Road network reserve capacity with stochastic user equilibrium

Abstract

To relax the strong assumption associated with User Equilibrium (UE) in the previous research of network reserve capacity conducted by Gao and Song (2002), this paper assumes that the drivers all make route choices based on Stochastic User Equilibrium (SUE) principle. Similarly, two bi-level programs are formulated to study the network reserve capacity with SUE problem. The first bi-level program is developed to maximize the network reserve capacity by optimizing signal settings while the traffic demands are reassigned by SUE model. The second program extends the research with Continuous Network Design (CND) problem to find the maximum possible increase in reserve capacity through optimizing allocation of network investment. Two methods, i.e. the sensitivity analysis-based method and Genetic Algorithm (GA), are detailed formulated to solve the bi-level reserve capacity problem. Application of the proposed model and its solution algorithms on two numerical examples find that the network reserve capacity does not always increase with improved quality of drivers’ information. Besides, CND can not only help to increase network reserve capacity, but also can help to make more use of physical capacity of road network at Deterministic User Equilibrium (DUE) condition, thus reduces the difference of reserve capacity between the assumptions of SUE and DUE.

First Published Online: 27 Mar 2015

Keyword : reserve capacity, stochastic user equilibrium, sensitivity analysis, genetic algorithm, continuous network design