Pareto Efficiency Research Of Reliability-based User Equilibrium

The uncertainty of travel time in transportation networks affects travelers’ route choice,travelers may be faced with several criteria to select the routes.Empirical research shows that with the continuous improvement of people’s travel time values,travelers are not only concerned with travel time,but more attention to the actual arrival time,and not only to minimize travel time,but also pay more attention to travel time reliability and volatility.In order to analyze how the traveler consider the equivalence relationship between travel time and travel risk,and risk-taking behavior of travelers under the uncertain environment,the concept of Pareto efficiency is introduced from the point of view of resource allocation.It aims to analyze how the traveler makes a tradeoff between the travel time reliability and travel time risk to make optimal route choice.At the same time,considering the effect of travel costs on traveler’s route choice,and further analyzes the route selection behavior of travelers under the influence of travel time and travel cost.This thesis firstly introduces methods for modeling route choice among uncertain environment,which including stochastic dominance method and mean risk method.Then,four commonly used reliability-based user equilibrium models are introduced:the percentile travel time model,travel time budget model,average excess travel time model,combined mean travel time model and quadratic disutility function model,and the basic ideas and characteristics of each model are expounded.Afterwards,considering the different types of travelers,the definition of the Pareto efficiency route flow pattern under the equilibrium state is introduced in combination with the mean-variance non-dominated route.And whether these models possess,and in what cases possess the Pareto efficiency route flow pattern is analyzed.Furthermore,analysis of geometric characteristics of traveler risk-taking behavior is developed,and using the mean-standard deviation indifference curve to analyze the various types of risk-taking behavior of travelers.And the geometric characteristics and equilibrium states of the reliability-based user equilibrium model are verified by the case analysis,which suppose the route travel time follows a normal distribution.Based on the travel time variability,this thesis then considers both travel time,travel time reliability and travel cost in the route choice decision process,a bi-objective user equilibrium model is constructed,and the definition of Pareto-efficiency route flow pattern under double target is given.In the case of a dual-target user equilibrium state,no traveler can reduce his/her travel time or travel cost or both without worsening the other objective by unilaterally changing routes.And the Pareto efficiency route flow pattern in this case is an effective solution.In order to solve the non-uniqueness of the solution in proposed model,a multi-criteria multi-class Pareto user equilibrium model from the perspective of traveler classification is constructed.In this network state,for travellers who are more concered about travel time,no one can reduce his/her travel time by unilaterally changing routes.And for travellers who are more concered about travel cost,no one can reduce his/her travel cost by unilaterally changing routes.Then,a Quasi Method of Successive Average(QMSA)algorithm is designed to solve the model in this chapter.In order to find out the most probable solution of the bi-objective user equilibrium model,a maximum entropy multi-class multi-criteria Pareto user equilibrium model whose feasible set is the solution set of bi-objective user equilibrium model is proposed.Then,a route-based solution algorithm,the partial linearization algorithm is developed to solve the model,and the example analysis clearly illustrates the traffic flow transfer process during the route flow reaches equilibrium and proves the convergence and efficiency of the algorithm.The TTB-C indifference curve was constructed to analyze the behavior of traveler’s route choice under the dual-target.Finally,the main research results,innovations and shortcomings of this thesis are summarized.