Abstract:This paper considers the bilevel mixed network design problem (MNDP) used in dynamic traffic assignment (DTA) and the simulation-based optimization solution. The upper level of the bilevel MNDP minimizes the network cost in terms of average travel time by the expansion of existing links and the addition of new candidate links. The lower level is a dynamic user-optimal condition that can be formulated as a variational inequality problem. The MNDP simultaneously finds optimal capacity expansions of existing links and new link additions. A surrogate-based optimization (SBO) framework is proposed for solving the MNDP that is characterized by expensive-to-evaluate objective functions. Because simulation was applied to evaluate those objective functions, additional complexity arose from the fine-grained representation of traffic dynamics in large-scale networks, which were not fully considered by the traditional static user equilibrium. SBO methods enjoy both the advantages of simulation in time-varying network performance evaluation and the efficiency of mathematical optimization. To be more specific, SBO produces computational time savings by exploring the input–output mapping surface in a more systematic and efficient way. For demonstrative purposes, a case study was conducted on the large-scale Montgomery County network in Maryland. In this example, a mesoscopic simulation-based DTA model, DTALite, was used to evaluate the system performance in response to various network design strategies. Results showed that the optimal investment with a moderate budget could reduce 17.73% of the network average travel time in the morning peak. The proposed framework is a general approach, which is ready for application to either continuous or discrete network design problems.

The Method of Successive Averages and Method of Successive Weighted Averages for Logit-Based Stochastic User Equilibrium Assignment

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

A Modified Stochastic User Equilibrium Based Back-Propagation Method of Transportation Network State Estimation

Logit-Based Stochastic User Equilibrium Problem for Entry-Exit Toll Schemes

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

Joint BS-User Association, Power Allocation, and User-Side Interference Cancellation in Cell-free Heterogeneous Networks

Origin-Based Partial Linearization Method for the Stochastic User Equilibrium Traffic Assignment Problem

A faster path-based algorithm with Barzilai-Borwein step size for solving stochastic traffic equilibrium models

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

Stochastic User Equilibrium Assignment with Elastic Demand and Its Application

Stochastic User Equilibrium Assignment In Schedule-Based Transit Networks With Capacity Constraints

Weibull Distribution Based Multi-Class Stochastic User Equilibrium Model

Stochastic User Equilibrium Model with a Bounded Perceived Travel Time

Maximizing Network Throughput under Stochastic User Equilibrium with Elastic Demand

Stochastic User Equilibrium Assignment Based on Traffic State

A COMPARISON OF NETWORK INEFFICIENCY BETWEEN LOGIT-BASED STOCHASTIC USER EQUILIBRIUM AND DETERMINISTIC USER EQUILIBRIUM

Joint User Association and Resource Allocation for Energy-Efficient Multi-Stream Aggregation

A Day-to-Day Dynamical Approach to the Most Likely User Equilibrium Problem

Energy-Efficient User Association and Resource Allocation for Multistream Carrier Aggregation.

Stochastic User Equilibrium and System Optimum with Heterogeneous Path Choice Sets

User Equilibrium Traffic Assignment with Morgenstern Utility Theory