Katrina M. Groth

A data-informed PIF hierarchy for model-based Human Reliability Analysis

Abstract This paper addresses three problems associated with the use of Performance Shaping Factors in Human Reliability Analysis. (1) There are more than a dozen Human Reliability Analysis (HRA) methods that use Performance Influencing Factors (PIFs) or Performance Shaping Factors (PSFs) to model human performance, but there is not a standard set of PIFs used among the methods, nor is there a framework available to compare the PIFs used in various methods. (2) The PIFs currently in use are not defined specifically enough to ensure consistent interpretation of similar PIFs across methods. (3) There are few rules governing the creation, definition, and usage of PIF sets. This paper introduces a hierarchical set of PIFs that can be used for both qualitative and quantitative HRA. The proposed PIF set is arranged in a hierarchy that can be collapsed or expanded to meet multiple objectives. The PIF hierarchy has been developed with respect to a set fundamental principles necessary for PIF sets, which are also introduced in this paper. This paper includes definitions of the PIFs to allow analysts to map the proposed PIFs onto current and future HRA methods. The standardized PIF hierarchy will allow analysts to combine different types of data and will therefore make the best use of the limited data in HRA. The collapsible hierarchy provides the structure necessary to combine multiple types of information without reducing the quality of the information.

Hybrid causal methodology and software platform for probabilistic risk assessment and safety monitoring of socio-technical systems

Abstract This paper introduces an integrated framework and software platform for probabilistic risk assessment (PRA) and safety monitoring of complex socio-technical systems. An overview of the three-layer hybrid causal logic (HCL) modeling approach and corresponding algorithms, implemented in the Trilith software platform, are provided. The HCL approach enhances typical PRA methods by quantitatively including the influence of soft causal factors introduced by human and organizational aspects of a system. The framework allows different modeling techniques to be used for different aspects of the socio-technical system. The HCL approach combines the power of traditional event sequence diagram (ESD)event tree (ET) and fault tree (FT) techniques for modeling deterministic causal paths, with the flexibility of Bayesian belief networks for modeling non-deterministic cause–effect relationships among system elements (suitable for modeling human and organizational influences). Trilith enables analysts to construct HCL models and perform quantitative risk assessment and management of complex systems. The risk management capabilities included are HCL-based risk importance measures, hazard identification and ranking, precursor analysis, safety indicator monitoring, and root cause analysis. This paper describes the capabilities of the Trilith platform and power of the HCL algorithm by use of example risk models for a type of aviation accident (aircraft taking off from the wrong runway).

Bridging the gap between HRA research and HRA practice: A Bayesian network version of SPAR-H

The shortcomings of Human Reliability Analysis (HRA) have been a topic of discussion for over two decades. Repeated attempts to address these limitations have resulted in over 50 HRA methods, and the HRA research community continues to develop new methods. However, there remains a gap between the methods developed by HRA researchers and those actually used by HRA practitioners. Bayesian Networks (BNs) have become an increasingly popular part of the risk and reliability analysis framework over the past decade. BNs provide a framework for addressing many of the shortcomings of HRA from a researcher perspective and from a practitioner perspective. Several research groups have developed advanced HRA methods based on BNs, but none of these methods has been adopted by HRA practitioners in the U.S. nuclear power industry or at the U.S. Nuclear Regulatory Commission. In this paper we bridge the gap between HRA research and HRA practice by building a BN version of the widely used SPAR-H method. We demonstrate how the SPAR-H BN can be used by HRA practitioners, and we also demonstrate how it can be modified to incorporate data and information from research to advance HRA practice. The SPAR-H BN can be used as a starting point for translating HRA research efforts and advances in scientific understanding into real, timely benefits for HRA practitioners.

Deriving causal Bayesian networks from human reliability analysis data: A methodology and example model

Within the probabilistic risk assessment community, there is a widely acknowledged need to improve the scientific basis of human reliability analysis (HRA). This has resulted in a number of independent research efforts to gather empirical data to validate HRA methods and a number of independent research efforts to improve theoretical models of human performance used in HRA. This paper introduces a methodology for carefully combining multiple sources of empirical data with validated theoretical models to enhance both qualitative and quantitative HRA applications. The methodology uses a comprehensive set of performance influencing factors to combine data from different sources. Further, the paper describes how to use data to gather insights into the relationships among performance influencing factors and to build a quantitative HRA causal model.  To illustrate how the methodology is applied, we introduce the Bayesian network model that resulted from applying the methodology to two sources of human performance data from nuclear power plant operations. The proposed model is introduced to demonstrate how to develop causal insights from HRA data and how to incorporate these insights into a quantitative HRA model. The methodology in this paper provides a path forward for carefully incorporating emerging sources of human performance data into an improved HRA method. The proposed model is a starting point for the next generation of data-informed, theoretically-validated HRA methods.

A Bayesian method for using simulator data to enhance human error probabilities assigned by existing HRA methods

In the past several years, several international organizations have begun to collect data on human performance in nuclear power plant simulators. The data collected provide a valuable opportunity to improve human reliability analysis (HRA), but these improvements will not be realized without implementation of Bayesian methods. Bayesian methods are widely used to incorporate sparse data into models in many parts of probabilistic risk assessment (PRA), but Bayesian methods have not been adopted by the HRA community. In this paper, we provide a Bayesian methodology to formally use simulator data to refine the human error probabilities (HEPs) assigned by existing HRA methods. We demonstrate the methodology with a case study, wherein we use simulator data from the Halden Reactor Project to update the probability assignments from the SPAR-H method. The case study demonstrates the ability to use performance data, even sparse data, to improve existing HRA methods. Furthermore, this paper also serves as a demonstration of the value of Bayesian methods to improve the technical basis of HRA.

Capturing cognitive causal paths in human reliability analysis with Bayesian network models

reIn the last decade, Bayesian networks (BNs) have been identified as a powerful tool for human reliability analysis (HRA), with multiple advantages over traditional HRA methods. In this paper we illustrate how BNs can be used to include additional, qualitative causal paths to provide traceability. The proposed framework provides the foundation to resolve several needs frequently expressed by the HRA community. First, the developed extended BN structure reflects the causal paths found in cognitive psychology literature, thereby addressing the need for causal traceability and strong scientific basis in HRA. Secondly, the use of node reduction algorithms allows the BN to be condensed to a level of detail at which quantification is as straightforward as the techniques used in existing HRA. We illustrate the framework by developing a BN version of the critical data misperceived crew failure mode in the IDHEAS HRA method, which is currently under development at the US NRC [45]. We illustrate how the model could be quantified with a combination of expert-probabilities and information from operator performance databases such as SACADA. This paper lays the foundations necessary to expand the cognitive and quantitative foundations of HRA.

Early-Stage Quantitative Risk Assessment to Support Development of Codes and Standard Requirements for Indoor Fueling of Hydrogen Vehicles

Sandia National Laboratories is developing the technical basis for assessing the risk of hydrogen infrastructure for use in the development of relevant codes and standards. The development of codes and standards is an important step in ensuring the safe design and operation of the hydrogen fuel cell infrastructure. Codes and standards organizations are increasingly using risk-informed processes to establish code requirements. Sandia has used Quantitative Risk Assessment (QRA) approaches to risk-inform safety codes and standards for hydrogen infrastructures. QRA has been applied successfully for decades in 3 many industries, including nuclear power, aviation, and offshore oil. However, the hydrogen industry is a relatively new application area for QRA, and several gaps must be filled before QRA can be widely applied to reduce conservatisms that influence the safety requirements for hydrogen installations. This report documents an early-stage QRA for a generic, code-compliant indoor hydrogen fueling facility. The goals of conducting this activity were threefold: to provide initial insights into the safety of such facilities; to recommend risk-informed changes to indoor fueling requirements in safety codes and standards; and to evaluate the quality of existing models and data available for use in hydrogen installation QRA. The report provides several recommendations for code changes that will improve indoor fueling safety. Furthermore, the report provides insight into gaps in the QRA process that must be addressed to provide greater confidence in the QRA results.

Hydrogen quantitative risk assessment workshop proceedings.

The Quantitative Risk Assessment (QRA) Toolkit Introduction Workshop was held at Energetics on June 11-12. The workshop was co-hosted by Sandia National Laboratories (Sandia) and HySafe, the International Association for Hydrogen Safety. The objective of the workshop was twofold: (1) Present a hydrogen-specific methodology and toolkit (currently under development) for conducting QRA to support the development of codes and standards and safety assessments of hydrogen-fueled vehicles and fueling stations, and (2) Obtain feedback on the needs of early-stage users (hydrogen as well as potential leveraging for Compressed Natural Gas [CNG], and Liquefied Natural Gas [LNG]) and set priorities for %E2%80%9CVersion 1%E2%80%9D of the toolkit in the context of the commercial evolution of hydrogen fuel cell electric vehicles (FCEV). The workshop consisted of an introduction and three technical sessions: Risk Informed Development and Approach; CNG/LNG Applications; and Introduction of a Hydrogen Specific QRA Toolkit.

Hybrid methodology and software platform for probabilistic risk assessment

This paper introduces the software implementation of a hybrid methodology for probabilistic risk assessment (PRA) of complex systems. The software, called IRIS (Integrated Risk Information System) combines a user-friendly graphical interface with a powerful computational engine. The framework includes a multi-layered modeling approach, combining Event Sequence Diagrams, Fault Trees, and Bayesian Belief Networks in a hybrid causal logic (HCL) model. This allows the most appropriate modeling techniques to be applied in the different domains of the system. At its core IRIS brings related perspectives of system safety, hazard analysis, and risk analysis into a unifying framework.

Challenges in leveraging existing human performance data for quantifying the IDHEAS HRA method

This article documents an exploratory study for collecting and using human performance data to inform human error probability (HEP) estimates for a new human reliability analysis (HRA) method, the IntegrateD Human Event Analysis System (IDHEAS). The method was based on cognitive models and mechanisms underlying human behaviour and employs a framework of 14 crew failure modes (CFMs) to represent human failures typical for human performance in nuclear power plant (NPP) internal, at-power events [1]. A decision tree (DT) was constructed for each CFM to assess the probability of the CFM occurring in different contexts. Data needs for IDHEAS quantification are discussed. Then, the data collection framework and process is described and how the collected data were used to inform HEP estimation is illustrated with two examples. Next, five major technical challenges are identified for leveraging human performance data for IDHEAS quantification. These challenges reflect the data needs specific to IDHEAS. More importantly, they also represent the general issues with current human performance data and can provide insight for a path forward to support HRA data collection, use, and exchange for HRA method development, implementation, and validation.