基于混合Logit模型的高速公路交通事故严重程度分析
|
摘要
为从多方面掌握影响高速公路事故严重程度的因素,基于统计分析方法构建事故严重程度模型,分析其与道路、环境、驾驶员及车辆等因素间关系。鉴于多项Logit模型难以解析异质性及各因素对事故影响的交互作用,构建了混合Logit模型,并提出了刻画参数间相关性的方法。结果表明,考虑参数间相关性的混合Logit模型比多项Logit模型有更好的拟合优度,且能更合理地反映各因素对事故严重程度的作用效果;碰撞护栏或桥墩、女性驾驶员或驾驶员超过56岁时,更易产生受伤和死亡事故;能见度低于200m、驾驶员驾龄小于3年或超过10年、责任车辆为重型货车或车辆变更车道时,发生财产损失事故的概率增加,而发生死亡和受伤事故的概率有所降低;湿滑路面将导致受伤事故的概率增加3.7%,而混凝土护栏和夜间无照明时将使死亡事故的概率分别增加8.7%和28.8%。
In order to investigate factors affecting severity of traffic crashes,a model based on statistical analysis is developed,and relationships between severity of traffic crashes and influencing factors including road,environment,driver,and vehicle is analyzed.For a multinomial Logit model can barely capture unobserved heterogeneity and interactions among effects of factors on crashes,a mixed Logit model is developed and a method to analyze correlation between parameters is proposed.The results show that compared with the multinomial model,the mixed Logit model which considers correlation between parameters has a better goodness-of-fit and can more reasonably reflect effects of influencing factors on severity of traffic crashes.Colliding with barriers or piers,female drivers,or drivers over 56 are more likely to result in injuries and fatal crashes.Traffic crashes caused by visibility less than 200 m,driving experience of drivers is less than3 years or over 10 years,and wrongdoers are heavy vehicles or lane-changing vehicles,are all associated with increasing probability of property-damage-only crashes,while with decreasing probability of injuries and fatal crashes.Wet and slippery road surfaces increases probability of injury crashes by 3.7%,while a concrete barriers and unlighted conditions at night result in increasing fatal crashes by 8.7% and 28.8%,respectively.
In order to investigate factors affecting severity of traffic crashes,a model based on statistical analysis is developed,and relationships between severity of traffic crashes and influencing factors including road,environment,driver,and vehicle is analyzed.For a multinomial Logit model can barely capture unobserved heterogeneity and interactions among effects of factors on crashes,a mixed Logit model is developed and a method to analyze correlation between parameters is proposed.The results show that compared with the multinomial model,the mixed Logit model which considers correlation between parameters has a better goodness-of-fit and can more reasonably reflect effects of influencing factors on severity of traffic crashes.Colliding with barriers or piers,female drivers,or drivers over 56 are more likely to result in injuries and fatal crashes.Traffic crashes caused by visibility less than 200 m,driving experience of drivers is less than3 years or over 10 years,and wrongdoers are heavy vehicles or lane-changing vehicles,are all associated with increasing probability of property-damage-only crashes,while with decreasing probability of injuries and fatal crashes.Wet and slippery road surfaces increases probability of injury crashes by 3.7%,while a concrete barriers and unlighted conditions at night result in increasing fatal crashes by 8.7% and 28.8%,respectively.
引文
[1]World Health Organization.Global status report on road safety 2016:Supporting a decade of action[R].Switzerland:World Health Organization,2016.
[2]公安部交通管理局.中国道路交通事故统计年报[R].北京:公安部交通管理局,2017.Ministry of Public Security,Transportation Bureau.The road traffic accidents statistics report in China[R].Beijing:Ministry of Public Security,Transportation Bureau,2017.(in Chinese)
[3]ZHAO J,DENG W.Traffic accidents on expressways:New threat to China[J].Traffic Injury Prevention,2012,13(3):230-238.
[4]王长君,王励旸,李瑞敏.道路因素对典型较严重道路交通事故严重性的影响分析[J].交通信息与安全,2016,34(3):10-17.WANG Changjun,WANG Liyang,LI Ruimin.An analysis on the impact of highway factors on severity of typical traffic crashes[J].Journal of Transportation Information and Safety,2016,34(3):10-17.(in Chinese)
[5]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.MA Fuming,LIU Hao.The influence of snow weather affects fast road traffic characteristics and its countermeasures[J].Technology&Economy in Areas of Communications,2016,18(4):53-56.(in Chinese)
[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.ZHAO Yuefeng,ZHANG Shengrui,MA Zhuanglin.Analysis of traffic accident severity on highway tunnels using the partial proportion odds model[J].China Journal of Highway Transportation,2018,31(9):159-166.(in Chinese)
[7]宗芳,许洪国,张慧永.基于Ordered Probit模型的交通事故受伤人数预测[J].华南理工大学学报(自然科学版),2012,40(7):41-45.ZONG Fang,XU Hongguo,ZHANG Huiyong.Forecast injury number due to traffic accident based on ordered probit model[J].South China University of Technology:Natural Science Edition,2012,40(7):41-45.(in Chinese)
[8]KHATTAK A J,PAWLOVICH M D,SOULEY-RETTE R R,et al.Factors related to more severe older driver traffic crash injuries[J].Journal of Transportation Engineering,2002,128(3):243-249.
[9]HU W,DONNELL E T.Median barrier crash severity:some new insights[J].Accident Analysis&Prevention,2010,42(6):697-704.
[10]FILTNESS A J,ARMSTRONG K A,WATSONA,et al.Sleep-related vehicle crashes on low speed roads[J].Accident Analysis&Prevention,2017(99):79-286.
[11]ANARKOOLI A J,HOSSEINPOUR M,KAR-DAR A.Investigation of factors affecting the injury severity of single-vehicle rollover crashes:a random-effects generalized ordered probit model[J].Accident Analysis&Prevention,2017(106):399-410.
[12]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.WANG Lei,LYU Pu,LIN Yongjie.Traffic accidents on freeways:Influencing factors analysis and injury severity evaluation[J].China Safety Science Journal,2016,26(3):86-90.(in Chinese)
[13]SAVOLAINEN P T,MANNERING F L,LORDD,et al.The statistical analysis of highway crash-injury severities:A review and assessment of methodological alternatives[J].Accident Analysis and Prevention,2011,43(5):1666-1676.
[14]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.MA Zhuanglin,ZHANG Weiwei,YANG Yang,et al.Research on models for predicting severity of traffic accident in highway tunnel[J].China Safety Science Journal,2015,25(5):75-79.(in Chinese)
[15]刘海珠.道路交通事故严重程度影响因素分析及预测模型建立[D].长春:吉林大学,2014.LIU Haizhu.The analysis of influencing factors for crash severity and the establishment of predication model[D].Changchun:Jilin University,2014.(in Chinese)
[16]STIPANCIC J,ZANGENEHPOUR S,MIRAN-DA-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.
[17]陈艳艳,李向楠,孙智源,等.基于Logistics回归的道路交通事故影响因素辨识方法[J].交通科技与经济,2018,20(5):1-5.CHEN Yanyan,LI Xiangnan,SUN Zhiyuan,et al.Identification method of factors affecting traffic accident based on logistics[J].Technology&Economy in Areas of Communications,2018,20(5):1-5.(in Chinese)
[18]杨晔,王立晓,左志.考虑乘客的交通事故受伤严重程度模型[J].交通科技与经济,2018,20(4):23-26.(in Chinese)YANG Ye,WANG Lixiao,ZUO Zhi.The traffic accident injury severity model considering passenger[J].Technology&Economy in Areas of Communications,2018,20(4):23-26.(in Chinese)
[19]ULFARSSON G F,MANNERING F L.Differences in male and female injury severities in sport-utility vehicle,minivan,pickup and passenger car accidents[J].Accident Analysis&Prevention,2004,36(2):135-147.
[20]CELIK,A K,OKTAY E.A multinomial logit analysis of risk factors influencing road traffic injury severities in the Erzurum and Kars provinces of Turkey[J].Accident Analysis&Prevention,2014(72):66-77.
[21]MALYSHKINA N V,MANNERING F L.Markov switching multinomial logit model:An application to accident-injury severities[J].Accident A-nalysis&Prevention,2009,41(4):829-838.
[22]MALYSHKINA N V,MANNERING F L.Empirical assessment of the impact of highway design exceptions on the frequency and severity of vehicle accidents[J].Accident Analysis&Prevention,2010,42(1):131-139.
[23]HOU Q,MENG X,LENG J,et al.Application of a random effects negative binomial model to examine crash frequency for freeways in China[J].Physica A:Statistical Mechanics&Its Applications,2018(509):937-944.
[24]SHAHEED M S,GKRITZA K,ZHANG W,et al.A mixed logit analysis of two-vehicle crash severities involving a motorcycle[J].Accident A-nalysis&Prevention,2013(61):119-128.
[25]KIM J K,ULFARSSON G F,SHANKAR V N,et al.A note on modeling pedestrian-injury severity in motor-vehicle crashes with the mixed logit model[J].Accident Analysis&Prevention,2010,42(6):1751-1758.
[26]HOU Q,TARKO A P,MENG X.Investigating factors of crash frequency with random effects and random parameters models:New insights from Chinese freeway study[J].Accident Analysis&Prevention,2018(120):1-12.
[2]公安部交通管理局.中国道路交通事故统计年报[R].北京:公安部交通管理局,2017.Ministry of Public Security,Transportation Bureau.The road traffic accidents statistics report in China[R].Beijing:Ministry of Public Security,Transportation Bureau,2017.(in Chinese)
[3]ZHAO J,DENG W.Traffic accidents on expressways:New threat to China[J].Traffic Injury Prevention,2012,13(3):230-238.
[4]王长君,王励旸,李瑞敏.道路因素对典型较严重道路交通事故严重性的影响分析[J].交通信息与安全,2016,34(3):10-17.WANG Changjun,WANG Liyang,LI Ruimin.An analysis on the impact of highway factors on severity of typical traffic crashes[J].Journal of Transportation Information and Safety,2016,34(3):10-17.(in Chinese)
[5]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.MA Fuming,LIU Hao.The influence of snow weather affects fast road traffic characteristics and its countermeasures[J].Technology&Economy in Areas of Communications,2016,18(4):53-56.(in Chinese)
[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.ZHAO Yuefeng,ZHANG Shengrui,MA Zhuanglin.Analysis of traffic accident severity on highway tunnels using the partial proportion odds model[J].China Journal of Highway Transportation,2018,31(9):159-166.(in Chinese)
[7]宗芳,许洪国,张慧永.基于Ordered Probit模型的交通事故受伤人数预测[J].华南理工大学学报(自然科学版),2012,40(7):41-45.ZONG Fang,XU Hongguo,ZHANG Huiyong.Forecast injury number due to traffic accident based on ordered probit model[J].South China University of Technology:Natural Science Edition,2012,40(7):41-45.(in Chinese)
[8]KHATTAK A J,PAWLOVICH M D,SOULEY-RETTE R R,et al.Factors related to more severe older driver traffic crash injuries[J].Journal of Transportation Engineering,2002,128(3):243-249.
[9]HU W,DONNELL E T.Median barrier crash severity:some new insights[J].Accident Analysis&Prevention,2010,42(6):697-704.
[10]FILTNESS A J,ARMSTRONG K A,WATSONA,et al.Sleep-related vehicle crashes on low speed roads[J].Accident Analysis&Prevention,2017(99):79-286.
[11]ANARKOOLI A J,HOSSEINPOUR M,KAR-DAR A.Investigation of factors affecting the injury severity of single-vehicle rollover crashes:a random-effects generalized ordered probit model[J].Accident Analysis&Prevention,2017(106):399-410.
[12]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.WANG Lei,LYU Pu,LIN Yongjie.Traffic accidents on freeways:Influencing factors analysis and injury severity evaluation[J].China Safety Science Journal,2016,26(3):86-90.(in Chinese)
[13]SAVOLAINEN P T,MANNERING F L,LORDD,et al.The statistical analysis of highway crash-injury severities:A review and assessment of methodological alternatives[J].Accident Analysis and Prevention,2011,43(5):1666-1676.
[14]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.MA Zhuanglin,ZHANG Weiwei,YANG Yang,et al.Research on models for predicting severity of traffic accident in highway tunnel[J].China Safety Science Journal,2015,25(5):75-79.(in Chinese)
[15]刘海珠.道路交通事故严重程度影响因素分析及预测模型建立[D].长春:吉林大学,2014.LIU Haizhu.The analysis of influencing factors for crash severity and the establishment of predication model[D].Changchun:Jilin University,2014.(in Chinese)
[16]STIPANCIC J,ZANGENEHPOUR S,MIRAN-DA-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.
[17]陈艳艳,李向楠,孙智源,等.基于Logistics回归的道路交通事故影响因素辨识方法[J].交通科技与经济,2018,20(5):1-5.CHEN Yanyan,LI Xiangnan,SUN Zhiyuan,et al.Identification method of factors affecting traffic accident based on logistics[J].Technology&Economy in Areas of Communications,2018,20(5):1-5.(in Chinese)
[18]杨晔,王立晓,左志.考虑乘客的交通事故受伤严重程度模型[J].交通科技与经济,2018,20(4):23-26.(in Chinese)YANG Ye,WANG Lixiao,ZUO Zhi.The traffic accident injury severity model considering passenger[J].Technology&Economy in Areas of Communications,2018,20(4):23-26.(in Chinese)
[19]ULFARSSON G F,MANNERING F L.Differences in male and female injury severities in sport-utility vehicle,minivan,pickup and passenger car accidents[J].Accident Analysis&Prevention,2004,36(2):135-147.
[20]CELIK,A K,OKTAY E.A multinomial logit analysis of risk factors influencing road traffic injury severities in the Erzurum and Kars provinces of Turkey[J].Accident Analysis&Prevention,2014(72):66-77.
[21]MALYSHKINA N V,MANNERING F L.Markov switching multinomial logit model:An application to accident-injury severities[J].Accident A-nalysis&Prevention,2009,41(4):829-838.
[22]MALYSHKINA N V,MANNERING F L.Empirical assessment of the impact of highway design exceptions on the frequency and severity of vehicle accidents[J].Accident Analysis&Prevention,2010,42(1):131-139.
[23]HOU Q,MENG X,LENG J,et al.Application of a random effects negative binomial model to examine crash frequency for freeways in China[J].Physica A:Statistical Mechanics&Its Applications,2018(509):937-944.
[24]SHAHEED M S,GKRITZA K,ZHANG W,et al.A mixed logit analysis of two-vehicle crash severities involving a motorcycle[J].Accident A-nalysis&Prevention,2013(61):119-128.
[25]KIM J K,ULFARSSON G F,SHANKAR V N,et al.A note on modeling pedestrian-injury severity in motor-vehicle crashes with the mixed logit model[J].Accident Analysis&Prevention,2010,42(6):1751-1758.
[26]HOU Q,TARKO A P,MENG X.Investigating factors of crash frequency with random effects and random parameters models:New insights from Chinese freeway study[J].Accident Analysis&Prevention,2018(120):1-12.