A Statistical Analysis of Hamburg Wheel Tracking Device Testing Results

Runhua Guo,Jorge A. Prozzi
DOI: https://doi.org/10.1080/14680629.2009.9690249
2009-01-01
Road Materials and Pavement Design
Abstract:ABSTRACT The Hamburg Wheel Tracking Device (HWTD) was originally developed to assess the rutting potential and stripping (moisture susceptibility) of hot-mix asphalt mixes (HMA). Several Departments of Transportation (DOT) have recently incorporated HWTD testing for mix acceptance. DOT adopted cylindrical specimens for test because specimens can readily be prepared by means of the Superpave Gyratory Compactor (SPC) and also field pavement cores can be tested. Traditionally, the rut depth development has been represented by the largest deformation. Although the point of maximum deformation is often located towards the center of the specimen, in general, it is randomly located. Most importantly, its location is affected by test characteristics such as specimen preparation, material type, operator, etc. This leads to the test result not necessarily measuring a material characteristic or property but often the characteristics of the testing conditions. Current DOT specification is written based on the maximum deformation at any of the 11 points measured. This statistics is highly variable and often characterize test variability rather than material properties. If maximum deformation were to be used in the specification, it is recommended that at least four specimens be tested. Alternatively, more stable statistics should be used such as average deformation of some trimmed means. This paper explains the steps followed for the development of the Hamburg Wheel Tracking Device Database and reports on some of the data processing. In addition, some basic statistical analysis was carried out that highlighted some interesting results, which could help DOT in reviewing current specifications.
What problem does this paper attempt to address?