Award Date

12-1-2022

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Civil and Environmental Engineering and Construction

First Committee Member

Jee Woong Park

Second Committee Member

Pramen Shrestha

Third Committee Member

Jin Ouk Choi

Fourth Committee Member

Ying Tian

Fifth Committee Member

Ashok Singh

Abstract

Construction researchers have made a significant effort to improve the safety of scaffolding structures, as a large proportion of workers are involved in construction activities requiring scaffolds. However, most past studies focused on design and planning aspects of scaffolds. While limited studies investigated scaffolding safety during construction, they are limited to simple cases only with limited failure modes and simple scaffolds. In response to this limitation, this study aims to develop an automated scaffold monitoring approach capable of monitoring large scaffolds. Accordingly, this study developed an automated scaffold safety monitoring framework that leverages sensor data collected from a scaffold, scaffold modeling techniques, and a machine-learning approach. The proposed framework is based on the capability of the machine-learning approach to identify patterns, which in this study are the patterns of the scaffold structural response based on different loads acting on it. Due to the cost and safety issues related to testing an actual scaffold with varying load applications, the scaffold monitoring framework was experimentally tested under a controlled laboratory setting with a single-bay two-story scaffold with four safety cases. After the field trial, this approach was applied on a four-bay and three-story scaffold involving 1,411 safety cases through computational exploration. During this process, this study integrated a divide-and-conquer strategy with machine-learning models to improve the performance of large-scale classification. The results show that the proposed scaffold monitoring approach is capable of large-scale classification of scaffold safety status. Therefore, this approach can be reliably applied to monitor similar scaffolds on construction sites. Further, this approach is replicable to solve other classification problems. In addition, this study is expected to encourage the use of sensing technologies and data analysis techniques to develop automated monitoring approaches.

Keywords

construction safety; divide-and-conquer; large-scale classification; machine learning; real-time monitoring; temporary structures

Disciplines

Civil Engineering | Occupational Health and Industrial Hygiene

File Format

pdf

File Size

3000 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Rights

IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/

Download


Share

COinS