A Cognitive-Mechanistic Human Reliability Analysis Framework: A Nuclear Power Plant Case Study

TL;DR

The paper addresses the limitations of traditional human reliability analysis by introducing a cognitively grounded framework (COGMIF) that couples ACT-R-based operator cognition with TimeGAN-generated data to augment the IDHEAS-ECA methodology in HTGR contexts. By building a mechanistic digital twin, augmenting data withGAN-generated time-series, and framing the results within a Bayesian network, COGMIF enables scalable, explainable estimation of human error probabilities ($P_{HFE}$) that reflect cognitive, procedural, and temporal factors. The approach is validated against HTGR simulations, showing temporal accuracy and robust HEP estimates with comparable alignment to SPAR-H, while revealing key drivers of risk through influence-strength analyses. This methodology offers a credible, computationally efficient pathway to integrate cognitive theory into industrial HRA, with practical implications for next-generation reactor safety assessment and operator performance optimization.

Abstract

Traditional human reliability analysis (HRA) methods, such as IDHEAS-ECA, rely on expert judgment and empirical rules that often overlook the cognitive underpinnings of human error. Moreover, conducting human-in-the-loop experiments for advanced nuclear power plants is increasingly impractical due to novel interfaces and limited operational data. This study proposes a cognitive-mechanistic framework (COGMIF) that enhances the IDHEAS-ECA methodology by integrating an ACT-R-based human digital twin (HDT) with TimeGAN-augmented simulation. The ACT-R model simulates operator cognition, including memory retrieval, goal-directed procedural reasoning, and perceptual-motor execution, under high-fidelity scenarios derived from a high-temperature gas-cooled reactor (HTGR) simulator. To overcome the resource constraints of large-scale cognitive modeling, TimeGAN is trained on ACT-R-generated time-series data to produce high-fidelity synthetic operator behavior datasets. These simulations are then used to drive IDHEAS-ECA assessments, enabling scalable, mechanism-informed estimation of human error probabilities (HEPs). Comparative analyses with SPAR-H and sensitivity assessments demonstrate the robustness and practical advantages of the proposed COGMIF. Finally, procedural features are mapped onto a Bayesian network to quantify the influence of contributing factors, revealing key drivers of operational risk. This work offers a credible and computationally efficient pathway to integrate cognitive theory into industrial HRA practices.

Figures (17)

• Figure 1: IDHEAS-ECA HRA Process xing2020integrated
• Figure 2: The Proposed Framework COGMIF
• Figure 3: Content of Experimented Emergency Operational Procedures
• Figure 4: Integrated Schematic Diagram of Exp2 Basic Operations and Refined Protocols for LISP Model Construction
• Figure 5: Refined Protocols of Exp1 and Exp3 for LISP Model Construction