Using ATSPM Data for Traffic Data Analytics

The Utah Department of Transportation (UDOT) performs about 6000 short-duration counts over a three-year cycle to estimate and report annual average daily traffic (AADT) on roadways per Federal Highway Administration (FHWA) requirements. Many of these roadways are in proximity to signalized intersections equipped with radar detectors that provide approach and turning movement counts. This research investigates the use of volume data obtained from traffic-signal radar detectors (i.e. Advance and Matrix detectors) to estimate AADTs and related traffic engineering factors. An assessment of the accuracy of these radar detectors may enable the elimination of selected short-duration counts, and possibly complement Continuous Count Station (CCS) data for estimating seasonal factors. In this research, 27 Matrix detectors and 33 Advance detectors proximate to CCS sites were identified. The hourly count data for an entire year, 2017, were collected from Automated Traffic-Signal Performance Measures (ATSPM) data archive and mapped with the associated CCS hourly counts as ground-truth. An anomaly detection method was implemented to clean the dataset of count data when significant outliers were identified. The accuracy of detector hourly counts was measured using linear regression with and without adjustment factors. The results show that hourly counts from Matrix detectors hourly are more accurate (i.e. average R-squared value of 0.93) than Advance detectors’ hourly counts (i.e. average R-squared value of 0.79). AADTs estimated from Matrix detectors had an 88 percent accuracy, with a range of -21% - +7%. Matrix detectors are sufficiently accurate for estimating AADT as the current methods utilizing short-duration counts have been estimated to be less than 80% accurate. The Matrix detectors are also very accurate in estimating the seasonal factors (i.e. about 97% accurate) and thus can be used to complement CCSs in calculating them. This would be particularly valuable to UDOT in measuring seasonal factors for lower functional class roadways which have sparse coverage by CCS sites.

• Record URL:
  https://drive.google.com/file/d/1CXvMtv2BpattDUq2JsoUZLhdIyt-XdER/view?usp=sharing
• 
• Record URL:
  https://rosap.ntl.bts.gov/view/dot/61721
• Corporate Authors:

  Resource System Group

  Crane Building
  307 W 200 S, Suite 2004
  Salt Lake City, UT  United States  84101

  Utah Department of Transportation

  4501 South 2700 West
  Salt Lake City, UT  United States  84114-8410

  Federal Highway Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Authors:
  • Chamberlin, Robert
  • Fayyaz, Kiavash
• Publication Date: 2019-9

Language

• English

Media Info

• Media Type: Digital/other
• Edition: Final Report
• Features: Figures; References; Tables;
• Pagination: 53p

Subject/Index Terms

• TRT Terms: Accuracy; Annual average daily traffic; Linear regression analysis; Signalized intersections; Traffic counts; Traffic data; Vehicle detectors
• Identifier Terms: Automated Traffic Signal Performance Measure
• Geographic Terms: Utah
• Subject Areas: Highways; Operations and Traffic Management;

Filing Info

• Accession Number: 01725607
• Record Type: Publication
• Report/Paper Numbers: UT-19.22
• Contract Numbers: 19-8273
• Files: NTL, TRIS, ATRI, USDOT, STATEDOT
• Created Date: Dec 18 2019 3:45PM