Study on Bounded Rational Route Choice Behavior and Modelling when Real-time Traffic Information Provided
Tao LI,Hong-zhi GUAN
DOI: https://doi.org/10.3969/j.issn.1002-0268.2018.03.013
2018-01-01
Journal of Highway and Transportation Research and Development
Abstract:Providing real-time traffic information helps travelers to judge the utility of different routes. However,the decision behavior of traveler is still bounded rationality,sometimes called non optimal choice, in some cases due to the vagueness of information or the limited computing power. Therefore, it is assumed that the traveler cannot fully predict the consequences of all alternatives which means maximizing utility based on Simon's theory of bounded rationality, and put forward a bounded rationality model. An experiment to verify the bounded rational decision behavior in traffic route choice is also designed. In the experiment,real-time traffic information is released to the subjects before each trip,the data such as the decision results and decision time of the subjects are recorded after each route choice. The bounded rationality route choice behavior is analyzed, and a bounded rationality route choice model is established to describe the bounded rationality of travelers. It is found that (1) when the utility difference of the two paths is less than a certain threshold, travelers cannot realize the rational judgment and utility maximization, and only the random decision can be made;(2) when the utility difference is larger than the threshold,the rational judgment can be made and the utility is maximized. According to this phenomenon, the maximization hypothesis of traditional Logit model is relaxed to the limited utility maximization hypothesis, and the Logit model under bounded rationality(bounded rational binary Logit model,BRBL) is established. Through the analysis of the model,it is found that the sensitivity of the model in the vicinity of 0 point is less than that of traditional Logit model. After calibrating BRBL model and BL(binary Logit) model,it is found that the improved model can explain some bounded rational behavior in the decision making of route choice, and the calibration result shows that the accuracy of BRBL model is better than that of traditional BL model, the new model is more suitable for the route choice problem when the road information provided.