Identifying reproducible macroscopic traffic patterns in a year-long data set

The macroscopic Fundamental Diagram (MFD) provides a novel perspective on urban traffic that facilitates new policies and strategies to cope with recurring congestion. The MFD is assumed to be well-defined and reproducible, but, so far, no long-term empirical evidence for this shape assumption exists. In this paper, the authors use an extensive one-year traffic data set from Lucerne, recent advances in modeling the MFD shape as well as established similarity measures (Dynamic time warping and Fr\'{e}chet distance) and k-medoid clustering to investigate this assumption. The authors first find that they can reduce the complexity of urban traffic throughout the year to only three or five clusters depending on the selected similarity measure. Furthermore, the authors reveal that the MFD shape in the loading phase of the network is very similar across the observed inflow patterns over the course of a year, but less so for the unloading and full-day MFD.

  • Supplemental Notes:
    • This paper was sponsored by TRB committee AHB45 Standing Committee on Traffic Flow Theory and Characteristics.
  • Corporate Authors:

    Transportation Research Board

  • Authors:
    • Ambuhl, Lukas
    • Loder, Allister
    • Leclercq, Ludovic
    • Axhausen, Kay W
    • Menendez, Monica
  • Conference:
  • Date: 2019


  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; References;
  • Pagination: 8p

Subject/Index Terms

Filing Info

  • Accession Number: 01697740
  • Record Type: Publication
  • Report/Paper Numbers: 19-03614
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 7 2018 9:36AM