Estudo comparativo de diferentes métodos de dimensionamento de pavimentos flexíveis, utilizando-se um sistema computacional.

Abstract

This thesis shows the results of a comparative analysis between three flexible pavement design methods; the flexible pavement design method used by DNER (DNER method) the method based on the elasticity theory (elastic method) and another method developed for tropical countries (called "Tropical Inglês" method). All those methods use a computacional program called "DIMP". The obtained results showed that the DIMP program may be used for flexible pavement design under any loading system and any layer materials configuration, allowing the investigation of a larger number of possible solutions to be simulated. From the dimensioning results it may be conclude that, for the conditions of our investigation, the total thicknees of the pavement is affected neither by the elastic modulus of the base layer nor by its CBR value. This thickness depends only on the subgrade layer CBF, values and on the traffic level when both DNER and tropical ingles methods are used and that this total thickness depends on the elastic moduli of the subgrade and subbase layers and on the traffic level when the elastic method is used for design. The DNER method gave a higher thickness for all our tested conditions CBR and elastic moduli values of the layers when the traffic level was maintained between 10 5 and 5x10 6 vehicles/year. For a traffic level varying between 10 3 to 10 4 vehicles/year or between 10 7 to 10 8 vehicles/year, the DNER method gave a smaller total thickness of the conditions than the one given by the elastic method. The results showed that, under all the conditions of our investigations, the Tropical Ingles led to the smallest total thickness of the pavement.