Use of mechanistic-empirical method of pavement design for performance sensitivity analysis to asphalt pavement fatigue

Abstract

In the new Brazilian mechanistic-empirical design method of asphalt pavements, MeDiNa, the characterization of permanent deformation (PD) for the selection of soils and gravel is based on tests performed with at least 150,000 loading cycles for each of the nine specimens indicated in the DNIT standard. Despite providing information about the material behavior under a wide range of testing conditions, the experimental program related to these PD characterizations is time consuming and it is believed that it can be optimized. This paper evaluates the influence of the number of loading cycle applications on the characterization of the materials. For this purpose, seven materials were analyzed at their optimum moisture content (OMC) and one of them was also compacted in a condition above the OMC, in a total of eight data sets. Statistical regression analyzes were performed to identify the parameters of the predictive model for different numbers of cycles and the PD predictions for the different materials were compared. From these results, simulations were performed in the MeDiNa software to predict the performance of the materials. Four different N values were evaluated, considering 150,000 cycles as reference: discarding the 500 first cycles, but considering the PD accumulated in that interval; discarding the 500 first cycles and considering the PD accumulated in that interval; final N of 80,000; and final N of 100,000. For the analyzed materials, no significant differences were observed in the PD prediction, even considering tests with 50,000 or 70,000 cycles less than the 150,000 cycles required in the standard. This indicates that, although characterization is recommended following standardized procedures, the experimental program of the current PD standard can possibly be significantly optimized by reducing the number of cycles applied to materials in laboratory tests. This possibility must be analyzed for each material.