Cost-Effective Detection and Repair of Moisture Damage in Pavements
-
2021-12-01
-
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report, 05/2018 – 05/2021
-
Corporate Publisher:
-
Abstract:The major objective of this study was to evaluate the use of non-destructive evaluation (NDE) test methods to identify moisture damage in the field, to classify surface cracking in flexible pavements as top-down or bottom-up cracking, and to predict roughness conditions from surface images. NDE test methods evaluated in this study included Traffic Speed Deflection Device (TSDD) and Ground Penetrating Radar (GPR). In addition, the performance and cost-effectiveness of treatment methods including asphalt concrete (AC) overlay and chip seal were analyzed with and without moisture-induced damage. The study also developed a decision-making tool to identify top-down, bottom-up, and cement treated (CT) reflective cracking in in-service flexible pavements using Convolutional Neural Networks (CNN) model and Artificial Neural Networks (ANN). The developed CNN model was found to achieve an accuracy of 93.8% and 91.0% in the testing and validation phases, respectively. The ANN based decision-making tool achieved an overall accuracy of 92% indicating its effectiveness in cross-checking the prediction from the CNN model. Measurements from the Rolling Wheel Deflectometer (RWD) were evaluated to identify pavement sections that may suffer from stripping damage. Statistical and ANN models that used RWD measured deflections and pavement characteristics were developed to predict the probability of stripping damage in the tested sections. A regression-based classification tree was also developed that is easy to interpret and is convenient for highway agencies for preliminary stripping evaluation. A moisture detection protocol was also developed based on GPR measurements as a noninvasive and continuous evaluation technique to detect moisture damages in flexible pavements. A novel GPR-based indicator, known as the Accumulating In-layer Peaks (AIP), was introduced to detect stripping damage in asphalt pavements. The AIP predicted accuracies for stripped and non-stripped sections were 82% and 96%, respectively, indicating its effectiveness to detect AC stripping damage in flexible pavements. The study also evaluated the effects of AC stripping damage on the performance and cost-effectiveness of chip seal and AC overlays in pavement maintenance and rehabilitation. Results showed that for chip seal, moisture damage negatively affected the performance of the sections especially for low traffic volumes. For AC overlays, moisture-induced damage significantly affected the long-term pavement performance at all traffic levels. On average, moisture-induced damage decreased the extension in pavement service life and cost-effectiveness by 4.6 years and 0.5%, respectively. The study also employed CNN to classify pavement sections into different International Roughness Index (IRI) categories and to predict IRI values using Three-Dimensional (3D) pavement images. The developed CNN model classified pavement sections according to their roughness conditions with an accuracy of 93.4% and 89.6% in the training and validation stages, respectively. In addition, a CNN model was developed for the prediction of IRI values from 3D pavement images. The model yielded accurate predictions with a coefficient of determination (R2) of 0.985 and an average error of 5.9%.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: