Applications of Dynamic Human Reliability Analysis (dHRA) for Context Aware Operations

Abstract

Human Reliability Assessment is an essential tool for estimating a human operator’s propensity to commit errors based on available information and estimation of performance shaping factors. Even when estimates are approximate, HRA can provide valuable insights into task performance and vulnerabilities to accidental or volitional human actions. Performance Shaping Factors (PSFs) for Human Reliability Analysis are traditional determined as static across all operating conditions. However, it is likely that the PSFs are dynamic and dependent on the state of the plant. For example, available time, stress, and complexity factors change rapidly if the plant enters abnormal operating conditions. Nuclear Power Plants (NPP) are modernizing control rooms with digital systems and accompanying Human Machine Interfaces (HMIs) to enable operation past their planned 40-year life expectancies. New research is needed to identify how digital HMIs interact with dynamic PSFs to influence Human Error Reliability. Here we examine an experimental paradigm utilizing a microworld with non-expert operators to assess how context-aware HMIs could reduce human error by adapting the presentation of information based on the dynamic PSFs estimated from the plant’s current state.