Development of a risk based methodology to consider influence of human failure in industrial plants operation

Human Failures are one of the most unexplored causes in industrial accidents. Since there is still lack of heeds to qualify as well as quantify Human Errors, in this paper the authors attempt to highlight the importance of paying attention to qualitative methods in implementing quantitative risk analyses mainly in the framework of estimating more accurate Human Error Probability (HEP). A key point in evaluating such a risk is considering non-linear socio-Technical interaction in system to develop causal network for the accident scenario. An application of qualitative and quantitative Bayesian Network (BN) is therefore presented. The study shows that human performance has the most changes in the light of evidences. The developed methodology applied to a case study of an operation in field of Oil and Gas.