BibTeX

@inproceedings{GrothPSAM2012_HRA,
title = {A Model-Based Approach to HRA: Example Application and Quantitative Analysis},
author = {Katrina M Groth and Song-Hua Shen and Johanna Oxstrand and Ali Mosleh and Dana Kelly},
year = {2012},
date = {2012-06-01},
booktitle = {Proceedings of the International Conference on Probabilistic Safety Assessment and Management (PSAM 11)},
address = {Helsinki, Finland},
abstract = {Separate papers in this conference describe a model-based approach to Human Reliability Analysis (HRA) and the corresponding qualitative portion developed as part of various research efforts to improve the robustness of HRA methods. This paper describes the application of the methodology to an example from a nuclear plant probabilistic risk assessment (PRA). The goal of this exercise was to ensure that the research effort’s goals were met and to identify potential areas for improvement and to assess the methodology’s usefulness for supporting qualitative HRA. The example is based on a typical scenario from nuclear plant probabilistic risk assessment (PRA), a Steam Generator Tube Rupture (SGTR) scenario. We demonstrate the construction of the corresponding Crew Response Tree (CRT), and identify Crew Failure Modes (CFM) for the CRT scenarios using a fault tree (FT) approach. The qualitative methodology is intended to be method-agnostic from the perspective of quantification, and the outputs should be able to be used in a number of existing and future approaches to quantification. In this paper a Bayesian Network (BN) approach was selected for quantification of the impact of relevant Performance Shaping Factors (PSFs) on crew failure modes. The linking of various logic models (CRT, FTs, and BNs) is made using the Hybrid Causal Logic (HCL) methodology, which is also used to quantify HFEs of the scenario from the integrated model. This paper provides a detailed description of the exercise as well as the identified areas for improvements. The current state of both the qualitative approach and the linkage to Bayesian Networks for quantification are discussed.},
keywords = {causal models, crew failure modes, human error, Human Reliability Analysis (HRA), nuclear power},
pubstate = {published},
tppubtype = {inproceedings}
}

Abstract

Separate papers in this conference describe a model-based approach to Human Reliability Analysis (HRA) and the corresponding qualitative portion developed as part of various research efforts to improve the robustness of HRA methods. This paper describes the application of the methodology to an example from a nuclear plant probabilistic risk assessment (PRA). The goal of this exercise was to ensure that the research effort’s goals were met and to identify potential areas for improvement and to assess the methodology’s usefulness for supporting qualitative HRA. The example is based on a typical scenario from nuclear plant probabilistic risk assessment (PRA), a Steam Generator Tube Rupture (SGTR) scenario. We demonstrate the construction of the corresponding Crew Response Tree (CRT), and identify Crew Failure Modes (CFM) for the CRT scenarios using a fault tree (FT) approach. The qualitative methodology is intended to be method-agnostic from the perspective of quantification, and the outputs should be able to be used in a number of existing and future approaches to quantification. In this paper a Bayesian Network (BN) approach was selected for quantification of the impact of relevant Performance Shaping Factors (PSFs) on crew failure modes. The linking of various logic models (CRT, FTs, and BNs) is made using the Hybrid Causal Logic (HCL) methodology, which is also used to quantify HFEs of the scenario from the integrated model. This paper provides a detailed description of the exercise as well as the identified areas for improvements. The current state of both the qualitative approach and the linkage to Bayesian Networks for quantification are discussed.