Análise comparativa entre os métodos de dimensionamento DNER e Medina em pavimentos projetados no estado do Paraná

Abstract

The highway network is a fundamental part of Brazil's infrastructure, playing an important economic and logistical role. However, it is reported that the conditions of the roads are not ideal for are not ideal for their functionality, since long before the end of the project's useful life they present many pathologies that harm the safety and quality of the pavement. One factor that may be related to this is the way the structures are dimensioned. The work provides a comparative analysis between the DNER methods (empirical) and the Medina design method (mechanistic-empirical), using two structures designed in the state of Paraná. Data from project memorials found in digital media were used, such as traffic, soil, and proposed sizing data. With the help of the Medina software, the projection of the cracked area and the sinking of the wheel track into the pavement during its useful life was carried out, both with the structures dimensioned using the DNER method and those dimensioned using the Medina method. It was found that there was a difference in pavement thickness using the different methods, and in projects that anticipated a greater volume of traffic, the use of only the DNER method was considered insufficient. The use of empirical design is not the only determining factor in inadequate pavements, however, it is clear that design based on experiments, simulations, and with a mechanistic bias presents greater reliability. The main stresses acting on pavements designed using different methods are also verified and how small changes in layer thickness affect the pavement's resistance. It was verified, as well as other studies carried out, that the Medina method presents an important solution for the sizing of pavements, and that by carrying out experimental segments over time it will be possible to improve the method by seeking a reliable approximation of the estimate of software with reality.