Considerations for generating meaningful HRA data: Lessons learned from HuREX data collection

To enhance the credibility of human reliability analysis, various kinds of data have been recently collected and analyzed. Although it is obvious that the quality of data is critical, the practices or considerations for securing data quality have not been sufficiently discussed. In this work, based on the experience of the recent human reliability data extraction projects, which produced more than fifty thousand data-points, we derive a number of issues to be considered for generating meaningful data. As a result, thirteen considerations are presented here as pertaining to the four different data extraction activities: preparation, collection, analysis, and application. Although the lessons were acquired from a single kind of data collection framework, it is believed that these results will guide researchers to consider important issues in the process of extracting data.