Implementation of Statistical Analysis Makes IDOT Traffic Counts More Efficient

Last year, ICT completed a project on behalf of the Illinois Department of Transportation (IDOT) to evaluate the feasibility of decreasing the time frame for traffic counts. IDOT had been seeking to extend an existing waiver from the Federal Highway Administration (FHWA) on Highway Performance Monitoring System (HPMS) sections to conduct 24-hour traffic counts instead of the 48-hour counts required by FHWA. These counts are used to determine the average annual daily traffic (AADT) for roads.

The waiver extension would then allow the continued efficiency and integration within the existing traffic counting program. Using a shorter time frame offers a number of benefits:

  • IDOT can employ fewer consultants and better coordinate its own staff.
  • A higher set of counts could be produced in the same time frame, giving IDOT a better understanding of the distribution of traffic and reducing the per-count cost.
  • The shorter time frame would allow the same equipment to be used at different locations.
  • A higher frequency of samples could be obtained because personnel could go to the same site more than once.

FHWA has maintained a 48-hour rule for these traffic counts for decades, and they seem to be a sound practice. Therefore, to demonstrate that 24-hour time frames were viable, the research team needed to show that a 24-hour count was just as reliable as a 48-hour count.

James Hall, associate professor emeritus in the Department of Management Information Systems at the University of Illinois Springfield (UIS), was principal investigator for the project. He explains: “This project reviewed 2008–2012 data from 103 automated traffic recorder sites over the typical IDOT traffic counting season. To represent typical counting days, the months April through October and the days Monday through Thursday were selected for analysis. One challenge was selection of an appropriate statistical analysis method. Chung-Hsien Sung, chair of the Department of Mathematical Sciences at UIS and a critical member of the project team, employed a paired t-test statistical approach to compare the predicted 48-hour count, based on the 24-hour count for the first day, with the actual 48-hour count.”

Map indicating data collection sites for the study.

Hall and his team determined that, with the application of appropriate daily traffic count adjustment factors, the two count durations were statistically comparable, and IDOT was able to obtain the waiver from FHWA.

IDOT Planning & Systems Section Chief William Morgan was the Technical Review Panel chair for this project and led IDOT’s efforts in securing the waiver. “This time savings on the counts allows IDOT to better integrate the counts into the annual IDOT counting program and allow for additional counts to be conducted with the same resources by bringing the HPMS sections into alignment with the other locations. IDOT is able to conduct around 20,000 annual traffic counts with this waiver,” Morgan says.

He adds, “Traffic data, and its use in calculating vehicle miles of travel, is a major economic indicator locally, statewide, and nationally. There is no question that accuracy is important. The challenge then, as now, was how to align resources to efficiently collect high-quality traffic data within shrinking budgets.”

The full project report, Analysis of 24-Hour Versus 48-Hour Traffic Counts for HPMS Sampling, is available on ICT’s website. The current annual traffic statistics for Illinois can be found on the reports tab on the IDOT Internet site.

Analysis of traffic, by day, of a typical downstate Illinois rural interstate.

Posted July 14, 2015