In the Glasgow experiment for computer control of traffic it has been sound that, with normal traffic conditions, the use of optimum fixed time plans, prepared off-line using historical data, gave average journey times in the area which could be decreased neither by superimposing local responsive control at the signals in the network nor by the use of a very flexible platoon identification system. The fixed time scheme incorporated plans calculated for a small number of two-hourly periods which were changed according to time of day. As a consequence of this work and simular work in West London the fixed time approach is now the standard for Britain. This paper investigates the maximum benefits in terms of reduced journey times which could be expected on a typical day, when operating either ideal on-line signal plan generation or perfect short term plan selection from a large number of stored plans. It is assumed initially that there is perfect flow prediction with perfect plan changing and then levels of prediction error and plan change losses which might be expected are considered. In no case do the results indicate that a decrease in average journey time would have been detected if a floating car survey had taken place, on the day investigated, to compare these techniques with the standard approach. /Author/

  • Corporate Authors:

    Newcastle-upon-Tyne University, England

    Newcastle-upon-Tyne,   England 
  • Authors:
    • Ferguson, J A
    • Jenkins, I A
  • Publication Date: 1973-11


  • English

Media Info

  • Features: Figures; References; Tables;
  • Pagination: 13 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00080848
  • Record Type: Publication
  • Source Agency: Transport and Road Research Laboratory (TRRL)
  • Report/Paper Numbers: Working Paper No. 1
  • Files: ITRD, TRIS
  • Created Date: May 29 1975 12:00AM