Journal of Traffic and Transportation Engineering (English Edition)

Journal of Traffic and Transportation Engineering (English Edition)

Review Commodity

Research and applications of bogus neural network in pavement engineering: A country-of-the-fine art review

Under a Artistic Commons license

Open access

Highlights

• Frontiers of artificial neural network (ANN) in pavement design, construction, inspection and maintenance were reviewed.

• Three mainstream ANN architectures for dissimilar awarding scenarios were summarized.

• Five enquiry challenges and prospects of ANN application in pavement engineering were analyzed.

• Standardized literature search and nomenclature methods were implemented.

Abstract

Given the not bad advancements in soft calculating and information science, artificial neural network (ANN) has been explored and applied to handle complicated problems in the field of pavement technology. This study conducted a land-of-the-art review for surveying the recent progress of ANN awarding at dissimilar stages of pavement engineering science, including pavement design, structure, inspection and monitoring, and maintenance. This study focused on the papers published over the last three decades, especially the studies conducted since 2013. Through literature retrieval, a full of 683 papers in this field were identified, among which 143 papers were selected for an in-depth review. The ANN architectures used in these studies mainly included multi-layer perceptron neural network (MLPNN), convolutional neural network (CNN) and recurrent neural network (RNN) for processing one-dimensional information, two-dimensional data and fourth dimension-series data. CNN-based pavement wellness inspection and monitoring attracted the largest enquiry interest due to its potential to replace man labor. While ANN has been proved to exist an effective tool for pavement material blueprint, cost analysis, defect detection and maintenance planning, it is facing huge challenges in terms of data collection, parameter optimization, model transferability and low-toll data annotation. More attending should be paid to bring multidisciplinary techniques into pavement engineering to tackle existing challenges and widen future opportunities.

Keywords

Pavement engineering

Pavement design

Artificial neural network

Deep learning

Pavement life cycle

Health inspection and monitoring

Cited by (0)

Xu Yang received the bachelor's degree from Southeast University, China, in 2009, and the PhD degree from Michigan Technological University, The states, in 2015. He is currently a professor at the School of Highway, Chang'an University, China. He has published over 120 papers with more 2000 citations (Google Scholar) in peer reviewed journals and briefing proceedings. His research interests include advanced road pavement construction and maintenance technology, deep learning for automated pavement distress detection, and numerical simulation for civil engineering science materials.

Jinchao Guan received the bachelor's degree from Nanjing Tech University, China, in 2018. He is currently a PhD pupil at the School of Highway, Chang'an University. His electric current enquiry interests include 3D reconstruction for transportation infrastructure, deep learning for pavement health monitoring, and mathematical optimization for pavement maintenance.

Ling Ding received the PhD degree from Southeast Academy, China. She is currently a lecturer at the College of Transportation Engineering, Chang'an University. Her research interests include remote command for route activities, travel beliefs analysis using neural network and transportation infrastructure health monitoring.

Zhanping You received the PhD degree in civil engineering from the University of Illinois at Urbana-Champaign, U.s.a., in 2003. He is currently a distinguished professor at the Section of Civil and Environmental Engineering, Michigan Technological University. He has published over 300 papers in peer reviewed journals and briefing proceedings. His inquiry interests include transportation materials, pavement design, asphalt materials, and transportation engineering.

Vincent C.South. Lee received the PhD degree from University of New Castle, NSW, Australia. He is currently an associate professor at Auto Learning and Deep Learning Discipline of the Section of Data Scientific discipline and Artificial Intelligence, Kinesthesia of Data Technology, Monash University, Australia. He has published over 200 papers in peer reviewed journals and conference proceedings. His current enquiry interests are multidisciplinary spreading across signal processing; adaptive noesis representation and information engineering; data, text, and graph mining for cognition discovery; determination theory; information organisation research based on design science epitome; and neuro-financial technology, ICU patient wellness support systems, thyroid cancer prognostic and treatment analytics.

Mohd Rosli Mohd Hasan received the PhD caste in civil applied science (transportation materials) from Michigan Technological University, USA. He is currently an academic staff in highway engineering science at the School of Civil Engineering science, Universiti Sains Malaysia (Engineering Campus). He has published over lx papers in several local and international journals, and conference proceedings. His research focuses on sustainable transportation materials and asphalt engineering.

Xiaoyun Cheng received the B.South. and Thou.S. degrees in highway and railway engineering from Chang'an University, Shaanxi, People's republic of china and the PhD caste in transportation engineering from Tongji University, Shanghai, in 2015. She is currently an banana professor of transportation engineering at the College of Transportation Engineering, Chang'an Academy. Her research interests include transportation planning, travel behavior analysis, big data in transportation and statistical modeling.