الفهرس 
AbstractOnly 14 pages are availabe for public view
 |  | from | 155 |  |  |
| | | from | 155 | | |
Abstract
Non-structural factors such as surface distresses and ride quality have been commonly used as indicators of pavement conditions. Recently, the concept of implementing a structural condition index in Pavement Management System (PMS) to complement functional condition indices has become an important goal for many highway agencies. The Rolling Wheel Deflecto meter (RWD) provides the ability to measure pavement deflection while operating at traffic speed causing no delays. The objective of this study was two-fold. First, this study developed a model to predict pavement structural capacity based on RWD measurements and assessed its effectiveness in identifying structurally deficient pavement sections. Second, this study introduced a framework, along with the required implementation tools, for incorporating pavement structural conditions into the Louisiana PMS decision matrix. The proposed framework aims at filling the gap between network level and project level decisions and eventually, allowing more accurate budget estimation. RWD data collected from 153 road sections in District 05 of Louisiana were utilized in this study. The predicted Structural Number (SNRWD0.1) showed an acceptable accuracy with a Root Mean Square Error (RMSE) of 0.8 and coefficient of determination (R2) of 0.80 in the validation stage. Core samples showed that sections that were predicted to be structurally deficient suffered from asphalt stripping and material deterioration distresses. Results support that the developed model is available tool that could be used in PMS at the network level to predict pavement structural condition with an acceptable level of accuracy. With respect to the implementation of RWD in Louisiana PMS, two enhanced decision trees were developed, such that both functional and structural pavement conditions are considered in the decision making process. Implementation of RWD in the decision-making process is demonstrated and is expected to improve the overall performance of the pavement network. Furthermore, the enhanced decision trees are expected to reduce the total maintenance and rehabilitation (M&R) construction costs if applied to relatively high volume roads. Based on the results of this study, a one-step enhanced decision-making tool, which considers both structural and functional pavement conditions in treatment selection, was developed using an Artificial Neural Network (ANN)-based pattern recognition system.