Comparison of measured vs. predicted performance of jointed plain concrete pavements using the Mechanistic–Empirical Pavement Design Guideline

Abstract

This research evaluates the ability of the Mechanistic–Empirical Pavement Design Guide (MEPDG) to accurately predict the performance of jointed plain concrete pavements (JPCPs). This is accomplished by comparing predicted performances with observed performances for the in-service mainline test cells at Mn/ROAD. These comparisons indicate that MEPDG performance predictions for JPCP are most accurate when the default (constant) built-in equivalent temperature difference of − 5.5°C is used instead of a site-dependent value. It appears that significant portions of the error of estimation can be explained by the sensitivity of the performance models to variability in hardened concrete properties (modulus of rupture, modulus of elasticity and coefficient of thermal expansion) and pavement structural features (slab thickness, joint spacing, subbase type and bond condition). Predictions of slab cracking were found to be highly sensitive to these parameters. In addition, the MEPDG cracking model seemed not to fit local cracking observations for the Minnesota test cells. New calibration factors are needed to more accurately predict Minnesota JPCP slab cracking. This study also included comparisons of predicted service lives for the Mn/ROAD test cells using different design methodologies and as-built input parameters. In most cases considered, the MEPDG predicted longer service lives than did the 1993 AASHTO procedure. The MEDPG also predicted longer service lives than the PCA procedure for the 5-year cells but shorter service lives for the 10-year cells. This infers that, when holding service life constant, the MEPDG generally results in thinner concrete pavement sections than the 1993 AASHTO procedure.

Keywords:

• MEPDG
• built-in temperature difference
• jointed plain concrete pavement
• cracking
• service life
• performance prediction

Acknowledgements

The authors would like to gratefully acknowledge the Federal Highway Administration for its financial support. The contributions of Mr James Sherwood and Ms Katherine Petros to this research effort are greatly appreciated. The authors would also like to extend their gratitude to Mr Timothy Cline for providing some of the performance data for the Mn/ROAD test sections. Finally, the authors would like to thank Mr Lev Khazanovich for providing the climatic database and information on the traffic data used in this study.