A method for performing human reliability assessment (HRA), called Petro-HRA, has recently been developed for the Norwegian oil and gas industry. A main objective of this method is to provide input to the quantitative risk assessment (QRA) in the form of human error probabilities. It is based on the SPAR-H method (developed by the US Nuclear Regulatory Commission), which has been used within the nuclear industry for years, and is seen as a strong basis for HRA, but which is a method that is currently not widely applied in other industries. An attempt is now made to apply it in the oil and gas industry. One possible challenge then is that the method provides input to the QRA by adopting a different perspective on risk, compared with that used in the nuclear industry. In this paper, we discuss whether the shift, from application in the nuclear industry to an industry where risk might be interpreted differently, influences the appropriateness of the method.
A main finding is that, to ensure quality input to the QRA, the Petro-HRA method should be further aligned with the risk perspective of the Norwegian oil and gas industry. It evidently differs from that of the nuclear industry, which is of importance to the quality of the results and input to the QRA. The nuclear industry perceives risk as a combination of consequences and probability, while the petroleum industry sees risk, rather, as a combination of consequences and associated uncertainties. In the paper, we clarify why it is important to achieve such alignment, by showing how the method could produce different results with a higher focus on uncertainties. We further indicate how to improve the method and guidelines though better treatment of uncertainty in the HRA. The objective is to propose a way to better align the Petro-HRA method, in light of the risk perspective of the Norwegian oil and gas industry.