State DOT Implementation of Statistical Analysis and Percent Within Limits

Verification testing, conducted by many state agencies, should serve an integral role in a quality assurance program. Many agencies seek a balance of physical and financial resources, as resources are routinely a limiting factor. For the Wisconsin Department of Transportation (WisDOT), it is datasets collected by both the contractor (quality control, QC) and department (quality verification, QV) at differing frequencies compared in a statistically valid method that provide balance. Statistical analyses, specifically F- and t-testing, have been chosen by WisDOT to accompany a Percent Within Limits (PWL) specification for hot mix asphalt (HMA). The analyses evaluate QC and QV data variances, means, standard deviations, and so forth, and determine whether it is reasonable to assume the test results of each dataset come from the same population, that is, are representative of the same material. If comparable, the more frequent QC data are considered verified and subsequently used in PWL calculations to approximate the percentage of material within specification limits. Anything less than comparable is subject to further investigation, for example, a dispute resolution process, to determine which test results most reliably represent the physical properties of the material, and are thus justifiably used in latter calculations. Such an approach has allowed WisDOT to adjust pay in accordance with material quality and consistency, ultimately connecting pay to anticipated long-term performance. This modified system of monitoring HMA materials has motivated multiple changes to industry processes and planning, and resulted in more consistent pavements, which should reduce the potential for isolated premature distresses.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01693791
  • Record Type: Publication
  • Report/Paper Numbers: 19-04277
  • Files: TRIS, TRB, ATRI
  • Created Date: Feb 1 2019 1:40PM