O-D MATRIX PRODUCTION FROM CORDON SURVEY DATA

This article describes the methodology utilised in a study commissioned by Merseyside County Council for Applied Research of Cambridge Ltd and Marcial Echenique & Partners jointly to develop, test and implement a suite of programs to process a freight flow survey for Merseyside. The purpose of the study was to take the file of information collected at the roadside interviews giving the origins and destinations of the sample of lorries interviewed, and then to produce from this an estimated expanded matrix of freight flows between all zone pairs in Merseyside. This involved a number of stages. Firstly, for zone pairs whose trips are adequately captured in the survey it is necessary to scale up the trips using an expansion factor to take account of the size of the sample. Next, it is necessary to average across cordons those trips which are captured by more than one cordon in order to avoid double-counting. Having now produced an expanded partial matrix a gravity-type model can be calibrated for this partial matrix and then used to produce estimates for the flows between the unobserved zone pairs in a consistent fashion. The final part of the paper describes a test of the methodology which was carried out. This demonstrates that in a situation where 27 per cent of the zone pairs are missing, the estimated trips suffer little in accuracy from those estimated using the full matrix. (Author/TRRL)

  • Availability:
  • Corporate Authors:

    Printerhall Limited

    29 Newmart Street
    London W1P 3PE,   England 
  • Authors:
    • Echenique, M L
    • WILLIAMS, I N
  • Publication Date: 1982-12

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00372533
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • Report/Paper Numbers: HS-034 256
  • Files: ITRD, TRIS
  • Created Date: May 31 1983 12:00AM