Quantifying Workers’ Hazard Identification Using Fuzzy Signal Detection Theory

Abstract

Workers' safety and health is a primary concern in the construction industry due to the significant number of occupational injuries and fatalities experienced. Previous research indicates that such injuries and fatalities are multi-causal, and one of which may be workers drifting towards hazard. The purpose of this study was to develop a method to quantify the ability of construction workers to identify hazards such that this drifting can be minimized. The paper is not reporting on a specific finding regarding hazard identification ability amongst the population of workers surveyed in the research. Rather, it is a foray into a new method that augments assessment of hazard identification abilities of workers. This research applied a hybrid model, Fuzzy Signal Detection Theory (FSDT) to quantify workers' ability to identify occupational hazards. Data was collected using a survey designed to assess construction workers' perceptions of safety specific to the risk of fall accidents for structural steel workers. An eighteen-question survey was based on standards set by national safety agencies in the United States and on a mix of conditions that were with no violation, a complete violation, and a partial violation. Both the conventional Signal Detection Theory (SDT) and FSDT were used for analysis and the results were compared. The comparison of results indicated that the FSDT model provided a richer framework to study worker hazard perception on a construction site. The methodology is expected to provide guiding framework for similar studies. Increasing data collection of the type presented in this paper will enable refinement and revision of construction occupational safety and health regulations by national agencies.