EFFECTS OF VARIOUS EXTRAPOLATION TECHNIQUES FOR ABBREVIATED DYNAMIC MODULUS TEST DATA ON THE MEPDG RUTTING PREDICTIONS

Abbreviated dynamic modulus test protocols have been used to reduce the time and cost of running the test for routine uses. The abbreviated test results can then be extrapolated to estimate the complete array of the dynamic modulus values required for the Level 1 MEPDG analysis. Multiple approaches are available for the extrapolation depending on a master curve and shift factor functions chosen. The objective of this study was to identify the best practice of extrapolating the abbreviated dynamic modulus data by evaluating the accuracy of performance predictions through MEPDG pavement analysis. Two common sigmoid functional forms (i.e., the MEPDG Sigmoid and a Generalized Logistic functions) were compared for master curve construction. Four different shift factor functions were used to estimate the shift factors beyond the abbreviated temperature limits. The Arrhenius, Williams-Landel-Ferry (WLF), modified WLF by Kaelble, and quadratic functions were investigated. Through extrapolation, significant underestimations of the dynamic modulus values at 54.4 degrees C were obtained from all eight approaches. Among the eight approaches, the Generalized Logistic function coupled with the quadratic shift factor function showed relatively lower extrapolation errors. On the other hand, the MEPDG performance predictions using the eight sets of extrapolated dynamic modulus input showed less than 1 mm differences in the terminal rut depth when compared to the performance predictions using the measured data sets.