Modeling heterogeneous traffic flow: A pragmatic approach

Zhen (Sean) Qian, Jia Li, Xiaopeng Li, Michael Zhang, Haizhong Wang

Research output: Contribution to journalArticlepeer-review

39 Scopus citations

Abstract

Modeling dynamics of heterogeneous traffic flow is central to the control and operations of today's increasingly complex transportation systems. We develop a macroscopic heterogeneous traffic flow model. This model considers interplay of multiple vehicle classes, each of which is assumed to possess homogeneous car-following behavior and vehicle attributes. We propose the concepts of road capacity split and perceived equivalent density for each class to model both lateral and longitudinal cross-class interactions across neighboring cells. Rather than leveraging hydrodynamic analogies, it establishes pragmatic cross-class interaction rules aspired by capacity allocation and approximate inter-cell fluxes. This model generalizes the classical Cell Transmission Model (CTM) to three types of traffic regimes in general, i.e. free flow, semi-congestion, and full congestion regimes. This model replicates prominent empirical characteristics exhibited by mixed vehicular flow, including formation and spatio-temporal propagation of shockwaves, vehicle overtaking, as well as oscillatory waves. Those features are validated against numerical experiments and the NGSIM I-80 data. Realistic class-specific travel times can be computed from this model efficiently, which demonstrates the feasibility of applying this multi-class model to large-scale real-world networks.

Original languageEnglish
Pages (from-to)183-204
Number of pages22
JournalTransportation Research Part B: Methodological
Volume99
DOIs
StatePublished - May 1 2017

Keywords

  • Data driven
  • Fundamental diagram
  • Heterogeneous traffic flow
  • LWR
  • Multi-class
  • Multi-modal
  • NGSIM

Fingerprint

Dive into the research topics of 'Modeling heterogeneous traffic flow: A pragmatic approach'. Together they form a unique fingerprint.

Cite this