The Box Hill study: queue length estimation in scats

This paper describes a field study for the validation of algorithms developed by ARRB for the online estimation of queue lengths in scats. The algorithms include link flow estimation by the recursive least squares (rls) technique, and queue length estimation by platoon dispersion modelling. Two types of queue lengths were measured in the survey. They are the queue length at the start of a green period, and the maximum back of queue as a platoon joins the stationary queue at green start. The survey was carried out on one road link in Canterbury Road, Box Hill, Melbourne, for a measurement period of six hours. The results indicate that the scats counts and flow profiles are sufficiently accurate for online queue length and delay estimation. The rls estimator was successfully employed in link flow estimation. It achieved a high correlation coefficient between the measured and predicted link flow, and a mean percentage difference of eight per cent. Platoon dispersion modelling at a step size of 4 s was also found to produce satisfactory correlation between measured and predicted queue lengths. The mean percentage differences are eight per cent for the queue at green start and 25 per cent for the maximum back of queue. Queue length estimation on a lane by lane basis using the ARRB method was found to be feasible. The success of the study is attributed to the accurate departure flow profiles available from the stop line detectors in scats (a).

• Record URL:
  http://arrbknowledge.com
• Authors:
  • Luk, J Y
  • Besley, M
  • Doughty, B W
• Conference:
  • Australian Road Research Board (ARRB) Conference, 14th, 1988, Canberra
• Publication Date: 1988

Language

• English

Media Info

• Pagination: 157-72
• Monograph Title: Development of comprehensive level of service criteria for the maintenance of traffic light signals, 2: derivation of delay thresholds from traffic behaviour
• Serial:
  • Volume: 14
  • Issue Number: 2

Subject/Index Terms

• TRT Terms: Accuracy; Bunching; Data analysis; Data collection; Mathematical models; Traffic control; Traffic counts; Traffic delays; Traffic flow; Traffic queuing; Traffic signals; Vehicle detectors
• ATRI Terms: Accuracy; Area traffic control; Bunching; Data analysis; Data collection; Delay; Modelling; Queue length; Traffic count; Traffic flow; Traffic signal; Vehicle detector
• Subject Areas: Operations and Traffic Management;

Filing Info

• Accession Number: 01429687
• Record Type: Publication
• Source Agency: ARRB
• Files: ATRI
• Created Date: Aug 24 2012 3:33PM