考虑出行者偏好和经验的路径选择行为研究
|
摘要
出行者的路径选择行为是包括自身特性在内多种因素共同作用的结果。为分析出行者偏好对路径选择行为的影响,首先假设出行者从路网中获取的信息为不完全历史信息,建立了理解行程时间及其更新模型,然后给出了基于经验-加权吸引力(EWA)学习模型和累计强化学习模型的两种偏好动态更新规则,最后通过Dogit模型将理解行程时间和路径偏好共同纳入出行者的路径选择决策中。在此基础上,对比分析了不考虑路径偏好、路径偏好为固定值、基于EWA学习模型更新路径偏好和基于累计强化学习模型更新路径偏好4种不同偏好情况下网络交通流的演化情况。算例结果表明:相比利用Logit模型不考虑路径偏好的流量分配结果,利用Dogit模型考虑路径偏好的流量分配结果更为均衡,且在考虑偏好时,路径偏好为固定值、基于EWA学习模型更新路径偏好和基于累计强化学习模型更新路径偏好3种情况下路径的均衡流量间差异较小;偏好动态更新时,基于EWA学习模型的路径偏好动态更新规则较累计强化学习模型能更好地捕捉出行个体的路径偏好,但由累计强化学习模型得到的路网流量分配结果更为均衡;偏好为固定值时,路径的均衡流量介于EWA学习模型和累计强化学习模型两种偏好动态更新规则下路径的均衡流量之间。
Traveler's route choice behavior is the result of multiple factors including their own characteristics. In order to analyze the influence of traveler's preference on route choice behavior, first, assuming that the information obtained from the road network is incomplete historical information, the understanding travel time model and its update model are established. Then, 2 dynamical updating rules based on the preference of experience-weighted attraction(EWA) learning model and the cumulative reinforcement learning model are given. Finally, through the Dogit model, understanding travel time and route preference are integrated into traveler's route choice decision. On this basis, the evolutions of network traffic flow under 4 preferences(without considering route preference, route preference as a fixed value, updating route preference based on EWA learning model, and updating route preference based on cumulative reinforcement learning model) are compared and analyzed. The result of the example shows that(1) compared with the traffic distribution using the Logit model without considering the path preference, the traffic distribution using the Dogit model considering path preference is more balanced, and there is little difference among the route equilibrium traffic volumes under the route preference as a fixed value, the updating route preference based on EWA learning model and cumulative reinforcement learning model;(2) when dynamical updating the preference, the route preference dynamical update rule based on the EWA learning model can capture traveler's individual route preference better than the cumulative reinforcement learning model, but the road network traffic distribution result obtained by the cumulative reinforcement learning model is more balanced;(3) when the preference is a fixed value, the route equilibrium volume is between those under the rules of the EWA learning model and the cumulative reinforcement learning model.
Traveler's route choice behavior is the result of multiple factors including their own characteristics. In order to analyze the influence of traveler's preference on route choice behavior, first, assuming that the information obtained from the road network is incomplete historical information, the understanding travel time model and its update model are established. Then, 2 dynamical updating rules based on the preference of experience-weighted attraction(EWA) learning model and the cumulative reinforcement learning model are given. Finally, through the Dogit model, understanding travel time and route preference are integrated into traveler's route choice decision. On this basis, the evolutions of network traffic flow under 4 preferences(without considering route preference, route preference as a fixed value, updating route preference based on EWA learning model, and updating route preference based on cumulative reinforcement learning model) are compared and analyzed. The result of the example shows that(1) compared with the traffic distribution using the Logit model without considering the path preference, the traffic distribution using the Dogit model considering path preference is more balanced, and there is little difference among the route equilibrium traffic volumes under the route preference as a fixed value, the updating route preference based on EWA learning model and cumulative reinforcement learning model;(2) when dynamical updating the preference, the route preference dynamical update rule based on the EWA learning model can capture traveler's individual route preference better than the cumulative reinforcement learning model, but the road network traffic distribution result obtained by the cumulative reinforcement learning model is more balanced;(3) when the preference is a fixed value, the route equilibrium volume is between those under the rules of the EWA learning model and the cumulative reinforcement learning model.
引文
[1] CHORUS C G,MOLIN E J E,ARENTZE T A,et al.Validation of a Multimodal Travel Simulator With Travel Information Provision[J].Transportation Research Part C:Emerging Technologies,2007,15(3):191-207.
[2] 李志纯,黄海军.先进的旅行者信息系统对出行者选择行为的影响研究[J].公路交通科技,2005,22(2):95-99.LI Zhi-chun,HUANG Hai-jun.Modeling the Impacts of Advanced Traveler Information Systems on Travelers’ Travel Choice Behaviors[J].Journal of Highway and Transportation Research and Development,2005,22(2):95-99.
[3] 肖浩汉,蒋慧园,刘义,等.城市道路拥挤收费下路径选择的演化博弈分析[J].武汉理工大学学报,2015,37(9):53-59.XIAO Hao-han,JIANG Hui-yuan,LIU Yi,et al.Evolutionary Game Analysis of Route Choices under Urban Road Congestion Charging[J].Journal of Wuhan University of Technology,2015,37(9):53-59.
[4] VERPLANKEN B,AARTS H.Habit,Attitude,and Planned Behavior:Is Habit an Empty Construct or an Interesting Case of Goal-directed Automaticity?[J].European Review of Social Psychology,1999,10(1):101-134.
[5] XIE C,LIU Z.On the Stochastic Network Equilibrium with Heterogeneous Choice Inertia[J].Transportation Research Part B:Methodological,2014,66:90-109.
[6] 刘诗序,关宏志.出行者有限理性下的逐日路径选择行为和网络交通流演化[J].土木工程学报,2013,46(12):136-144.LIU Shi-xu,GUAN Hong-zhi.Travelers’ Day-to-day Route-choice Behavior and Evolution of Network Traffic Flow Based on Bounded Rationality[J].China Civil Engineering Journal,2013,46(12):136-144.
[7] CAMERER C,HO T H.Experience-weighted Attraction Learning in Normal Form Games[J].Econometrica,1999,67(4):827-874.
[8] CHEN Y,KHOROSHILOV Y.Learning under Limited Information[J].Games and Economic Behavior,2003,44(1):1-25.
[9] 夏新海.面向城市自适应交通信号控制的强化学习方法研究[D].广州:华南理工大学,2013.XIA Xin-hai.Study of Reinforcement Learning Towards Urban Self-adaptive Traffic Signal Control Environment[D].Guangzhou:South China University of Technology,2013.
[10] 段勇,徐心和.基于多智能体强化学习的多机器人协作策略研究[J].系统工程理论与实践,2014,34(5):1305-1310.DUAN Yong,XU Xin-he.Research on Multi-robot Cooperation Strategy Based on Multi-agent Reinforcement Learning[J].Systems Engineering-Theory & Practice,2014,34(5):1305-1310.
[11] 徐超,周宗放.基于EWA学习机制的联保贷款组织还款策略选择行为[J].系统工程,2015,33(6):111-116.XU Chao,ZHOU Zong-fang.Repayment Strategy Behavior of Guaranteed Loan Organizations Based on EWA Learning Model[J].Systems Engineering,2015,33(6):111-116.
[12] 宋妍.基于EWA学习的共享资源捐赠习俗演化仿真研究[J].系统科学学报,2013,21(4):78-81.SONG Yan.Simulation Research for the Contribution Convention Evolutionary of Shared Resource Based on EWA Learning[J].Journal of Systems Science,2013,21(4):78-81.
[13] GAUNDRY M J I,DAGENAIS M G.The Dogit Model[J].Transportation Research Part B:Methodological,1979,13(2):105-111.
[14] BEN-AKIVA M.Choice Models with Simple Choice Set Generating Processes[R].Cambridge:MIT,1977.
[15] GAUDRY M J I,WILLS M J.Testing the Dogit Model with Aggregate Time-series and Cross-sectional Travel Data[J].Transportation Research Part B:Methodological,1979,13(2):155-166.
[16] LI G,ZHANG J,NUGROHO S B,et al.Analysis of Paratransit Drivers’ Stated Job Choice Behavior under Various Policy Interventions Incorporating the Influence of Captivity:A Case Study in Jabodetabek Metropolitan Area,Indonesia[J].Journal of the Eastern Asia Society for Transportation Studies,2011,9:1144-1159.
[17] BORDLEY R F.The Dogit Model Is Applicable Even without Perfectly Captive Buyers[J].Transportation Research Part B:Methodological,1990,24(4):315-323.
[18] 刘新民,鲁晓燕,孙秋霞.基于不同偏好的出行者路径选择行为研究[J].重庆交通大学学报:自然科学版,2017,36(10):102-106.LIU Xin-min,LU Xiao-yan,SUN Qiu-xia.Traveler's Behavior of Path Selection Based on Different Preferences[J].Journal of Chongqing Jiaotong University:Natural Science Edition,2017,36(10):102-106.
[19] 刘诗序,关宏志,严海.预测信息下的驾驶员逐日路径选择行为与系统演化[J].北京工业大学学报,2012,38(2):269-274.LIU Shi-xu,GUAN Hong-zhi,YAN Hai.Driver's Day-to-day Route Choice Behavior and System Evolution under Forecast Information[J].Journal of Beijing University of Technology,2012,38(2):269-274.
[20] 田丽君,江晓岚,刘天亮,等.基于Dogit模型考虑路径偏好的日常出行行为研究[J].交通运输系统工程与信息,2016,16(6):228-235.TIAN Li-jun,JIANG Xiao-lan,LIU Tian-liang,et al.Day-to-day Route Choice Behavior Considering Route Preference in a Dogit Model[J].Journal of Transportation System Engineering and Information Technology,2016,16(6):228-235.
[21] 曾鹦.考虑乘客选择行为的城市公交客流分配及系统演化[D].成都:西南交通大学,2014.ZENG Ying.Urban Public Transit Assignment and System Evolution Considering Passengers’ Choice Behavior[D].Chengdu:Southwest Jiaotong University,2014.
[2] 李志纯,黄海军.先进的旅行者信息系统对出行者选择行为的影响研究[J].公路交通科技,2005,22(2):95-99.LI Zhi-chun,HUANG Hai-jun.Modeling the Impacts of Advanced Traveler Information Systems on Travelers’ Travel Choice Behaviors[J].Journal of Highway and Transportation Research and Development,2005,22(2):95-99.
[3] 肖浩汉,蒋慧园,刘义,等.城市道路拥挤收费下路径选择的演化博弈分析[J].武汉理工大学学报,2015,37(9):53-59.XIAO Hao-han,JIANG Hui-yuan,LIU Yi,et al.Evolutionary Game Analysis of Route Choices under Urban Road Congestion Charging[J].Journal of Wuhan University of Technology,2015,37(9):53-59.
[4] VERPLANKEN B,AARTS H.Habit,Attitude,and Planned Behavior:Is Habit an Empty Construct or an Interesting Case of Goal-directed Automaticity?[J].European Review of Social Psychology,1999,10(1):101-134.
[5] XIE C,LIU Z.On the Stochastic Network Equilibrium with Heterogeneous Choice Inertia[J].Transportation Research Part B:Methodological,2014,66:90-109.
[6] 刘诗序,关宏志.出行者有限理性下的逐日路径选择行为和网络交通流演化[J].土木工程学报,2013,46(12):136-144.LIU Shi-xu,GUAN Hong-zhi.Travelers’ Day-to-day Route-choice Behavior and Evolution of Network Traffic Flow Based on Bounded Rationality[J].China Civil Engineering Journal,2013,46(12):136-144.
[7] CAMERER C,HO T H.Experience-weighted Attraction Learning in Normal Form Games[J].Econometrica,1999,67(4):827-874.
[8] CHEN Y,KHOROSHILOV Y.Learning under Limited Information[J].Games and Economic Behavior,2003,44(1):1-25.
[9] 夏新海.面向城市自适应交通信号控制的强化学习方法研究[D].广州:华南理工大学,2013.XIA Xin-hai.Study of Reinforcement Learning Towards Urban Self-adaptive Traffic Signal Control Environment[D].Guangzhou:South China University of Technology,2013.
[10] 段勇,徐心和.基于多智能体强化学习的多机器人协作策略研究[J].系统工程理论与实践,2014,34(5):1305-1310.DUAN Yong,XU Xin-he.Research on Multi-robot Cooperation Strategy Based on Multi-agent Reinforcement Learning[J].Systems Engineering-Theory & Practice,2014,34(5):1305-1310.
[11] 徐超,周宗放.基于EWA学习机制的联保贷款组织还款策略选择行为[J].系统工程,2015,33(6):111-116.XU Chao,ZHOU Zong-fang.Repayment Strategy Behavior of Guaranteed Loan Organizations Based on EWA Learning Model[J].Systems Engineering,2015,33(6):111-116.
[12] 宋妍.基于EWA学习的共享资源捐赠习俗演化仿真研究[J].系统科学学报,2013,21(4):78-81.SONG Yan.Simulation Research for the Contribution Convention Evolutionary of Shared Resource Based on EWA Learning[J].Journal of Systems Science,2013,21(4):78-81.
[13] GAUNDRY M J I,DAGENAIS M G.The Dogit Model[J].Transportation Research Part B:Methodological,1979,13(2):105-111.
[14] BEN-AKIVA M.Choice Models with Simple Choice Set Generating Processes[R].Cambridge:MIT,1977.
[15] GAUDRY M J I,WILLS M J.Testing the Dogit Model with Aggregate Time-series and Cross-sectional Travel Data[J].Transportation Research Part B:Methodological,1979,13(2):155-166.
[16] LI G,ZHANG J,NUGROHO S B,et al.Analysis of Paratransit Drivers’ Stated Job Choice Behavior under Various Policy Interventions Incorporating the Influence of Captivity:A Case Study in Jabodetabek Metropolitan Area,Indonesia[J].Journal of the Eastern Asia Society for Transportation Studies,2011,9:1144-1159.
[17] BORDLEY R F.The Dogit Model Is Applicable Even without Perfectly Captive Buyers[J].Transportation Research Part B:Methodological,1990,24(4):315-323.
[18] 刘新民,鲁晓燕,孙秋霞.基于不同偏好的出行者路径选择行为研究[J].重庆交通大学学报:自然科学版,2017,36(10):102-106.LIU Xin-min,LU Xiao-yan,SUN Qiu-xia.Traveler's Behavior of Path Selection Based on Different Preferences[J].Journal of Chongqing Jiaotong University:Natural Science Edition,2017,36(10):102-106.
[19] 刘诗序,关宏志,严海.预测信息下的驾驶员逐日路径选择行为与系统演化[J].北京工业大学学报,2012,38(2):269-274.LIU Shi-xu,GUAN Hong-zhi,YAN Hai.Driver's Day-to-day Route Choice Behavior and System Evolution under Forecast Information[J].Journal of Beijing University of Technology,2012,38(2):269-274.
[20] 田丽君,江晓岚,刘天亮,等.基于Dogit模型考虑路径偏好的日常出行行为研究[J].交通运输系统工程与信息,2016,16(6):228-235.TIAN Li-jun,JIANG Xiao-lan,LIU Tian-liang,et al.Day-to-day Route Choice Behavior Considering Route Preference in a Dogit Model[J].Journal of Transportation System Engineering and Information Technology,2016,16(6):228-235.
[21] 曾鹦.考虑乘客选择行为的城市公交客流分配及系统演化[D].成都:西南交通大学,2014.ZENG Ying.Urban Public Transit Assignment and System Evolution Considering Passengers’ Choice Behavior[D].Chengdu:Southwest Jiaotong University,2014.