Validation of Streetlight Insight® 2021 Vehicle Volume Metrics in Maine

Details

• Creators:
  Grond, Kathryn ; Rubin, Jonathan
• Corporate Creators:
  University of Maine. Transportation Infrastructure Durability Center (TIDC) ; University of Maine. Margaret Chase Smith Center for Public Policy
• Corporate Contributors:
  Maine. Department of Transportation ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology ; United States. Department of Transportation. University Transportation Centers (UTC) Program
• Subject/TRT Terms:
  Analysis Datasets Estimates Low Volume Roads Pedestrians Regression Analysis Statistical Analysis Time Intervals Traffic Lanes Validation
• Publication/ Report Number:
  69A3551847101
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States ; Maine
• Edition:
  Final Report
• Corporate Publisher:
  University of Maine. Transportation Infrastructure Durability Center (TIDC)
• Abstract:
  The University of Maine's Margaret Chase Smith Policy center undertook a validation effort on behalf of MaineDOT to better understand the accuracy of StreetLight Insight's vehicle volume metrics for monthly, daily, and hourly time periods. We examine the impact of two characteristics of the traffic counter location on the accuracy and precision: the annual average daily traffic (AADT) and MaineDOT's factor group classification. The factor groups were created to classify roads with high and low seasonal variability of traffic volume due to tourism. We also present a preliminary analysis of StL's turning movement counts (TMC). We employ similar statistical methodologies to those published by StL and the FHWA (Federal Highway Administration) to validated StL's AADT metrics. We find that the accuracy of the MADT and DOW estimates are very similar to the AADT estimates and should therefore be considered sufficiently accurate in most cases. The use ofStL's MADT estimates for low-volume roads (under 5,000 AADT) will require judgment on the part ofMaineDOT's staff and consultants before use in transportation planning. The Levene's test of equality of variance finds that the variance of StL metrics is different between Factor Group I & II, meaning that Factor Group two is significantly affecting the precision of the MADT, whereas the AADT Range does not significantly change the variance. In the analysis of the Day of the Week (DOW estimates), there is a significant difference in accuracy between weekdays (Mon-Fri) and weekends (Sat-Sun). The analysis of the Hour of the Day (HOD) estimates found that StL underestimates traffic volume in the morning, and increasingly overestimates traffic in the afternoon and evening. Using the average difference of short-term counts of 1.8% (as a percent of intersection traffic), 62.5% of turning movements could be considered within an acceptable error margin.
• Format:
  PDF
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:1312489d751b28ee54eacfb1671445ca2ad9965835f678d3d4ac03b12540e5fbcbd82b35f8df4048fa5e0e2cacd775b3149705fb5e12c5d1d6b6a73c2b247156
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/88949/dot_88949_DS1.pdf
• File Type:
  [PDF - 34.20 MB ]