Long-Term Modelling of Composite Pavement Performance

The rehabilitation of rigid (concrete) pavements with the placement of an asphalt overlay is a common maintenance technique, which results in a structure known as a composite pavement. The most common form of distress in this type of pavement is reflection cracking which can be due to traffic and/or climatic loading. This thesis is focused on the prediction of reflective cracking as a composite pavement response due to traffic and thermal loading. The model is the first of its kind to include factors such as dynamic vehicle loading in modelling the long-term performance of flexible composite pavement structures. The mechanisms of traffic driven reflective cracking are investigated, as well as a thermal driven cracking model. The above models are combined with a dynamic vehicle model into a whole-life model and its framework is presented. A parametric analysis investigates the sensitivity of the predictions to the variation of parameters: asphalt elastic modulus and thickness, concrete elastic modulus and thickness, subbase elastic modulus and thickness and subgrade elastic modulus. Finally, the main validation analysis is presented, with application of the whole model for representative pavement sections from 6 states in the USA. The input data is found in the Long-Term Pavement Performance Infopave database, which contains performance and traffic data from monitored in-service road sections in North America. A comparison is made between the outcome predictions and actual field measurements of transverse cracks on the relevant pavement sections. The selected pavement sections belong to two climatic regions: wet/freeze zone, wet/non-freeze. Predictions for certain states show matching results approximately to the actual development of reflective cracking at different periods after the placement of the asphalt overlay. It was found that the asphalt material properties have a big influence on the final outcome making critical its estimation. Temperature variations have also been taken in account and it has been seen that they can influence the result showing that thermal reflective cracking can be dominant. It was also discovered that knowledge of the state (crack spacing, severity of cracks) is highly important to the identification and prediction of reflective cracks. It was important to also investigate the process of both directions of cracking, bottom to up and top to down, identifying their dominance in each case. In the final chapter conclusions and recommendations for future work are presented