Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Artificial neural network prediction model for in situ resilient modulus of subgrade soils for pavement design applications

Artificial neural network prediction model for in situ resilient modulus of subgrade soils for... Subgrade resilient modulus (Mr) plays an important role in designing a pavement structure and complex traditional regression analysis-based model are still in use to predict Mr. Therefore, there is a dire need for the development of a simple, standalone model for predicting the resilient modulus of subgrade soils while bypassing the need to utilize many complex experimental factors. This study utilizes an artificial neural network (ANN) framework for developing a model to predict Mr. The data required for the analysis is obtained from 30 Long-Term Pavement Performance-Seasonal Monitoring Program (LTPP-SMP) pavement sections. A multilayer feed-forward ANN with only six neurons was utilized for the model training, and it is found that the developed model has an excellent prediction capability with an R-squared value of 0.84, which vastly outperformed models found in the literature. The developed model can be a perfect fit for various departments of transportation in the quick prediction of Mr of subgrade soils without the need of performing sophisticated tests. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Innovative Infrastructure Solutions Springer Journals

Artificial neural network prediction model for in situ resilient modulus of subgrade soils for pavement design applications

Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/artificial-neural-network-prediction-model-for-in-situ-resilient-YZYzNFO4xP

References (27)

Publisher
Springer Journals
Copyright
Copyright © Springer Nature Switzerland AG 2021
ISSN
2364-4176
eISSN
2364-4184
DOI
10.1007/s41062-021-00659-x
Publisher site
See Article on Publisher Site

Abstract

Subgrade resilient modulus (Mr) plays an important role in designing a pavement structure and complex traditional regression analysis-based model are still in use to predict Mr. Therefore, there is a dire need for the development of a simple, standalone model for predicting the resilient modulus of subgrade soils while bypassing the need to utilize many complex experimental factors. This study utilizes an artificial neural network (ANN) framework for developing a model to predict Mr. The data required for the analysis is obtained from 30 Long-Term Pavement Performance-Seasonal Monitoring Program (LTPP-SMP) pavement sections. A multilayer feed-forward ANN with only six neurons was utilized for the model training, and it is found that the developed model has an excellent prediction capability with an R-squared value of 0.84, which vastly outperformed models found in the literature. The developed model can be a perfect fit for various departments of transportation in the quick prediction of Mr of subgrade soils without the need of performing sophisticated tests.

Journal

Innovative Infrastructure Solutions – Springer Journals

Published: Feb 1, 2022

Keywords: Prediction models; Artificial neural network; LTPP; Subgrade; Moisture content; Resilient modulus

There are no references for this article.