Work safety behaviour model for improving safety in construction industry using Bayesian Network

• Main Author: Zid, Chaher
• Format: Thesis
• Language: English; English; English
• Published: 2020
• Subjects: TH1000-1725 Systems of building construction. Including fireproof construction, concrete construction
• Online Access: http://eprints.uthm.edu.my/4836/1/24p%20CHAHER%20ZID.pdf; http://eprints.uthm.edu.my/4836/2/CHAHER%20ZID%20COPYRIGHT%20DECLARATION.pdf; http://eprints.uthm.edu.my/4836/3/CHAHER%20ZID%20WATERMARK.pdf; http://eprints.uthm.edu.my/4836/

The construction industry involves one of the most hazardous occupations for workers due to complex management processes, environmental issues, work pressure and heavy equipment involved in modern construction projects. Despite the advancement of technology and the new approaches which have been developed and adopted to tackles the accidents in construction industry in Malaysia, an escalating number of fatal accidents are still occurring because of mainly the human errors and the unsafe behaviours. In the same vein, improving the safety behaviour became the main attention of the researchers recently in construction industry to reduce the accidents at workplace. Therefore, this study determined the key factors and the relationship between all safety behaviour factors (organizational factors, safety climate factors and individual factors) in order to build a safety behaviour model for improving safety and minizing the fatal accidents in the construction industry. In this research, all critical safety behaviour factors were defined from literature review in order to develop a comprehensive conceptual framework. After performing the pilot study, defining the proportionate stratified sampling, the data was collected accordingly from different construction projects. The key factors, then, were finalised by using explanatory factor analysis using SPSS. With the assistant of the experts in the safety field and by applying Dempster and Shafer theory, the structure of Bayesian network was built accordingly. Furthermore, Netica software embraced the latest shape of developed safety behaviour model. The developed Bayesian network model elaborated generally that the individual factors have registred the highest proportion of influence on safety behaviour while there was a mediocre influence by the safety climate factors. On the other hand, the safety behaviour received a low influence by the Organizational factors. Besides, Join Strategy assisted to deduce the best safety behaviour strategy to upgrade the safety at the workplace in the construction industry.The results revealed also that learning, safety compliance and safety knowledge have an enormous influence on safety behaviour and they should be well-taken care by the safety management to ameliorate the safety at workplace. All in all, safety behaviour can record a high performance among the workers by promoting these factors. Hence, the leaders and decision-makers are advised to enhance the organizations’ learning, safety knowledge and safety compliance to upgrade the safety behaviour of the employees in the construction site, and therefore this will improve the safety performance and minimise the fatal accidents in the construction industry.