Please use this identifier to cite or link to this item:
Title: Relationships between social distance, environmental factors and workers' intrusion behaviors on construction site
Authors: Dong, Shuang
Degree: Ph.D.
Issue Date: 2017
Abstract: Construction is one of the most dangerous industries in the world. Apart from physical suffering, the safety issue is exacerbated by the delays and economic loss caused by safety-related incidents in construction. Given the significance of maintaining good site safety, an increasing number of researchers and practitioners have investigated the causes of the aforementioned situation and attempted to determine effective methods and tools to manage safety in construction sites. Previous studies attributed the occurrence of accidents in construction sites to two main reasons, namely, unsafe behaviors (i.e., behavior or activity of a person that deviates from generally accepted safety procedures) and unsafe conditions (i.e., hazard or unsafe mechanical or physical environment). A thorough understanding and assessment of these rule-breaking behaviors are necessary to prevent accidents because these behaviors are believed to contribute to over 80% of all workplace accidents. Intrusion, which means to enter dangerous areas in a construction site without any permission, is one of the most serious rule-breaking behaviors and could lead to many types of major accidents and injuries, such as falling from a height, falling on the floor, electrocution, burn, and being hit by moving objects, among others. However, present studies on the assessment of factors that influence rule-breaking behaviors in construction sites are limited and inefficient in terms of three areas: (1) outcome-based lagging indicators, (2) nonpersonal-level evaluation, and (3) manual and subjective observation. The current study aims to solve these problems in two aspects: (1) determining and proposing a novel approach to automatically identify and assess intrusion behaviors in a construction site and (2) exploring the manner by which social distance factors and the working environment are related to these types of rule-breaking behavior.
In this study, a real-time location-tracking system is selected and applied to effectively assess and record intrusion behaviors. Thereafter, a five-month on-site experiment is conducted to analyze the relationships among social distance factors (e.g., gender, age, hometown, and trade), working environment (e.g., weather, temperature, and temperature difference), and intrusive behaviors. The data analysis is conducted and finds out workers and days with special characteristics that may cause extensive intrusion behaviors. In particular, these characteristics are male, middle age, small working areas with danger zones around, low related experience, minimal safety training, good weather, and uncomfortable temperature condition Through on-site interviews, the corresponding reasons are found to be: (1) lack of safety perception, (2) shortcuts due to unsafe site layout, (3) lack of safe places to rest, (4) tight work schedule, and (5) poor safety attitude, particularly the "tough guy" mindset. Then recommendations to control this type of risk-taking behavior are proposed like the application of location-tracking technology, timely and targeted safety training, safe layout and reasonable workload. This research mainly provides a new method of managing intrusion behaviors, including automatic verification of intrusion, identification of influencing factors, and control strategies. This study fills in the gap in the research on intrusion behaviors in construction site and provides a series of practical intrusion management methods for the construction industry.
Subjects: Hong Kong Polytechnic University -- Dissertations
Construction industry -- Safety measures
Building -- Safety measures
Construction workers -- Health and hygiene
Pages: xxxi, 341 pages : color illustrations
Appears in Collections:Thesis

Show full item record

Page views

Last Week
Last month
Citations as of Nov 26, 2023

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.