The contributing factors, such as roadway geometrics, traffic characteristics, land usage and rainfall were collected. Incidence Rate Ratio (IRR) and homogeneous segments were employed for analysis and prediction of crash count, which contains Property-Damage-Only (PDO) crashes, injury crashes, fatal crashes, and total crashes. Meanwhile, F test and Hausman test, log-likelihood value and Deviance Information Criterion (DIC) were used to compare the performance of the three types of models. It is found that the effects and statistical significance of covariates on the four types of crashes are not the same. For example, if the speed limit is increased by 1.609 km/h, the PDO rate, injury crash rate and total crash rate are reduced by 3.89%, 2.24% and 2.79%, respectively; however, the fatal crash rate is increased by 6.38%. Results show that the fixed-effects panel negative binomial model surpasses the pooled model and random-effect model, and the modeling is even better for crash count associated with more severe injuries and fatalities.

Key words: traffic and transportation safety engineering, panel negative binomial model, fixed-effects model, land use, rainfall

中图分类号: 

  • U491.31
[1] 马壮林,邵春福,李霞. 基于Logistic模型的公路隧道交通事故严重程度的影响因素[J]. 吉林大学学报: 工学版,2010,40(2):423-426.
Ma Zhuang-lin, Shao Chun-fu, Li Xia. Analysis of factors affecting accident severity in highway tunnels based on logistic model[J]. Journal of Jilin University(Engineering and Technology Edition),2010,40(2):423-426.
[2] 李铁洪, 吴华金. 长直线接小半径曲线公路交通事故成因及预防对策[J]. 中国公路学报,2007,20(1):35-40.
Li Tie-hong, Wu Hua-jin. Causes and countermeasures of highway traffic accidents in long straight line combined with sharp curve[J]. China Journal of Highway and Transport,2007,20(1):35-40.
[3] 黄合来,邓雪,许鹏鹏. 考虑空间自相关的贝叶斯事故预测模型[J]. 同济大学学报:自然科学版,2013,41(9):1378-1383.
Huang He-lai,Deng Xue, Xu Peng-peng. Bayesian crash prediction model based on a consideration of spatial autocorrelation[J]. Journal of Tongji University (Natural Science),2013,41(9):1378-1383.
[4] Jones A P,Haynes R,Kennedy V, et al. Geographical variations in mortality and morbidity from road traffic accidents in England and Wales[J]. Health & Place,2008,14(3):519-535.
[5] Noland R B,Quddus M A. A spatially disaggregate analysis of road casualties in England[J]. Accident Analysis & Prevention,2004,36(6):973-984.
[6] Wang Y,Kockelman K M. A conditional-autoregressive count model for pedestrian crashes across neighborhoods[R]. The 92nd Annual Meeting of the Transportation Research Board, 2013.
[7] Ma J M,Kockelman K M,Boothe C. Bayesian multivariate Poisson regression for models of injury count, by severity[J]. Transportation Research Record,2006,1950(1):24-34.
[8] Abdel-Aty M, Radwan A E. Modeling traffic accident occurrence and involvement[J]. Accident Analysis & Prevention,2000,32(5):633-642.
[9] Wang C, Quddus M A, Ison S G. Predicting accident frequency at their severity levels and its application in site ranking using a two-stage mixed multivariate model[J]. Accident Analysis & Prevention,2011,43(6):1979-1990.
[10] Ma J M,Kockelman K M,Damien P. A multivariate Poisson-lognormal regression model for prediction of crash counts by severity, using Bayesian methods[J]. Accident Analysis & Prevention,2008,40(3):964-975.
[11] Lord D. The prediction of accidents on digital networks: characteristics and issues related to the application of accident prediction models[D]. Toronto: University of Toronto,2000.
[12] Park B J,Lord D. Application of finite mixture models for vehicle crash data analysis[J]. Accident Analysis & Prevention,2009,41(4):683-691.
[13] Lambert D. Zero-inflated Poisson regression with an application to defects in manufacturing[J]. Technometrics,1992,34(1):1-14.
[14] Qin X,Reyes P. Conditional quantile analysis for crash count data[J]. Journal of Transportation Engineering,2011,137(9):601-607.
[15] 王军雷,孙小端,贺玉龙,等. 交通事故宏观计量经济学模型[J]. 交通运输工程学报,2012,12(2):70-75.
Wang Jun-lei, Sun Xiao-duan, He Yu-long, et al. Macroscopic econometrics model of traffic accident[J]. Journal of Traffic and Transportation Engineering,2012,12(2):70-75.
[16] 徐婷, 孙小端, 王伟力,等. 基于Panel Data的高速公路事故预测模型[J]. 北京工业大学学报,2010,36(4):495-499.
Xu Ting,Sun Xiao-duan,Wang wei-li, et al. Highway accidents statistical analysis with panel data model[J]. Journal of Beijing University of Technology,2010,36(4):495-499.
[17] Kweon Y J,Kockelman K M. Spatially disaggregate panel models of crash and injury counts: the effect of speed limits and design[R]. Transportation Research Board 83rd Annual Meeting,2004.
[18] Openshaw S. The modifiable areal unit problem[J]. Concepts and Techniques in Modern Geography, 1984.
[19] Hausman J,Hall B,Griliches Z. Econometric models for count data with an application to the patents-R&D relationship[J]. Econometrica,1984,52(4):909-938.
[20] Oh J, Lyon C, Washington S P, et al. Validation of the FHWA crash models for rural intersections: lessons learned[J]. Transportation Research Record,2003,1840:41-49.
[1] 代存杰,李引珍,马昌喜,柴获,牟海波. 不确定条件下危险品配送路线多准则优化[J]. 吉林大学学报(工学版), 2018, 48(6): 1694-1702.
[2] 王芳荣, 郭柏苍, 金立生, 高琳琳, 岳欣羽. 次任务驾驶安全评价指标筛选及其权值计算[J]. 吉林大学学报(工学版), 2017, 47(6): 1710-1715.
[3] 谭立东, 刘丹, 李文军. 基于蝇复眼的交通事故现场全景图像阵列仿生设计[J]. 吉林大学学报(工学版), 2017, 47(6): 1738-1744.
[4] 李显生, 孟祥雨, 郑雪莲, 程竹青, 任圆圆. 非满载罐体内液体冲击动力学特性[J]. 吉林大学学报(工学版), 2017, 47(3): 737-743.
[5] 王占中, 赵利英, 曹宁博. 基于多层编码遗传算法的危险品运输调度模型[J]. 吉林大学学报(工学版), 2017, 47(3): 751-755.
[6] 徐进, 陈薇, 周佳, 罗骁, 邵毅明. 汽车转向盘操作与驾驶负荷的相关性[J]. 吉林大学学报(工学版), 2017, 47(2): 438-445.
[7] 郭应时, 付锐, 赵凯, 马勇, 袁伟. 驾驶人换道意图实时识别模型评价及测试[J]. 吉林大学学报(工学版), 2016, 46(6): 1836-1844.
[8] 王喆, 杨柏婷, 刘昕, 刘群, 宋现敏. 基于模糊聚类的驾驶决策判别[J]. 吉林大学学报(工学版), 2015, 45(5): 1414-1419.
[9] 马勇, 石涌泉, 付锐, 郭应时. 驾驶人分心时长对车道偏离影响的实车试验[J]. 吉林大学学报(工学版), 2015, 45(4): 1095-1101.
[10] 徐建, 孙璐. 解决交通事故数据分析中零值问题的模型[J]. 吉林大学学报(工学版), 2015, 45(3): 769-775.
[11] 金立生, 王岩, 刘景华, 王亚丽, 郑义. 基于Adaboost算法的日间前方车辆检测[J]. 吉林大学学报(工学版), 2014, 44(6): 1604-1608.
[12] 金立生,牛清宁,刘景华,秦彦光,吕欢欢. 不同道路线形下驾驶人认知分散状态监测[J]. 吉林大学学报(工学版), 2014, 44(3): 642-647.
[13] 詹伟, 吕庆, 尚岳全. 高速公路隧道群交通事故灰色马尔可夫预测[J]. 吉林大学学报(工学版), 2014, 44(01): 62-67.
[14] 李志斌, 刘攀, 金茂菁, 徐铖铖. 高速公路常发拥堵路段追尾事故风险实时预测[J]. 吉林大学学报(工学版), 2013, 43(06): 1482-1487.
[15] 郑雪莲, 李显生, 任园园, 王宇宁, 杨猛. 非满载汽车罐车液体冲击等效机械模型[J]. 吉林大学学报(工学版), 2013, 43(06): 1488-1493.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 《吉林大学学报(工学版)》编辑部
地址：长春市人民大街5988号　电话：+86-431-85095297　传真：+86-431-85094074　E-mail: xbgxb@jlu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发