Abstract
Subgrade moisture content significantly influences soil strength and pavement bearing capacity. Pavement moisture content varies greatly throughout the year, especially in cold regions. Thus, having a better understanding of seasonal variation in moisture content in the pavement is needed to be developed. This research aims to apply machine learning models to predict the moisture content of unbound materials in the pavement. Unfrozen volumetric moisture content measurements recorded at the Integrated Road Research Facility test road in Edmonton, Alberta were used to train machine learning models to predict moisture content at depths within 2.7 m of the road surface. Machine learning models were implemented based on three parameters of pavement temperature, day of the year and depth. The results from the machine learning model were compared with a statistical model and showed higher accuracy than the existing model, indicating that machine learning models could enhance moisture content prediction in the pavement.
Acknowledgements
Thanks to Lana Gutwin for her assistance in editing and preparing this paper.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available.