Quantitative risk assessment of industrial hot work using Adaptive Bow Tie and Petri Nets

Abstract

Hot work poses great hazards to industrial operation site. Although regulatory and guidance exist, accidents caused by hot work continue to occur. Risk assessment has proved to be an effective method for risk control. However, current studies paid less attention to the evolution process of industrial hot work accidents and the quantitative risk assessment seems to be insufficient. In this context, we aimed to perform comprehensive risk evolution analysis and quantitative risk assessment of hot work using adaptive bow tie (ABT) and Petri nets (PNs). The ABT model is improved to analyze deep causes, consequences, and their relationships. PNs are employed to conduct quantitative risk assessment. To address uncertainties, the triangular membership function and improved regional center method are proposed to calculate the confidences of all places and transitions in the PNs. The probabilities are finally obtained by data fitting method. The study found that the most likely consequence of hot work accident is near miss. The results also indicate that hot work accident prevention measures should emphasize on the isolation of workplace, combustible gases test, removal of combustible substances, job organization, and job training. Besides, emergency devices maintenance and emergency drilling are critical to prevent accidents from expanding.