用户名： 密码： 验证码：
Experimental study on departure time choice behavior in morning commute problem and its modeling
详细信息    查看官网全文
摘要
A fundamental problem in the study of traffic systems is to understand various user choice behaviors and their underlying mechanisms that account for the emergence of complex traffic phenomena.Despite much effort devoted to theoretically exploring departure time choice behavior,relatively large scale and systematic experimental tests of theoretical predictions are still lacking.In this paper,we aim to offer a more comprehensive understanding of departure time choice behavior in terms of a series of laboratory experiments under different traffic conditions and feedback information to commuters.In the experiment,users make choices from a few of discrete departure times and the number of recruited players are much larger than that of choices to better mimic the real scenario that a large number of commuters will depart simultaneously in a relative small time window.Sufficient numbers of rounds are conducted to ensure the convergence of the collective behavior.Experimental results demonstrate that mean payoffs and mean number of subjects at each departure time are close to that in user equilibrium state,regardless of different scale and traffic conditions.Moreover,the amount of feedback information has negligible influence on the collective behavior but has relatively stronger effect of individual choice behavior,as confirmed by correlation analysis and Yule coefficient analysis.Significant fluctuation in individual choice persists until the end in spite of the achievement of aggregated user equilibrium.Two models in virtue of reinforcement learning and Fermi learning are built to reproduce experimental results and uncover the underlying mechanism.The aggregate behavior of players in simulation results of using both models is in good agreement with experimental results.
A fundamental problem in the study of traffic systems is to understand various user choice behaviors and their underlying mechanisms that account for the emergence of complex traffic phenomena.Despite much effort devoted to theoretically exploring departure time choice behavior,relatively large scale and systematic experimental tests of theoretical predictions are still lacking.In this paper,we aim to offer a more comprehensive understanding of departure time choice behavior in terms of a series of laboratory experiments under different traffic conditions and feedback information to commuters.In the experiment,users make choices from a few of discrete departure times and the number of recruited players are much larger than that of choices to better mimic the real scenario that a large number of commuters will depart simultaneously in a relative small time window.Sufficient numbers of rounds are conducted to ensure the convergence of the collective behavior.Experimental results demonstrate that mean payoffs and mean number of subjects at each departure time are close to that in user equilibrium state,regardless of different scale and traffic conditions.Moreover,the amount of feedback information has negligible influence on the collective behavior but has relatively stronger effect of individual choice behavior,as confirmed by correlation analysis and Yule coefficient analysis.Significant fluctuation in individual choice persists until the end in spite of the achievement of aggregated user equilibrium.Two models in virtue of reinforcement learning and Fermi learning are built to reproduce experimental results and uncover the underlying mechanism.The aggregate behavior of players in simulation results of using both models is in good agreement with experimental results.
引文

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址：北京市海淀区学院路29号 邮编：100083

电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700