Building safe organisations: using machine learning to decode safety habits of blue-collar workers in the construction industry

This study provides a novel comprehensive framework for classifying the safety behaviours of blue-collar workers in the construction sector, while emphasizing employee safety as a fundamental concern for operational efficiency, productivity, and employee well-being in a competitive business context. The study applies some of the latest machine learning techniques, both for framework development and framework validation through categorization of construction workers' safety behaviour, using machine learning techniques such as logistic regression, support vector machine, and decision tree. For validation, the author used randomly sampled construction workers’ safety behaviour (most frequent safety behaviours). The results show that workplace safety behaviour (WSB) is predominantly influenced by supervisor support, the presence of reckless

behaviour, and the self-motivating safety behaviour of the workers. The study also provides a list of important sub-factors that contribute to establishing a safe working environment, such as strategies to prevent workplace accidents, enforcement of workplace safety rules, proper documentation of safety-related records, and training/instruction of staff. The study also recognises the positive impact of construction work technology on various construction fields, and the framework describes some factors that can further enhance workplace safety, such as comprehensive safety work education, anticipation of workplace accidents in various situations, and a balanced enforcement of rules. This framework also targets the construction industry and provides avenues for advancement to improve workplace safety management.

Message: Hi FE, I need help with “Building safe organisations: using machine learning to decode safety habits of blue-collar workers in the construction industry” in “Civil Engineering”