What problem does this paper attempt to address?

The main problem this paper attempts to address is: In multi-stage traffic flow assignment models, how to find sufficient conditions to transform the search for equilibrium into a convex optimization problem. Specifically, the paper explores the possibility of converting the equilibrium search problem in multi-stage models into a unified convex optimization problem and proposes sufficient conditions to achieve this goal. ### Background and Motivation 1. **Multi-stage Traffic Flow Models**: - Since the 1970s, multi-stage traffic flow models have rapidly developed. These models typically involve iterative solutions of a series of convex optimization problems, gradually approaching the desired equilibrium state. - However, this approach may require careful selection of initial points to converge in practical applications. 2. **Unified Convex Optimization Problem**: - As an alternative method, researchers have attempted to find a unified convex optimization problem whose solution can directly provide the desired equilibrium state. - The advantage of this method is that it can avoid convergence issues during the iterative process, but it may not be successful in all cases. ### Main Contributions of the Paper 1. **Proposing Sufficient Conditions**: - The paper proposes sufficient conditions to ensure that the equilibrium search in multi-stage models can be transformed into a unified convex optimization problem. - These conditions have not appeared in previous literature and are somewhat innovative. 2. **Generality of the Model**: - The paper allows for the selection of a stable dynamic model (instead of the commonly used Bekman model) in one stage of the multi-stage model and considers the possibility of different types of users and vehicles. - This generality makes the model more flexible and better suited to the complexities of the real world. 3. **Numerical Experiments**: - The paper also conducts numerical experiments to verify the effectiveness of the proposed model and method in actual traffic networks. - The experimental results show that using Nesterov's accelerated method combined with the Sinkhorn-Bregman-Shelikhovsky method can efficiently solve the bi-dual problem on modern multi-core laptops, significantly reducing the dual gap. ### Conclusion By proposing sufficient conditions, the paper successfully transforms the equilibrium search problem in multi-stage traffic flow models into a unified convex optimization problem, providing a new theoretical foundation and numerical methods for practical applications. This achievement is not only significant in theory but also demonstrates good performance in practice.

Sufficient conditions for multi-stages traffic assignment model to be the convex optimization problem

On the Primal and Dual Formulations of Traffic Assignment Problems with Perception Stochasticity and Demand Elasticity

Simulation-based Optimization of Mixed Road Pricing Policies in a Large Real-world Network

Surrogate-Based Optimization for Solving a Mixed Integer Network Design Problem

Social Welfare Maximization of Multimodal Transportation

Necessary and Sufficient Condition for Non-Concave Network Utility Maximisation.

Finding Equilibria in the Traffic Assignment Problem with Primal-Dual Gradient Methods for Stable Dynamics Model and Beckmann Model

Combined Traffic Assignment Model with Game Theory

Primal-Dual Gradient Methods for Searching Network Equilibria in Combined Models with Nested Choice Structure and Capacity Constraints

Generalization of Beckmann's Transformation for Traffic Assignment Models with Asymmetric Cost Functions

Efficient Computation of User Optimal Traffic Assignment Via Second-Order Cone and Linear Programming Techniques

Network equilibrium with combined modes: models and solution algorithms

Multicommodity routing optimization for engineering networks

Perturbed utility stochastic traffic assignment

Variational Inequality Equilibrium Model for Multiclass and Multicriteria Stochastic Traffic Networks with Elastic Demands

METHODS FOR THE SOLUTION OF THE COMBINED TRIP MATRIX ESTIMATION AND STOCHASTIC USER EQUILIBRIUM ASSIGNMENT PROBLEM

A convex formulation of traffic dynamics on transportation networks

An Equilibrium Dynamic Traffic Assignment Model with Linear Programming Formulation

A Fully Distributed Traffic Allocation Algorithm for Nonconcave Utility Maximization in Connectionless Communication Networks.

Stochastic User Equilibrium Assignment Based on Traffic State

Network Design for the Traffic Assignment Problem with Mixed-Integer Frank-Wolfe