How to quantify the quality of matrix estimation with traffic counts?

One classic way to calibrate traffic models is to compare the calculated flow values with traffic counts and to estimate the trip matrix which produces the smallest difference between assignment and counts. The results are normally good to very good for links with traffic counters. But the question arises: does the calibration also improve the quality of the model on links without counters? The authors developed a method to answer this question and tested it with online traffic data from the City of Zurich. The method can be easily implemented in any traffic model which calibrates matrices.

Language

  • English

Media Info

  • Pagination: 4p
  • Monograph Title: Moving towards an integrated Europe: intelligent transport systems and services: congress and exhibition, 24-26 May 2004, Budapest, Hungary

Subject/Index Terms

Filing Info

  • Accession Number: 01390870
  • Record Type: Publication
  • Source Agency: ARRB
  • Files: ATRI
  • Created Date: Aug 23 2012 5:07AM