The Use of Artificial Intelligence in Pavement Engineering

Abstract

The performance of asphalt pavements decreases with time because of traffic loading and environmental conditions. Performance decay models are needed in pavement management systems to program pavement preservation and rehabilitation treatments to extend the service life and improve the performance of flexible pavements. Many factors affect pavement performance, including the material properties and thickness of each layer, applied traffic, and environmental conditions. Performance models, including those for rutting, cracking, roughness, are often developed and used to forecast the future conditions of pavements. Meanwhile, to develop reliable performance models, numerous variables are needed in such models, and historical performance data are required. This study investigated and developed multiple types of artificial intelligence models to predict pavement performance. The study results demonstrated that random forests regression was best suited for the data utilized in this study. Multiple random forests regression models were developed to predict various indicators of pavement performance, such as the International Roughness Index (IRI), rutting, and cracking. These models utilized a theoretical dataset generated with the Pavement ME software and field data collected from the Long-Term Pavement Performance (LTPP) Program. There were good correlations between all the theoretical and predicted performance indicators. In addition, the predicted performance decay curves were found to closely simulate the measured decay curves. In addition, the results for the models developed with the field dataset demonstrated good correlations between measured and predicted performance indicators for some of the investigated performance indicators.