Dependency definitions for quantitative human reliability analysis

Abstract

Human reliability analysis (HRA) identifies the causal factors impacting the occurrence of human failure events and quantifies probabilities based on those causal factors. Quantification requires understanding the dependency structures that exist between failure events and causal factors. Most HRA methods incorporate the dependency framework established in the Technique for Human Error Rate Prediction (THERP), which approaches dependency as a simple multiplier on HEPs resulting from considering only a few factors. Accordingly, modern HRA methods have a limited ability and accuracy when characterizing dependency. The fundamental concept of dependency in HRA methods is approaching 60 years without a serious assessment of its veracity or applicability. Misconceptions and confusion surrounding fundamental concepts in HRA are largely responsible for the lack of substantive improvement in the technical basis of dependency. This paper defines dependency for HRA and provides clarification and unification around related concepts. We propose a standardized library of key terms and mathematics to provide a basis for the development of a dependency framework. We clarify the meaning of, and difference between, critical dependency concepts and propose a general lexicography for decomposing scenarios. We present a holistic linguistic and mathematical definition for dependency in HRA.