Structured information analysis for human reliability analysis of emergency tasks in nuclear power plants

Being supported by scarce empirical data, most of the performance influencing factors in human reliability analysis (HRA) have to be assessed on the basis of the analyst's knowledge on the human performance in given tasks and their context. Therefore, the outcome of HRA may only be warranted by a proper application of their knowledge based on sufficient information about the tasks and situations. However, most of the HRA methodologies, including the newly developed ones, focus on the provision of cognitive models, error mechanisms, error types and analysis method while leaving the information collection mostly in the hands of the analyst. This paper suggests structured information analysis (SIA), which helps HRA analysts in collecting and structuring such information on tasks and contexts. The SIA consists of three parts: the scenario analysis, the goal-means analysis, and the cognitive function analysis. An expert evaluation showed that this three-part information analysis allowed more expressiveness and hence more confidence on the error prediction than ASEP HRA.