CONTROL LOGIC FOR TRAFFIC SIGNAL NETWORKS

In an effort to develop and spread the know-how for effective innovative methods of computer traffic control, work is reported of the formulation, development and evaluation of an operational software system that represents a significant step in the direction of attaining self- adjusting real-time control applicable in both light traffic and heavily congested conditions. The ASCOT (Adaptive Signal Control Optimization Techniques) software system described here is being tested and evaluated on an installed control system with actual traffic. The system is designed to achieve very flexible modes of traffic control such as: on-line signal plan computation; cycle-to- cycle split variations; cycle-length adjustment, coupled with offset changes at congested intersections; and phase- skip on designated signal phases in the absence of demand. The logic can adapt to conditions affecting an entire network, such as adverse weather, as well as to localized conditions that may affect only a single intersection. The three levels of control of the logic are summarized in a table. The ASCOT software implementation in San Jose, California is outlined. Details are briefly described of the evaluation of the control logic which is based on comparison of total travel times for all vehicles operating in a controlled network. Travel time data are collected by floating-car technique, and expressed in vehicles per hour as a function of service rate in vehicle-miles per hour. An event recorder is described which is used for the field evaluation studies. It is expected that in the final report, ASCTO's functional specification will be sufficiently detailed to enable the widest possible applications on a variety of digital computers and a variety of city networks.

  • Digital Copy:
  • Supplemental Notes:
    • An NCHRP staff digest of the essential findings from the Interim Report of NCHRP Project 3-18(1), "Improved Control Logic for Use with Computer-Controlled Traffic," by D.W. Ross, T.L. Humphrey, and G.L. Williams, Stanford Research Institute.
  • Corporate Authors:

    Transportation Research Board

    500 Fifth Street, NW
    Washington, DC  United States  20001
  • Publication Date: 1973-5

Media Info

  • Features: Figures; Tables;
  • Pagination: 6 p.
  • Serial:
  • Publication flags:

    Open Access (libre)

Subject/Index Terms

Filing Info

  • Accession Number: 00080824
  • Record Type: Publication
  • Files: TRIS, TRB
  • Created Date: Mar 6 1975 12:00AM