DEVELOPING STOCHASTIC FLEXIBLE PAVEMENT DISTRESS AND SERVICEABILITY EQUATIONS

The development of a method of predicting pavement performance in terms of present serviceability index and four primary distress types (area and severity) and the application of the method to the design of flexible pavements are summarized. The method is based on an S-shaped performance curve, the curve fit parameters for which can be determined using the methodology developed by Garcia-Diaz and Riggins. These parameters have been found for 164 pavement test sections located throughout Texas. The pavement test sections were categorized as three main types: asphalt concrete pavement on unbound base course, asphaltic concrete pavement on bituminous base course, and asphaltic concrete overlay. Data describing the pavement structure, including the thickness and elastic modulus of each layer, the environment, and the traffic for these pavements, were used to develop regression models for the curve fit parameters. A sensitivity analysis was made of these models to determine the effects of climate on pavement performance in four widely separated highway districts in Texas. The regression models along with the proposed performance equations and the stochastic form of these equations have been incorporated in the Texas Flexible Pavement System (FPS) design computer program. The modified version of FPS provides a listing of the optimal pavement designs selected on the basis of least total cost including material and user costs, overlay costs, and salvage values.