基于交通行为分析的交通组织优化理论与方法研究
|
摘要
近年来,机动车保有量大幅度增加,其增长速度远远大于道路改扩建的速度,经济快速发展激励生成的交通需求远远大于已有道路资源的供给能力。然而,道路资源是有限的,而且在近期内也不可能无限度地增扩,如何在现有的道路资源基础上,通过寻求交通组织优化理论与方法,最大限度提升现有道路资源供给能力,同时引导或者改变交通需求,从而使得交通供需趋于平衡,是近年来交通领域研究热点之一。
本论文从消费者行为学理论中的消费购买决策过程角度,分析了出行者的交通行为过程,研究了目前在理论和应用层面比较成熟完善的Logit模型,分别推导选择概率对于选择方案属性和决策者特性两种变量的导数和弹性计算公式,从而得出了交通组织优化的理论依据。在模型的参数标定过程中,分别对两种不同变量进行了参数估计,完善了已有的参数估计方法。
最后进行案例分析,采用Logit模型对交通信息对驾驶人的路径选择影响调查数据建模并对参数进行标定,根据模型标定结果,预测了北京市道路网络中增设VMS后道路交通组织运行效率优化情况并进行了实证研究,结果表明路网负荷度变化比例与预测值相差在5个百分点以内。对比Logit模型与其他模型的预测结果则是下一步需要继续深入研究的内容。
The number of vehicles is increasing sharply in recent years, while the road network's improvement is far behind the rate of vehicle increase. Travel demand stimulated by the economic development has now exceeded the traffic supply capacity. It has been known that traffic capacity is limited and it could not be improved without any restriction in a short time, a common solution approach to improve the traffic supply capacity is optimization of traffic operation and management in the existing road network; simultaneously, inducing or changing travelers' traffic demand, therefore the balance between the supply and demand can be achieved.
In this paper, travelers' behavior is analyzed using the theory of consumer purchase-making decision process, and then the disaggregate Logit model is studied which is both developed in theory and application. Finally the theory foundation is achieved by derivatives and elasticities of choice probabilities with respect to attributes of alternatives and characteristics of decision maker. Parameters of these two variables are calibrated respectively, which improves the previous research in terms of parameter calibration method.
In the case study, the Logit model is adopted to calibrate traffic information effect on travelers' route choice behavior, and then the traffic operation optimization is predicted and demonstrated after setting up more Variable Message Signs (VMS) in Beijing road network according to the Logit model. The results show that there is a 5 Percentage Point Change between prediction and demonstration in the road network's Percentage Change for rate of volume to capacity. Compare Logit model with other models in prediction results aspects should be explored in future research.
本论文从消费者行为学理论中的消费购买决策过程角度,分析了出行者的交通行为过程,研究了目前在理论和应用层面比较成熟完善的Logit模型,分别推导选择概率对于选择方案属性和决策者特性两种变量的导数和弹性计算公式,从而得出了交通组织优化的理论依据。在模型的参数标定过程中,分别对两种不同变量进行了参数估计,完善了已有的参数估计方法。
最后进行案例分析,采用Logit模型对交通信息对驾驶人的路径选择影响调查数据建模并对参数进行标定,根据模型标定结果,预测了北京市道路网络中增设VMS后道路交通组织运行效率优化情况并进行了实证研究,结果表明路网负荷度变化比例与预测值相差在5个百分点以内。对比Logit模型与其他模型的预测结果则是下一步需要继续深入研究的内容。
The number of vehicles is increasing sharply in recent years, while the road network's improvement is far behind the rate of vehicle increase. Travel demand stimulated by the economic development has now exceeded the traffic supply capacity. It has been known that traffic capacity is limited and it could not be improved without any restriction in a short time, a common solution approach to improve the traffic supply capacity is optimization of traffic operation and management in the existing road network; simultaneously, inducing or changing travelers' traffic demand, therefore the balance between the supply and demand can be achieved.
In this paper, travelers' behavior is analyzed using the theory of consumer purchase-making decision process, and then the disaggregate Logit model is studied which is both developed in theory and application. Finally the theory foundation is achieved by derivatives and elasticities of choice probabilities with respect to attributes of alternatives and characteristics of decision maker. Parameters of these two variables are calibrated respectively, which improves the previous research in terms of parameter calibration method.
In the case study, the Logit model is adopted to calibrate traffic information effect on travelers' route choice behavior, and then the traffic operation optimization is predicted and demonstrated after setting up more Variable Message Signs (VMS) in Beijing road network according to the Logit model. The results show that there is a 5 Percentage Point Change between prediction and demonstration in the road network's Percentage Change for rate of volume to capacity. Compare Logit model with other models in prediction results aspects should be explored in future research.
引文
铩颷1]黄海军.交通行为建模-问题与机会.交通运输系统与工程信息,2002,2(1):24-28.
[2]章辉美,谷章原.交通社会学-对-门新生应用社会学的构想.湖南师范大学社会科学学报,2003,32(6):48-52.
[3]Adolf D.MAY.Model for Free way Corridor Analysis,TRB Special Report 194,Transportation,2000,27(1):1-4.
[4]Ben-Akiva,M.,Palma,A.D.,Kaysi,I.Dynamic network models and driver information systems.Transportation Research,1991,25A:251-266.
[5]Hussein Dia.An gent-based approach to modeling driver route choice behavior under theinfluence of real-time information.Transportation Research,2002,part C:331-349.
[6]Tsippy Lotan.Effects of Familiarity on Route Choice Behavior in the Presence of Information.Transportation Research Part C,1997,5(3):255-243.
[7]Wafaa Saleh,Seona Farrell.Investigation and analysis of evidence of asymmetric chum in traveldemand models.Transportation Research Part A,In Press,Corrected Proof,Available online 42006,12:671-702.
[8]Yaron Hollander.Direct versus indirect models for the effects of unreliability.TransportationResearch Part A,2006,40(9):699-711.
[9]Haitham m.al-deek,Asad j.khattak,Paramsothy thananjeyan.A combined traveler behavior andsystem performance model with advanced traveler information systems.TransportationResearch part A,1998,32(7):479-493.
[10]Kara maria Kockelman.A model for time- and budget-constrained activity demand analysis.Transportation Research part B,2001,35:255-269.
[11]李志瑶,隽志才,宗芳.居民出行时间选择及拥挤收费政策.交通运输工程学报,2005,5(3):105-110.
[12]林震,杨浩.交通信息诱导系统对出行选择的影响分析.交通运输系统与工程,2002,4:21-24.
[13]石小法,王炜,卢林,葛苏闽.交通信息影响下的动态路径选择模型研究.公路交通科技,2000,17(4):35-37.
[14]李佳,张成,范文博.私家车交通行为分析.交通与安全,2000,17(149):97-100.
[15]关宏志,绍洁,李亚茹,白洪岭.自驾车旅游交通的基础研究.北京工业大学学报,2005,31(2):151-154.
[16]王官城,消费心理学.北京:电子工业出版社.2004:17-21
[17]李东进,消费者行为学.北京:经济科学出版,2001:32-36.
[18]罗子明,消费心理学.北京:清华大学出版社.2002:14-17
[19]邵春福.交通规划原理.北京:中国铁道出版社,2004:99-120.
[20]Dijker T,etal.Car-following under congested condition[J].TRB,1998,1644:20-28.
[21]Frank S.Koppelman,Chandra Bhat.Yang D.A Self Instructing Course in Mode ChoiceModeling:Multinomail and Nested Logit Models.TRR,2006:57-94.
[22]徐丽丽,邵春福.交通信息对出行者路径选择行为影响调查结果分析与研究.第九届多国 城市交通学术会议论文集,2005:306-313.
铩颷23]徐丽丽,邵春福.基于双层规划模型的路径信息诱导研究.交通运输工程学报,2007,7(5):93-96.
[24]任福田,刘小明,荣建等.交通工程学.北京:人民交通出版社.2003:57-71.
[25]车广侠,石建军,隋莉颖.VMS交通诱导系统现状分析.道路交通与安全,2006,6(8):19-21.
[26]包煊,朱雪良,王义生,关积珍.VMS发布动态交通诱导信息的探讨.现代显示,2006,6:63-65.
[27]孙宝凤,隽志才,吴文静.高速公路VMS项目社会经济影响评价研究综述.交通运输系统工程与信息,2005,5(5):18-22.
[28]卢守峰,杨兆升,刘喜敏.基于元胞自动机的VMS诱导信息系统研究.交通与计算机,2006,24(2):9-12.
[29]史志法.非集计模型下公交规划的调查工作研究.交通与运输 2006,7:6-9.
[30]焦朋朋,陆化普.基于意向调查数据的非集计模型研究.公路交通科技,2005,22(6):114-116.
[31]隽志才,李志瑶,宗芳.基于活动链的出行需求预测方法综述.公路交通科技,2005,22(6):108-113.
[2]章辉美,谷章原.交通社会学-对-门新生应用社会学的构想.湖南师范大学社会科学学报,2003,32(6):48-52.
[3]Adolf D.MAY.Model for Free way Corridor Analysis,TRB Special Report 194,Transportation,2000,27(1):1-4.
[4]Ben-Akiva,M.,Palma,A.D.,Kaysi,I.Dynamic network models and driver information systems.Transportation Research,1991,25A:251-266.
[5]Hussein Dia.An gent-based approach to modeling driver route choice behavior under theinfluence of real-time information.Transportation Research,2002,part C:331-349.
[6]Tsippy Lotan.Effects of Familiarity on Route Choice Behavior in the Presence of Information.Transportation Research Part C,1997,5(3):255-243.
[7]Wafaa Saleh,Seona Farrell.Investigation and analysis of evidence of asymmetric chum in traveldemand models.Transportation Research Part A,In Press,Corrected Proof,Available online 42006,12:671-702.
[8]Yaron Hollander.Direct versus indirect models for the effects of unreliability.TransportationResearch Part A,2006,40(9):699-711.
[9]Haitham m.al-deek,Asad j.khattak,Paramsothy thananjeyan.A combined traveler behavior andsystem performance model with advanced traveler information systems.TransportationResearch part A,1998,32(7):479-493.
[10]Kara maria Kockelman.A model for time- and budget-constrained activity demand analysis.Transportation Research part B,2001,35:255-269.
[11]李志瑶,隽志才,宗芳.居民出行时间选择及拥挤收费政策.交通运输工程学报,2005,5(3):105-110.
[12]林震,杨浩.交通信息诱导系统对出行选择的影响分析.交通运输系统与工程,2002,4:21-24.
[13]石小法,王炜,卢林,葛苏闽.交通信息影响下的动态路径选择模型研究.公路交通科技,2000,17(4):35-37.
[14]李佳,张成,范文博.私家车交通行为分析.交通与安全,2000,17(149):97-100.
[15]关宏志,绍洁,李亚茹,白洪岭.自驾车旅游交通的基础研究.北京工业大学学报,2005,31(2):151-154.
[16]王官城,消费心理学.北京:电子工业出版社.2004:17-21
[17]李东进,消费者行为学.北京:经济科学出版,2001:32-36.
[18]罗子明,消费心理学.北京:清华大学出版社.2002:14-17
[19]邵春福.交通规划原理.北京:中国铁道出版社,2004:99-120.
[20]Dijker T,etal.Car-following under congested condition[J].TRB,1998,1644:20-28.
[21]Frank S.Koppelman,Chandra Bhat.Yang D.A Self Instructing Course in Mode ChoiceModeling:Multinomail and Nested Logit Models.TRR,2006:57-94.
[22]徐丽丽,邵春福.交通信息对出行者路径选择行为影响调查结果分析与研究.第九届多国 城市交通学术会议论文集,2005:306-313.
铩颷23]徐丽丽,邵春福.基于双层规划模型的路径信息诱导研究.交通运输工程学报,2007,7(5):93-96.
[24]任福田,刘小明,荣建等.交通工程学.北京:人民交通出版社.2003:57-71.
[25]车广侠,石建军,隋莉颖.VMS交通诱导系统现状分析.道路交通与安全,2006,6(8):19-21.
[26]包煊,朱雪良,王义生,关积珍.VMS发布动态交通诱导信息的探讨.现代显示,2006,6:63-65.
[27]孙宝凤,隽志才,吴文静.高速公路VMS项目社会经济影响评价研究综述.交通运输系统工程与信息,2005,5(5):18-22.
[28]卢守峰,杨兆升,刘喜敏.基于元胞自动机的VMS诱导信息系统研究.交通与计算机,2006,24(2):9-12.
[29]史志法.非集计模型下公交规划的调查工作研究.交通与运输 2006,7:6-9.
[30]焦朋朋,陆化普.基于意向调查数据的非集计模型研究.公路交通科技,2005,22(6):114-116.
[31]隽志才,李志瑶,宗芳.基于活动链的出行需求预测方法综述.公路交通科技,2005,22(6):108-113.