Automated Identification of Traffic Detector Malfunctions

To assess the quality of data provided by traffic detectors, and therefore detector health, research is necessary to bridge the gap between existing traffic theory and pattern recognition that can identify poor performance (through comparison of data to expected norms) in an automated fashion. To address this gap, the proposed objective is to develop a reliable and robust method of determining poor performance of a traffic detector based solely on historical data and traffic flow theory. It is proposed that this method will work at isolated signalized intersections, using data only from that intersection’s detectors for evaluation. Additionally, a system design of this method will be developed to assist Oregon Department of Transportation (ODOT) with implementation of the method.

Project

• Status: Active
• Funding: $193000
• Contract Numbers:

  SPR 837

• Sponsor Organizations:

  Federal Highway Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Managing Organizations:

  Oregon Department of Transportation

  555 13th Street NE
  Salem, OR  United States  97301
• Project Managers:

  Li, Joe

• Performing Organizations:

  Northern Arizona University

  Civil and Environmental Engineering
  PO box 15600
  Flagstaff, Arizona  United States  86011
• Principal Investigators:

  Smaglik, Edward

• Start Date: 20200101
• Expected Completion Date: 20211231
• Actual Completion Date: 0
• USDOT Program: Transportation, Planning, Research, and Development

Subject/Index Terms

• TRT Terms: Automation; Data analysis; Data quality; Performance measurement; Traffic flow theory; Vehicle detectors
• Identifier Terms: Oregon Department of Transportation
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management;

Filing Info

• Accession Number: 01725109
• Record Type: Research project
• Source Agency: Oregon Department of Transportation
• Contract Numbers: SPR 837
• Files: RIP, STATEDOT
• Created Date: Dec 12 2019 1:36PM