Introduction of an Expert System Based Traffic Signal Analysis Tool

An Expert System based traffic signal analysis tool is introduced in this paper. The objective of this tool is to assist traffic operators and engineers in analyzing traffic network data, identifying the anomalies or various technical issues, reasoning the cause and consequences, generating a set of response solutions including signal timing adjustments, and evaluating the effectiveness of the implemented solutions for future use. The tool makes full use of available computing power, collects system-wide historical data and real-time data, as well as the operation knowledge built from previous operations, to conduct comprehensive on-line analysis of traffic events. It is expected to get “smarter” over time by accumulating the operational knowledge. This tool, developed with FHWA’s sponsorship, could be an early step towards automated TMC operations. An example scenario of a freeway incident on I-95 causing traffic diversion to the parallel US 1 arterial is presented. The tool was collecting real-time data and found out significant congestion and signal timing issues in US 1. By comparing the real-time data and historical data, and applying rule-based and case-based knowledge, a comprehensive analysis was conducted to generate reasonable solutions and simulation was conducted for assessing the effectiveness of the solutions. A set of ranked solutions was then proposed to the user interface. The after implementation data would be collected and stored in the system for future use. Simulation results show that the suggested solution can reduce travel times of the diverted vehicles and increase the overall throughputs.

  • Supplemental Notes:
    • This paper was sponsored by TRB committee AHB25 Traffic Signal Systems.
  • Corporate Authors:

    Transportation Research Board

    500 Fifth Street, NW
    Washington, DC  United States  20001
  • Authors:
    • Zhao, Xiaoliang
    • Hsu, Chung-Jen
    • Xu, Roger
    • Huang, Peter
  • Conference:
  • Date: 2015

Language

  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; Maps; References; Tables;
  • Pagination: 24p
  • Monograph Title: TRB 94th Annual Meeting Compendium of Papers

Subject/Index Terms

Filing Info

  • Accession Number: 01556988
  • Record Type: Publication
  • Report/Paper Numbers: 15-2917
  • Files: TRIS, TRB, ATRI
  • Created Date: Mar 16 2015 11:35AM