Empirical Analysis On Traveler's Route Choice Behavior And Models Considering Bounded Rationality

The study of traveler's route choice behavior directly determines the result of traffic flow assignment,which is an important content of urban traffic system analysis.Under the development of behavior decision-making,it is necessary to systematically sort out and verify the different bounded rational behavior characteristics and influencing factors.In order to highlight the practical significance of the theoretical models,it also needs the support of corresponding empirical analysisTherefore,this article conducts an in-depth and systematic analysis of the traveler's route choice behavior,and the decision rules of bounded rational route choice from both theoretical and empirical perspectives focusing on the large-scale data sets.And then,carry on model fitting and prediction analysis to the mainstream bounded rational route choice models.On this basis,the route choice model considering adaptive value of time is further adapted to improve the model's ability to explain actual problems.The specific research contents and conclusions are as followsFirst of all,from the perspective of psychology and decision science,systematically sorts out and analyzed in detail the bounded rational factors;Compares the existing route choice models considering bounded rationality,including models considering the single bounded rational factors,such as the conventional Logit route choice model,the route choice model considering asymmetric preferences and the route choice model considering the inertia threshold;and the route choice model with adaptive value of time that comprehensively considered a variety of bounded rational factors.And then compared the differences and connections between them from multiple anglesThen,conducts an empirical verification of multiple bounded rational factors through Wilcoxon Rank Sum Test,Kruskal-Wallis Rank Sum Test and other non-parametric hypothesis test methods as well as Logistic Classification method,?-inference Heuristic method.The test results indicate that,under a certain level of significance,decision makers have asymmetric preference behavior characteristics and satisfactory behavior characteristicsFinally,a systematic comparative analysis and parameter estimation are carried out on the four existing route choice models,and an improved route choice model with adaptive value of time that can more accurately predict the actual route choice behavior is further proposed.The maximum likelihood estimation method,simulated maximum likelihood estimation method,global search and other methods are used to fit and predict the model.In addition,explores the ability of different models to explain the actual travel behavior of travelersIn summary,it systematically sorts out and analyze the characteristics of bounded rational behavior and influencing factors from a theoretical and empirical perspective and uses different parameter estimation methods for different route choice models.On this basis,an improved route choice model considering adaptive value of time with better fitting results is proposed,which improves the prediction accuracy and provides a basis for the selection of route choice model in the study of actual traffic problems.