Real-Time Anomaly Detection System for Signalized Intersections

Automated Traffic Signal Performance Metrices (ATSPMs) rely on high-resolution controller data to support signal operations and maintenance. While existing ATSPM Watchdog tools can identify certain detector malfunctions, they typically rely on static thresholds set across the entire system, limiting their ability to detect issues promptly and adapt to site-specific traffic patterns. This project develops and evaluates a statistical, spatiotemporal anomaly detection framework that leverages historical detector behavior and agreement among neighboring detectors to improve alert timeliness and accuracy. The research will characterize practitioner needs, define normal detector behavior, develop and test statistical detection and classification methods, and quantify benefits relative to static thresholds. Results will inform a roadmap for potential real-time implementation within Virginia Department of Transportation (DOT) systems.

Language

• English

Project

• Status: Active
• Funding: $444,060.00
• Contract Numbers:

  128794

• Sponsor Organizations:

  Virginia Transportation Research Council

  530 Edgemont Road
  Charlottesville, VA  United States  22903
• Managing Organizations:

  Virginia Transportation Research Council

  530 Edgemont Road
  Charlottesville, VA  United States  22903
• Project Managers:

  Lan, Chien-Lun

• Performing Organizations:

  Virginia Transportation Research Council

  530 Edgemont Road
  Charlottesville, VA  United States  22903

  University of Virginia, Charlottesville

  Center for Transportation Studies
  P.O. Box 400742, Thornton Hall, D228
  Charlottesville, VA  United States  22903
• Principal Investigators:

  Lan, Chien-Lun

  McNamara, Margaret

• Start Date: 20260301
• Expected Completion Date: 20280531
• Actual Completion Date: 0

Subject/Index Terms

• TRT Terms: Flaw detection; Signalized intersections; Statistical analysis; Traffic signal control systems
• Identifier Terms: Virginia Department of Transportation
• Geographic Terms: Virginia
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management;

Filing Info

• Accession Number: 01981045
• Record Type: Research project
• Source Agency: Virginia Department of Transportation
• Contract Numbers: 128794
• Files: RIP, STATEDOT
• Created Date: Feb 24 2026 10:25AM