Characterization of Hamburg Wheel Tracking Device Testing Results

Abstract
The Hamburg Wheel Tracking Device was developed as specification requirement for some of the heavy-duty road to measure rutting (potential) and stripping (moisture susceptibility). TxDOT adopted cylindrical specimens for testing with the HWTD (Hamburg Wheel Tracking Device) manufactured by PMW, Inc. Traditionally, the rut depth has been represented by the largest impression point which occurs randomly along the 11 points. The maximum impression is randomly located and more likely to be in the middle part of specimen. It is difficult to characterize with a single theoretical distribution such as a normal or other such function that defines one central tendency. This paper detailed the spatial distributions of Hamburg Wheel Tracking Device Test Results and demonstrated its predictive capability using statistics tools. In addition, the author conducted statistical analysis of all the test points in four patterns, and thus a comparison between all patterns were carried on. Finally, a more reasonable representative pattern is proposed. Central to the approach developed in this research was utilizing a statistics tool to better characterize the deformation. It is certainly more straightforward to represent them with maximum deformation rather than other averaged points. But in case where small amounts of observed data are available, large variance could prove prohibitive to reliable analysis, which will finally result in great uncertainty in predicting performance. Average of the next four maximum deformations presents a relative smaller variance compared to other method. In case observed data are limited, it will provide a relatively better choice.