Examining the scaling effect and overlapping problem in logit-based stochastic user equilibrium models

Anthony Chen, Surachet Pravinvongvuth, Xiangdong Xu, Seungkyu Ryu, Piya Chootinan

Research output: Journal article publicationJournal articleAcademic researchpeer-review

60 Citations (Scopus)

Abstract

The purpose of this paper is to examine the scaling effect and overlapping problem in a route choice context using the logit-based stochastic user equilibrium (SUE) principle to explicitly account for the congestion effect. Numerical experiments are performed on nine models: the deterministic user equilibrium model, the multinomial logit SUE model with and without scaling, the C-logit SUE model with and without scaling, the path-size logit SUE model with and without scaling, and the paired combinatorial logit SUE model with and without scaling. Sensitivity analysis is conducted to examine the effects of route sets, congestion levels, dispersion intensities, and network asymmetries. A real transportation network in the City of Winnipeg, Canada is also used to compare the network equilibrium flow allocations of different SUE models. The results of the sensitivity analysis and the Winnipeg network reveal that both scaling effect and overlapping problem can have a significant impact on the network equilibrium flow allocations.
Original languageEnglish
Pages (from-to)1343-1358
Number of pages16
JournalTransportation Research Part A: Policy and Practice
Volume46
Issue number8
DOIs
Publication statusPublished - 1 Oct 2012
Externally publishedYes

Keywords

  • Extended logit
  • Multinomial logit
  • Route choice model
  • Stochastic user equilibrium

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Transportation
  • Management Science and Operations Research

Cite this