Regional Calibration of Permanent Deformation Performance Models for Rehabilitated Flexible Pavements

The Mechanistic-Empirical Pavement Design Guide (MEPDG) is a pavement analysis system that has been gaining popularity as an analysis tool for new and rehabilitated pavement structures. A handful of states are already using the MEPDG for design. The performance models used in the MEPDG, developed under the National Cooperative Highway Research Program Projects NCHRP 1-37A and 1-40D, have been calibrated using sections spread throughout the United States. It is necessary to calibrate these models for specific states and regional conditions due to the differences in terms of materials, environmental conditions and construction practices. In general, a pavement design based on the nationally calibrated MEPDG will result in either an overestimate or an underestimate of the pavement layer thicknesses because of systematic errors due to local differences. An average does not necessary represent any specific condition. This deficiency calls for a local calibration of the performance models in the MEPDG, so that it can be used to design pavements at a regional level. Several states have already done so; others are in the process. This paper documents some of the calibration work in Texas. This study focuses on finding bias correction factors for systematic differences due to traffic, climatic conditions, and material properties that govern the initial condition as well as the progression of rutting over time. Bias correction factors for rutting in the subgrade were assumed based on inputs from other similar studies, which are properly referenced. The paper proposes a set of Level 2 bias correction factors which are good for analysis of rehabilitated pavements for the regions specified. It has been also observed that pavements rehabilitated in warmer climatic areas will experience lower initial rutting, but a higher rate of increase in the rut depth with time compared to those constructed in colder climatic areas. It should be emphasized that calibration factors for new and rehabilitated pavements are significantly different.