https://doi.org/10.24928/2023/0230

Efficient Pavement Distress Detection and Visual Management in Lean Construction Based on BIM and Deep Learnin

Ting Deng1 & Yi Tan2

1Assistant Professor, Key Laboratory for Resilient Infrastructures of Coastal Cities (Shenzhen University), Ministry of Education, Shenzhen University, Shenzhen, China, [email protected], orcid.org/0000-0001-8902- 4778
2Graduate Student, Sino-Australia Joint Research Center in BIM and Smart Construction, Shenzhen University, Shenzhen, China, [email protected], orcid.org/0000-0003-3064-8353

Abstract

With a wide range of road construction worldwide, the focus of road engineering has shifted to road maintenance and management. This paper presents a research aimed at developing a lean management framework that integrates BIM and deep learning technology to guide lean production applications in road maintenance management. Firstly, the pavement distress dataset is established based on the obtained road point cloud data. Secondly, a deep learning-based 3D object detection network is applied for automatically detect the pavement distress and improve the accuracy and reliability of the detection. After obtaining the detection information of the distress, Dynamo is utilized to realize the efficient visualization management of pavement distresses. Finally, an untrained road section is applied for the experiment. The predicted information of distress is integrated and visualized in BIM model can provide a better maintenance guidance and well promote the transformation of pavement intelligent maintenance management.

Keywords

Lean construction, template, formatting, instructions, references.

Files

Reference

Deng, T. & Tan, Y. 2023. Efficient Pavement Distress Detection and Visual Management in Lean Construction Based on BIM and Deep Learnin, Proceedings of the 31st Annual Conference of the International Group for Lean Construction (IGLC31) , 174-185. doi.org/10.24928/2023/0230

Download: BibTeX | RIS Format