Modelos de previsão de desempenho de pavimentos flexíveis para uma rodovia no Rio de Janeiro

Abstract

Brazilian Road Researches showed, in 2012, that public investment in highways in the country was insufficient to solve problems of road infrastructure. The low level of investment in maintenance affects the quality of the roads, this factor causes problems such as increasing the operating costs of transportation and the increased number of accidents. Use of Pavement Management System (PMS) by road agencies and concessionaires arise as an alternative to reverse the current situation of the sector and to provide an improvement in performance of pavements. Models performance prediction of the pavement are one of the main tools of a PMS. However, little attention has been given to the development, calibration and validation of those models. In general, it is adopted generic models and models obtained in completely different conditions existing in administered highways. The knowledge of the model which best represents the actual situation of a pavement performance, will allow the pavement management system make better planning of maintenance on roads and to implementate more efficiently the resources. This research is the development of equations to predict the deflection and irregularity of flexible pavement in a federal highway. The equations developed showed best approaches, comparing with actual data and the values predicted by models, reinforcing the importance of developing specific models for highways, mainly for institutions working on project level managing. An important factor noted was the need to mount a robust database and reliable informations to minimize errors incorporated in the regression, increasing the number of independent variables to the development of more reliable models and to execute the analysis based on trustfull interval.