Empirical-Mechanistic Method Based Stochastic Modeling of Fatigue Damage to Predict Flexible Pavement Cracking for Transportation Infrastructure Management

Abstract

In the purely theoretical approach of pavement design, percentage fatigue cracking is related to damage in a probabilistic manner according to the Miner’s law. Two methods that are currently widely in use are based on assumptions of damage distribution. One method assumes fatigue damage being normally distributed, while the other one assumes fatigue damage being lognormally distributed. Since mechanistic-empirical pavement design and pavement management require precise forecasting of pavement fatigue cracking, much effort should be taken to characterize and predict fatigue cracking in terms of damage distribution. In this paper, we formulate the probability density distribution of fatigue damage of flexible pavements according to the underlying structure of fatigue cracking equations so that pavement fatigue-cracking damage can be interpreted in a more meaningful way. Numerical computation is conducted for a case study. It is found that damage is neither normally nor lognormally distributed. It is therefore recommended that methodology and damage distribution model established in this paper be used in practice to predict damage distribution and percentage cracking so that a better estimation of fatigue cracking can be made.