ABSTRACT
In Probabilistic Safety Assessments, Human Reliability Analysis (HRA) methods are used to estimate the probability of human failure events. Starting with the earliest developments, data for HRA has been persistently viewed as lacking. Indeed, many sources of HRA-relevant data exist and many efforts to collect the data have been and are being pursued. This paper presents a review of the literature aimed at assessing the state of HRA data. Based on an overview of existing databases, it attempts to relate the various sources of data and databases by considering the relationships among the types of data and the associated applications of the data. There are key distinctions among the intended applications of the data, with implications for data they require. Even if HRA methods share common traits, the diversity of HRA methods has also resulted in method-specific data requirements. Moreover, the sensitivity and, in some cases, the proprietary character of the data represents additional challenges. Finally, the review highlights some trends in the types of data collected and their applications over time and some current developments.
