Journal Published Online: 19 October 2022
Volume 51, Issue 2

Structural Behavior Prediction Model for Asphalt Pavements: A Deep Neural Network Approach

CODEN: JTEVAB

Abstract

Structural behavior of pavements is assessed using various destructive and nondestructive tests, albeit they are found to be cost-intensive. There is a need to develop cost-effective structural condition evaluation methods that are scientifically sound so appropriate maintenance interventions can be performed at the right time. The objective of this research study was to develop a Deep Neural Network (DNN)–based approach to predict pavement structural condition using functional, traffic, and climatic characteristics. A DNN was developed to calculate the deflection bowl parameters along with peak surface deflections from roughness, traffic, pavement age, pavement temperature, and climatic conditions. Over 26,000 data points covering various geographic locations were used to establish a global model (R2 = 82 % for the test data) to evaluate the structural integrity of asphalt pavement layers. It is envisioned that this study would assist roadway agencies in assessing the overall condition of asphalt pavements synergizing functional and structural characteristics.

Author Information

Haridas, Aswani K.
Department of Civil and Environmental Engineering, Indian Institute of Technology Tirupati, Andhra Pradesh, India
Peraka, Naga Siva Pavani
Department of Civil Engineering, GMR Institute of Technology Rajam, Andhra Pradesh, India
Biligiri, Krishna Prapoorna
Department of Civil & Environmental Engineering, Indian Institute of Technology Tirupati, Andhra Pradesh, India
Pages: 31
Price: $25.00
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Stock #: JTE20210804
ISSN: 0090-3973
DOI: 10.1520/JTE20210804