Developing fuzzy signal detection theory for workers’ hazard perception measures on subway operations

The ability to perceive hazards has been suggested as being an important factor in determining safety risk. By measuring the ability to perceive hazard, at-risk workers could be identified and provided with additional training. This can reduce knowledge-based human errors and thereby help to avoid accidents. This paper proposes the use of fuzzy signal detection theory (FSDT), which combines fuzzy logic and conventional signal detection theory (SDT) to measure hazard perception. The flexibility of the proposed method allows multiple forms of qualitative and quantitative information to be used, including expert opinions and workers’ responses when quantitative data are lacking or when only qualitative aspects are known. Details have been provided on the basic postulates, formulas, and steps involved in FSDT analysis. A real-world example demonstrates the feasibility of the proposed framework. The results of a case study show that the participating workers had a medium level of hazard perception. This means that these workers were sensitive to the difference between hazard and safety. But there were still opportunities for improvement. Of the workers surveyed, five workers expressed conservative views regarding risk while the other workers expressed risky views. FSDT increases the applicability of conventional SDT analysis to many real-world settings, where the definition of a signal is less dichotomous than in a controlled laboratory environment.

Research highlights► This research proposes a fuzzy signal detection theory to measure hazard perception. ► The basic postulates, formulas, and steps of the new method are provided. ► Sensitivity and criteria are two parameters to measure hazard perception. ► With a case study, the participating workers had a medium level of hazard perception. ► Five workers express conservative views regarding risk in the nine workers.