- Title
- Assessing the influence of safety climate on off-site manufacturing safety performance: a Bayesian network approach
- Creator
- Ovitigala Vithanage Don, Sadith
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2022
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Off-site manufacturing (OSM) has been promoted as a contemporary production technique applied in the construction industry. It involves 60–90% of construction tasks being carried out in a manufacturing facility off-site prior to moving the product to the construction site. Several benefits are expected from OSM because it occurs in a controlled environment. Benefits include improved quality, time savings, increased resource efficiencies, reduction in construction waste and decreased on-site activity. Among these benefits, many industry practitioners and researchers believe that OSM offers a safer environment for workers because most of the activities are performed in a controlled environment, that is, a factory. However, safety management in OSM has become of particular concern with poor safety performance measures. Somewhat surprisingly, injury data from the USA indicate higher injury rates in OSM than in traditional construction. Accordingly, research is needed to empirically examine the factors that influence OSM safety performance and test these assertions. The aim of the present research was to examine the influence of safety climate, workers perception about safety state in an organisation, on safety performance in the OSM sector. Initially, key safety performance categories (i.e., human, organisational, and work environment) pertinent to OSM and safety climate dimensions were identified through an extensive literature review along with their underlying causal factors. Following this, a quantitative methodological approach was chosen, and a questionnaire survey designed to gather empirical data from OSM practitioners in Australia. Consequently, a Bayesian network approach was adopted to explore the probabilistic relationships between the identified factors associated with safety climate. A final OSM safety climate model was then validated using sensitivity analysis (e.g., tornado diagrams, derivatives of sensitivity), injury data gathered as part of the questionnaire survey and strength of influence analysis. The main findings included that OSM safety performance was influenced by safety climate factors related to human, organisational and work environment facets. It was discovered that OSM safety performance had contrasting effect from safety climate dimensions and their underlying causal factors. Three separate models were developed to present probabilities associated with different states of safety climate factors, identify interrelationships among factors and underlying causal factors and to explain the influence of these factors regarding OSM safety performance. In the first model, it was discovered that human safety influence was affected by different factors of worker involvement and co-worker support of safety. The results indicate that human safety influence can be improved by enhancing a co-worker’s safety value and practice, management safety response, as well as improving worker involvement to safety by providing appropriate safety training and reducing production pressure over safety. The second model concerning organisational safety influence suggests that there is a greater effect from safety climate dimensions such as management commitment, supervisor commitment and safety communication while comparatively lower effect was identified from safety training and safety rules and procedures. Several strategies were identified to improve organisational safety influence include, but not limited to, improvements in areas of management safety response, balance between safety and production, credibility of training, frequency of safety meetings, programs and campaigns, supervisors’ safety actions and expectations and feedback and feed-forward systems. The third model, work environment safety influence was assessed through the two lenses of technological and physical environment and identified that the former has a greater effect on OSM safety performance. The safety improvement strategies for work environment include enhancing machinery and equipment safety condition, factory housekeeping and supportiveness of factory layout for safety. Consideration on factory designing and planning at early stage is important particularly for OSM because the OSM workers are exposed to machinery, tools and equipment exclusive for OSM practice. Finally, these three models were integrated to develop an integrated OSM safety climate model to predict safety performance in an OSM organisation. The Bayesian network approach developed and used to underpin this research offers an effective tool to predict OSM safety performance by considering antecedents of OSM safety performance including human, organisational and work environment aspects. Researchers and practitioners in this domain will benefit from using the tool developed from this research to examine the interrelationships between factors associated with safety climate. This study also extends safety climate knowledge by investigating its influence in OSM and utilising a distinctive Bayesian network approach compared with conventional statistical analysis methods. For industry practitioners, it offers a platform to identify appropriate safety management strategies in an optimal safety expenditure such as savings from safety investments and additional training time allocation. The model also provides an understanding of the causality between several factors and underlying causes related to OSM safety. Therefore, industry stakeholders and practitioners in an OSM factory such as safety managers or factory managers can make informed decisions regarding safety management.
- Subject
- Bayesian network; off-site manufacturing; prefabricated construction; probabilistic assessment; safety climate; safety performance
- Identifier
- http://hdl.handle.net/1959.13/1508576
- Identifier
- uon:56137
- Rights
- Copyright 2022 Sadith Ovitigala Vithanage Don
- Language
- eng
- Full Text
- Hits: 141
- Visitors: 166
- Downloads: 35
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 7 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 124 KB | Adobe Acrobat PDF | View Details Download |