Pavement ME JPCP Transverse Cracking Model Calibration and Design Catalog Framework (Version 2.5.5)

The Mechanistic-Empirical Pavement Design Guide (MEPDG) is a comprehensive method, including models and guidance, developed in 2002 by the American Association of State Highway and Transportation Officials (AASHTO) to analyze and design both flexible and rigid pavements. The MEPDG is implemented in a software called Pavement ME. The MEPDG models were calibrated using data from the Long-Term Pavement Performance (LTPP) sections from throughout the United States, including some from California. The MEPDG recommends that nationally calibrated models be validated using local data and, if necessary, recalibrated. This recommendation is particularly applicable to the Caltrans road network, considering the climate and materials differences between California and the rest of the nation. The first step in recalibrating Pavement ME is to perform a sensitivity analysis to identify which variables are most important and to look for results that do not match expected performance. This was the subject of a previous report titled Pavement ME Sensitivity Analysis. Based on the sensitivity analysis, the decision was made to perform a new calibration of the MEPDG models as implemented in Pavement ME software. A new approach was developed for the calibration. This new approach uses network-level performance data from the pavement management system (PMS) with orders of magnitude more observations and length of pavement than are used in the traditional approach and in the national calibration of the MEPDG models. The framework does not require sampling of materials from specific sections in the network. Rather, it uses the statewide median values from mechanistic testing from a representative sample of materials across the network. Variability of performance and reliability of design (probability that the design will meet or exceed the design life) is accounted for through separate consideration of within-project and between-project variability. The calibration reduced significant bias in the application of the nationally calibrated models to California. This report presents the results of the application of the new procedure to calibrate the Pavement ME transverse cracking model for jointed plain concrete pavements (JPCP). The California pavement management system (PaveM) database—with about 4600 lane-miles of JPCP built on 446 lane replacement projects completed between 1947 and 2017—was used to calibrate the transverse cracking model in Pavement ME. The nationally calibrated Pavement ME transverse cracking model prediction on the PaveM performance database has bias and standard error of 13.3% and 23.03%, respectively. After calibration, the bias and standard error of the locally calibrated model decreased to 0.039% and 5.69%, respectively.