基于前景理论的出行路径选择模型研究

发布时间：2018-12-10 13:58

【摘要】：随着人们生活水平的提高,人均汽车拥有量逐年上升,也引起了很多交通问题,其中最严重的就是交通拥堵问题,然而传统的交通系统管理和交通需求管理已经无力缓解,国内大多数学者一致认为智能交通系统(ITS)和先进的出行者信息系统(ATIS)是解决交通拥堵问题的有效途径之一,而ATIS的核心就是路径选择问题,因此,研究路径选择问题对于完善ATIS有极大的作用和意义。国内外学者对于确定环境下的路径选择问题已经有系统和深入的研究,并拥有了一系列的研究成果,而不确定环境下的路径选择研究还是热点研究问题。期望效用理论是解决不确定环境下路径选择的有效方法,但是效用理论认为人们是完全理性的假设与现实择路行为有较大的差距。而前景理论假设人们是有限理性的,并通过国内外学者大量的实证研究证实前景理论可以用来解决不确定环境下的路径选择问题。于是本文通过前景理论来研究不确定环境下的出行者路径选择问题。首先解析了不确定环境下的路径选择问题的分类,并详细介绍了三种路径选择模型的理论方法,之后分析了不确定环境下路径选择问题的研究现状。然后阐述了前景理论的提出背景、思想框架以及发展应用,最后分析了前景理论对路径选择的适用性。其次考虑出行者对准时到达的可靠性要求并结合行程时间预算的相关定义求出了行程时间预算区间,在此基础上,将基于行程时间预算的到达时间作为本章模型的两个参考点,建立了基于出发时刻的价值函数模型,然后通过出行者的个人主观概率确定了决策权重函数,进而搭建了基于前景理论的出发时刻和路径选择模型。通过对通勤者的SP调查统计得到了不同类型的通勤者的到达时间分布,并结合一个算例进行了验证。最后通过出行者基于前景理论的决策过程建立了基于前景理论的动态路径选择模型,进而提出了该动态模型的算法,并通过一个算例进行验证,然后再通过一个算例对基于效用理论和前景理论的路径选择模型进行比较得出前景理论比效用理论能够更加准确地描述出行者在不确定性条件下的出行决策行为。
[Abstract]:With the improvement of people's living standards, the per capita car ownership has increased year by year, which has also caused many traffic problems, the most serious of which is the traffic congestion problem. However, the traditional traffic system management and traffic demand management have been unable to alleviate. Most domestic scholars agree that the intelligent transportation system (ITS) and the advanced traveler information system (ATIS) are one of the effective ways to solve the traffic congestion problem, and the core of ATIS is the path selection problem. The study of path selection is of great importance to the improvement of ATIS. Scholars at home and abroad have made systematic and in-depth research on the path selection problem in the deterministic environment, and have a series of research results. However, the research on path selection in uncertain environment is still a hot research issue. The theory of expected utility is an effective method to solve the problem of path choice in uncertain environments. However, the theory of utility holds that there is a big gap between the assumption that people are completely rational and the behavior of road choice in reality. The prospect theory assumes that people are bounded rational, and through a large number of empirical studies at home and abroad, the prospect theory can be used to solve the path selection problem in uncertain environment. So this paper studies the path selection problem in uncertain environment by foreground theory. Firstly, the classification of path selection problems in uncertain environment is analyzed, and the theoretical methods of three kinds of path selection models are introduced in detail, and then the current research situation of path selection problems in uncertain environments is analyzed. Then, the background, the frame of thought and the development and application of foreground theory are introduced. Finally, the applicability of foreground theory to path selection is analyzed. Secondly, considering the reliability requirements of travelers to arrive on time and combining with the definition of travel time budget, the travel time budget interval is obtained. On this basis, the arrival time based on travel time budget is regarded as the two reference points of this chapter model. The value function model based on the departure time is established, and then the decision weight function is determined by the individual subjective probability of the traveler, and the starting time and path selection model based on the foreground theory is built. Through the SP survey of commuters, the arrival time distribution of different types of commuters is obtained and verified by an example. Finally, the dynamic path selection model based on foreground theory is established through the decision process of traveler based on foreground theory, and the algorithm of the dynamic model is put forward, which is verified by an example. Then a numerical example is given to compare the path selection model based on utility theory and prospect theory. It is concluded that foreground theory is more accurate than utility theory in describing travelers' travel decision behavior under uncertain conditions.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491

【参考文献】