Calibrating Multilane First-Order Traffic Flow Model with Endogenous Representation of Lane-Flow Equilibrium

This study develops a multilane first-order traffic flow model for freeway networks. In the model, lane changes are considered as a stochastic behavior in that an individual driver decreases his or her disutility or cost and are represented as dynamics toward the equilibrium of lane-flow distribution along with longitudinal traffic dynamics. The proposed method can be differentiated from the previous studies in the following points: 1) The motivation of changing the lane is explicitly considered, and it is treated as utility defined by the current macroscopic traffic state; 2) A whole process of lane-changing is computed by macroscopic manner, that is, the extension of kinematic wave theory employing the information technology (IT) principle; and 3) In the model framework, the lane-flow equilibrium curve will be endogenously generated as a result of self-motivated lane changes. In addition, the parsimony representation enables the parameter calibration by using the data collected from the conventional loop detectors. As a result of the parameter calibration using the data collected at four different sites of Chugoku Expressway in Japan, including sag bottleneck, it is revealed that 1) the proposed method can represent the lane-flow distribution of any observation sites with high accuracy, and 2) the estimated parameters can reasonably explain the multilane traffic dynamics and the bottleneck phenomena on uphill of sag section.

Language

  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; References; Tables;
  • Pagination: pp 192-209
  • Monograph Title: Celebrating 50 Years of Traffic Flow Theory: A Symposium. August 11-13, 2014, Portland, Oregon
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 01604712
  • Record Type: Publication
  • Files: TRIS, TRB, ATRI
  • Created Date: Jul 6 2016 11:07AM