Comparison of Flexible Pavement Designs: Mechanistic-Empirical (NCHRP1-37A) Versus Empirical (AASHTO 1993) Flexible Pavement Design Using Available Local Calibration Models

This paper focuses on conducting a comparative study between mechanistic empirical (NCHRP1-37A) design method and empirical 1993 AASHTO design method. The flexible pavement sections selected for the study are from different regions in the United States with varying climatic conditions. The typical data required for defining and characterizing the pavement structure were gathered from FHWA's Long-Term Pavement Performance (LTPP) program. The pavement sections are initially designed using 1993 AASHTO design method at various conditions including different traffic levels, percentage of overloaded vehicles, and environmental conditions. Both default and available local calibration models for the selected regions are used to predict the performance of flexible pavement by mechanistic empirical pavement design guide (MEPDG) software. The results reveal that despite selecting the same serviceability loss in designing flexible pavement using 1993 AASHTO empirical method, the obtained results showed a noticeable variation in the predicted pavement performance while analyzing these sections using MEPDG design approach. The computed performance of AASHTO designed pavements using local and default calibration models are significantly different. Furthermore, as the percentage of overloaded trucks increases, pavement distresses computed at the end of the service life exceed the permissible distress limit in most cases. For pavements located in regions with higher average temperatures, the pavement performance exhibits the least total rutting performance. Finally, the worst pavement performance for alligator cracking distresses is observed for pavement sections designed with a strong subgrade for moderate and high traffic volume in warm climates.