Accuracy of the Brazilian mechanistic-empirical method in the analysis of asphalt pavements from two experimental sections

The first Brazilian mechanistic-empirical method was developed for designing asphalt pavements (MeDiNa) between 2014 and 2018, involving 41 test sections implemented and monitored. However, the validation process of the MeDiNa transfer function involved 13 experimental sections. In this study, an experimental segment with a length of 220 meters was implemented on the BR-420 highway in Bahia state to test two asphalt pavement solutions, with soil-gravel and cement-improved soil base layers, and to verify the accuracy of MeDiNa in estimating the short-term evolution of crack area (CA) and rutting. After implementing the experimental sections, monthly data collection was conducted after traffic was allowed on the road to determine the evolution of functional and structural parameters of these two constructed pavement sections. Using the geotechnical material characterization data and the asphalt mixture used in this experimental segment, the MeDiNa software was utilized to estimate the CA and rutting evolutions month by month for a 10-year project horizon. After 12 months of monitoring, the results indicate that the estimated values of rutting in MeDiNa diverge from those found in the field, where the onset of this pathology has already been detected. However, the pavement fatigue phase has not yet begun. Therefore, the AT values estimated in the software in the short-term analysis are inconsistent with those observed in the field. In addition to the lack of calibration, other factors, such as traffic and different climates, impact the low predictability.