01626265

Component

The sensitivity analysis-based linear approximation (SA-LA) method has two primary issues that cause the poor estimation of user equilibrium (UE) flows under large-scale perturbations. First, the linear approximation of perturbed UE link flows cannot capture the nonlinear UE flow variation with respect to perturbed parameters (such as link capacity, signal splits). Second, the SA-LA method uses the maximum set of equilibrated and linearly independent paths to calculate the gradient of equilibrium path flow and link flow in perturbation at the unperturbed state. The gradient cannot characterize the flow variation due to the change of the maximum set of equilibrated and linearly independent paths induced by larger-scale perturbation. To address the irst issue, a piecewise linear approximation technique is proposed to improve the estimation accuracy of the SA-LA method, by dividing the large-scale perturbation into a sequence of small- scale perturbations and sequentially estimating the perturbed UE path flows. To address the second issue, a revised sensitivity analysis-based method is developed upon the piecewise linear approximation technique. The authors label this method as the RSA-PLA method. The RSA-PLA method is constructed on a path flow variation strategy that specifies the adjustment of the maximum set of equilibrated and linearly independent paths. This strategy also regulates the path flow variations in three disjoint path sets, such that the proposed RSA-PLA method can provide more accurate link flow estimates under large-scale perturbation scenarios. A numerical example demonstrates that the proposed RSA-PLA method is superior to the SA-LA method in terms of estimation accuracy.

This paper was sponsored by TRB committee ADB30 Standing Committee on Transportation Network Modeling.

01618707

17-00779

English

21p

2017

Transportation Research Board 96th Annual Meeting

Digital/other

Figures; References; Tables

Approximation (Mathematics); Equilibrium (Systems); Link volume; Perturbations; Sensitivity analysis

Path flows; User equilibrium

Highways; Planning and Forecasting

Transportation Research Board Annual Meeting 2017 Paper #17-00779

TRIS, TRB, ATRI

Dec 8 2016 10:11AM