ML17268A110

Ehpg Paper - 2017-2

ML17268A110
Person / Time
Issue date:	09/28/2017
From:	Sean Peters NRC/RES/DRA/HFRB
To:
Sean Peters
Shared Package
ML17268A107	List: ML17268A109 ML17268A110
References
Download: ML17268A110 (10)

Text

The US Nuclear Regulatory Commission's Human Factors and Human Reliability Analysis Research Programs Sean Peters, US Nuclear Regulatory Commission Abstract Over the last decade, the NRC staff has made numerous advancements in the state of science in its human and organizational factors and human reliability analysis research programs. In the area of control room human factors, the program covered advanced technologies, new concepts of operation, adaptive automation, and computerized procedures. In other human and organizational factors areas, the staff has looked into non-destructive examination techniques, safety culture, and fitness for duty. Finally, in human reliability analysis (HRA), the staff worked to improve the consistency and effectiveness of HRA methods through benchmarking and method development activities, and has looked into expanding the use of HRA through new and novel domains such as fires, flooding, severe accidents, spent fuel pools, fuel handling operations, and by-product and medical applicati ons. The NRC continues to improve the state of science in human reliability analysis and human and organizational factors.

1.0 Background Like in other major socio-technical industries, humans are integral to the safe use of nuclear materials. As such, the NRC continues to focus its regulatory research programs on understanding the human contributions to safety in order to ensure adequate protection and common defense and security of nuclear materials. As technology evolves, so do the challenges that humans face in operating nuclear systems. To manage this technological shift, o ver the last decade, the NRC has made numerous adv ancements in the state of science in its human and organizational factors and human reliability analysis research programs.

2.0 Human Performance for New and Advanced Control Room Designs

2.1 Objective The objective of this work is to identify and prioritize human performance research that will be needed to provide a technical basis for developing regulatory guidance for review of licensees' implementation of new technology in new and advanced nuclear power plants (NPPs).

2.2 Research Approach NRC began this effort by evaluating current indu stry trends and organizing them into seven human factors engineering (HFE) topic areas: (1) Role of personnel and automation; (2) Staffing and training; (3) Normal operations management; (4) Disturbance and emergency management; (5) Maintenance and change management; (6) Plant design and construction; and (7) HFE methods and tools. Next, a panel of independent subject-matter experts representing various disciplines and backgrounds prioritized the areas. NUREG/CR-6947, "Human Factors Considerations with Respect to Emerging Tech nology in Nuclear Power Plants," documents the results of the study. The findings from the study are being used to develop long-term research plans addressing human performance within these technology areas. The three projects underway are described below.

Advances in Human Factors Engineering Methods and Tools

This project looks into current state of human factors engineering methods and tools (e.g., measures of complexity, methods for integrated system validation, etc.) and looks into their suitability for use in nuclear power plant licensing. This work has resulted in the development of detailed review guidance for applying human perfo rmance models to the evaluation of NPP designs.

Roles of Automation and Complexity in Control Rooms This study examines the impact of automation on control room design, specifically: (1) Operator performance during normal, abnormal, and emergency operations; (2) The reliability of operator's use of automation systems including existing methods for assessing impacts; and (3) Operator performance when the automation fails or is in a degraded state.

Update Existing Human Factors Engineering Regulatory Guidance

This project updates the human factors guidance that is currently being used in NPP licensing

reviews. Specifically, the following guidance documents are being updated, NUREG-0711, "Human Factors Engineering Program Review Model," and NUREG-0700, "Human-System Interface Design Review Guidelines," as the acceptance criteria. These updates will help the guidance keep pace with modern I&C systems and advancing levels of automation.

2.3 Status In addition to NUREG/CR-6947 mentioned above, two additional reports are under development: (1) Integrated System Validation, and (2) Cognitive Task Analysis. Two technical reports are currently available related to automation and complexity: (1) Human-System Interfaces to Automatic Systems, and (2) The Effects of Degraded Digital Instrumentation and

Control Systems on Human-system Interfaces and Operator Performance. A revision to NUREG-0711 was published in 2012. Due to its size, NUREG-0700 is being updated in two phases. The phase 1 update is complete and should be issued by early 2018, with the phase 2

update to follow.

3.0 Human Performance Test Facility Research

3.1 Objective The objective of this work is to conduct research assessing the impact of new designs on human performance with a larger and lower cost subject pool as a supplement to the research being performed at the Halden Reactor Project.

3.2 Research Approach To meet this objective, the NRC procured two copies of a desktop computer-based nuclear control room simulator to conduct this research-one copy is housed at NRC headquarters and the other is at the University of Central Florida (UCF) under contract with the NRC. The simulators have the following characteristics:

• Generic pressurized-water reactor.

• Westinghouse, 3-Loop.

• RETACT thermal-hydraulics code.

• Reprogrammable analog panel, soft controls, digital interfaces.

• Supporting documents (e.g., procedures, technical specifications).

The NRC and UCF are working together to des ign and conduct human-in-the-loop experiments. This research is expected to produce nuclear-specific human performance data that aid in the evaluation of prioritized issues identified in NUREG/CR-6947, "Human Factors Considerations with Respect to Emerging Technologies in Nuclear Power Plants." These issues include the

impact that new designs, technologies, and concepts of operations have on human performance.

3.3 Status The information gained will be incorporated in updat es to the NRC staff's human factors review guidance NUREG-0700, "Human-System Interface Design Review Guidelines;" NUREG-0711, "Human Factors Engineering Program Review Model;" and in updates to the NRC's Human Reliability Analysis method development initiatives.

4.0 Human Factors in Nondestructive Examination 4.1 Objective The objective of this work is to support the NRC regulatory offices by conducting research to identify, understand, and prioritize the human factors that are most likely to impact personnel performance during nondestructive examination (NDE) in nuclear power plants.

4.2 Research Approach NDE is a means of testing a specimen or co mponent without damaging or destroying it. NDE plays a vital role in ensuring the safety of nuclear power plant operations. The effective use of NDE to find flaws in a component can be dependent on the personnel performing the examination, the design of the task, along with the environmental and organizational conditions within which personnel carry out the task. These human factors issues must be considered in

order to have reasonable assurance that a licensee is meeting the NRC's regulatory requirements.

NRC is currently focused on understanding the human factors that can challenge performance in a specific type of NDE called ultrasonic testing (UT). The first stage of the research is to characterize the current state of human factors research in NDE by conducting a literature review. Next, NRC staff plans to conduct a task analysis to develop a thorough understanding of the process of performing NDE in nuclear power plants. Then, the RES staff plan to work with industry to prioritize the human factors identified. This research effort will assist the NRC in planning future research, evaluating whether changes are needed in regulatory requirements and providing technical justification for regulatory decisions.

4.3 Status The NRC staff completed the literature review of human factors considerations in non-destructive examinations in February, 2017. The technical letter report titled, "Review of Human Factors Research in Nondestructive Examination," is available in the Agencywide Document and Management System (ADAMS) at accession number ML17059D745. The task analysis is ongoing in 2017, and the prioritization of human factors issues is planned for 2018.

5.0 Fitness for Duty

5.1 Objective The objective of this work is to support the NRC regulatory offices in the development of the technical basis for rulemaking and implementation of the Fitness for Duty rule, Title 10 of the

Code of Federal Regulations (10 CFR) Part 26, "Fitness for Duty Programs."

5.2 Research Approach The NRC requires certain licensees to have an FFD program to provide reasonable assurance that licensee personnel (1) are trustworthy; (2) will perform their tasks in a reliable manner; (3) are not under the influence of any substance, legal or illegal, that may impair their ability to perform their duties; and (4) are not mentally or physically impaired from any cause that can adversely affect their ability to safely and competently perform their duties.

RES participates in and provides technical support to several working groups engaged in Fitness for Duty (FFD) rulemakings and program implementation. Two main initiatives related to Part 26 are described

below.

Fatigue Since the issuance of its regulation on fatigue for workers at NPPs, a number of issues have arose with the implementation of these requirem ents. The staff is working to resolve these issues. To support the rule, RES is developing guidance for implementing the fatigue management requirements, and RES has been looking at new methods to manage fatigue in the workplace and technologies for assessing fatigue as well as other possible types of impairment.

Drug and Alcohol Testing

NRC continues to evaluate the latest advancements in the area of drug and alcohol testing.

The latest topics of interest have included the use of alternate specimens such as breath and saliva for testing. This is following the recent policy adoptions of these new testing methods in the private sector and by the U.S. Department of Health and Human Services. In addition to rulemaking support, NRC staff has been assisting in the development of regulatory guidance that describes the methods that the staff considers acceptable for complying with the drug testing provisions in Part 26.

5.3 Status The results from the drug and alcohol initiatives will be published as a NUREG/CR in the

ongoing series of technical basis reports the NRC has periodically published since the FFD rule was first implemented in the early 1990s. The staff is currently working on an update to Regulatory Guide 5.73, "Fatigue Management for Nuclear Power Plant Personnel," that will incorporate implementation guidance based on these questions and lessons learned since the fatigue requirements went into effect in 2011.

6.0 Safety Culture 6.1 Objective The objective of this work is to provide technical expertise related to human and organizational performance to support the NRC's safety culture oversight and policymaking activities.

6.2 Research Approach The culture of an organization affects the performance of the people in it. Weaknesses in an organization's safety culture may set the stage for equipment failures and human errors that can have an adverse impact on safe performance. The NRC has long recognized the importance of

maintaining a positive safety culture in nuclear operations. The NRC's Safety Culture Policy Statement (76 FR 34773; June 14, 2011

) provides the Commission's expectation that individuals and organizations establish and maintain a positive safety culture commensurate with the safety and security significance of their activities and the nature and complexity of their organizations and functions.

RES provides ongoing technical support for safety culture activities across the NRC. Past projects have included conducting research to understand the underlying relationship between safety culture and safety performance, reviewi ng methods for assessing safety culture, and developing educational materials to increase awareness and understanding of the importance of a positive safety culture. Most recently, NRC staff have been working with the International Atomic Energy Agency (IAEA) to develop safety culture perception questionnaires for use by nuclear power plant licensees and regulatory agencies as one tool in an overall safety culture assessment. In addition, the NRC's Office of En forcement leads the Safety Culture Advisory Committee, which coordinates safety culture activities across the agency.

6.3 Status

Research on the relationship between safety culture and safety performance is documented in a technical report titled, "Independent Evaluation of INPO's Nuclear Safety Culture Survey and Construct Validation Study." This technical report can be found in the NRC's ADAMS at accession number ML12172A093. The manual for the IAEA safety culture perception questionnaire for license holders is publicly available as a working document on the IAEA's website. The manual includes a copy of the questionnaire and offers high-level guidance for how to conduct a safety culture survey. The questionnaire is currently used by the IAEA to support their independent safety culture assessment reviews.

Updates on safety culture activities and new educational materials can be accessed from the NRC's safety culture Web site at http://www.nrc.gov/about-nrc/safety-culture.html.

7.0 Human Reliability Analysis Consistent with the NRC's policy statement on the use of probabilistic risk assessment (PRA) and staff requirements memoranda (SRMs) for achieving an appropriate PRA quality for NRC risk-informed regulatory decisionmaking, t he NRC has established a phased approach to PRA quality. This phased approach includes an action plan for stabilizing the PRA quality expectations and requirements to address PRA technical issues. Human reliability analysis (HRA) is an important PRA element. HRA is a structured approach used to identify potential human failure events and to systematically estimate the probability of those errors using data, models, or expert judgment.

The Commission identified the need for HRA data and models in the SRMs M061020 , dated November 8, 2006, and M090204B, dated February 18, 2009. In SRM-M061020, the Commission directed the Advisory Committee on Reactor Safeguards (ACRS) to work with the staff and external stakeholders to evaluate different human reliability models in an effort to propose a single model for the agency to use or guidance on which model(s) should be used in specific circumstances.

In SRM-M090204B, the Commission directed the staff to work with industry and international partners to test the performance of U.S. nuclear power plant operating crews and to keep the Commission informed of the status of its HRA data and benchmarking projects. 8.0 Human Reliability Analysis Data Repository 8.1 Objective RES developed the human performance data collection method and tool (i.e., Scenario Authoring, Categorization, and Debriefing Application [SACADA]) to collect licensed operator simulator training data to inform human error probability (HEP) estimations in human reliability analysis (HRA).

The objective is to acquire a large quantity of licensed operator simulator exercise data to provide statistical indications of the reliability of performing various tasks in the main control room.

8.2 Research Approach The staff's approach is to use the similarity-matching concept to identify the empirical data that can be used to inform the HEPs of the human failure events of interest. Similarity matching is based on the situational profile in challenging nuclear power plant operators in detecting the cues of plant malfunctions, understanding the situation, making correct decisions, and executing correct actions with the additional consideration of team communication and supervision. This human-centered approach differs from traditional task-centered or component-centered approaches (e.g., turn a switch) and allows combining data of different tasks with similar situational profiles to inform HEP estimates.

This approach is expected to significantly increase data usability.

A successful data collection program should include high data reliability and a long collection period to acquire enough data for statistical indications. To achieve high data reliability, the SACADA data are entered by the plant staff (operator trainers and reactor operators) when the information is still fresh in the individuals' memories. The key SACADA human performance data can be divided into two types.

The first type of data is the situational or performance challenge profile, which is entered by the scenario designers (i.e., operator trainers). Each human task identified in the simulation scenario has its own situational profile that is represented by a set of performance-influencing factors whose states can be objectively identified. Therefore, the scenario designers could enter the data with high reliability.

The second type of data is the operators' performance results. This type of data includes the operators' performance in meeting the expectations and, if there are performance deficiencies, it includes the information related to the performance deficiency. This data is entered by the plant operating crew during post simulation debriefings to ensure data reliability. For both types of data, the master set of factors are provided by SACADA for the operator trainers and operators to identify the most appropriate factors and factor statuses to characterize the situational profile and operator performance results.

Finally, to achieve the objective of a long-term data collection, SACADA was designed to mutually benefit the data providers and the NRC.

The data providers are the plants' training department and the operations department, whose main interest is to improve human performance. The SACADA tool allows for the plants to replace their current tools in collecting operator simulator performance information. When used to replace the plants' existing tools, SACADA has streamlined data entry that, in turn, has reduced data entry effort for other plant training applications.

8.3 Status SACADA has been collecting operator training data from a U.S. nuclear power station since 2012. The Halden Reactor Project has used the tool to collect information from operator simulator experiments. A few international research institutes have signed agreements with the NRC to test the SACADA tool, and several U.S. and International utilities are evaluating SACADA for use at their facilities. So far, the SACADA database has collected enough data for demonstrations on how to use the data to inform HRA. An international data workshop is being planned for early 2018.

9.0 General Human Reliability Analysis Method 9.1 Objective This work addresses the issues identified by the NRC in SRM-M061020 regarding the use of different human HRA methods contributing to the variability of probabilistic risk assessment (PRA)/HRA results.

9.2 Research Approach The research includes three parts: (1) develop a cognitive basis framework for HRA; (2) develop a stand-alone HRA method that reduces analyst-to-analyst variability for internal, at-power scenarios (referred to as "Integrated Human Event Analysis System" [IDHEAS])

and (3) develop a comprehensive HRA methodology that can be used to build HRA models for general applications, e.g. external events, shutdown, Level-3 PRA, and non-reactor applications The cognitive framework, while developed as the technical basis for IDHEAS, is a stand-alone product. The staff conducted a literature review to document the understanding of the cognitive aspects of nuclear power plant (NPP) crew behavior in response to plant upsets based on research results and findings in cognitive psychology, human factors, and organizational

behavior. The framework was developed to organize the results of cognitive research related to human performance in NPPs and to identify relevant performance influencing factors (PIFs) leading to crew failure. The outcome of the literature review and the cognitive basis framework for HRA were documented in NUREG-2114.

The NRC staff worked with the Electric Power Research Institute (EPRI) under a memorandum of understanding to develop a stand-alone HRA method that reduces analyst-to-analyst variability for internal, at-power scenarios. The method, IDHEAS for Nuclear Power Plant Internal Events at-Power Applications (NUREG-2199, Vol. 1), integrates the strengths of existing methods and addresses key sources contributing to analyst-to-analyst variability. The project team addressed four key sources of variability by incorporating the following features in IDHEAS:

• Integrating qualitative analysis guidance in existing HRA methods and developing additional guidance for task analysis.

• Incorporating the cognitive framework of the mechanisms underlying human errors as the human performance model for HRA.

• Developing the IDHEAS human error quantificat ion model based on the cognitive framework and experts' understanding of operator actions in internal, at-power scenarios.

• Verifying the quantification model and estimating the base HEPs through an expert panel that consists of human factors/cognitive engineers, PRA/HRA analysts, and operational personal from U.S. NPPs.

The NRC staff is also developing a version of the IDHEAS method to be used to create HRA methods for general HRA applications (known as IDHEAS-G). The general method is based on the cognitive basis framework and models human errors in four basic cognitive functions: (1) detecting information, (2) understanding and assessing plant status, (3) making decisions and planning actions, and (4) executing planned actions. The method models a broad set of factors that may lead to human errors under various conditions and can be used for both reactor and non-reactor applications.

9.3 Status The staff published the cognitive framework report, NUREG-2114 in January 2016, and the IDHEAS method for at-power applications, NUREG-2199, Vol. 1 in May 2016. The staff is currently engaged in testing the at-power method, finishing the development of the general method, and developing a simplified IDHEAS method for events and condition assessment (known as IDHEAS-ECA) in support of the NRC's Significant Determination Process and Accident Sequence Precursor program. The staff plans to deliver completed IDHEAS-G and IDHEAS-ECA methods for agency use in 2018, 10.0 Potential Human Errors for Medical Applications of Byproduct Materials

10.1 Objective

The objective of this effort is to support the Office of Nuclear Material Safety and Safeguard (NMSS) by (1) developing a report of understanding human error in radiation therapy, (2) publishing human reliability analysis (HRA)-informed training materials, and (3) demonstrating how to use the HRA informed job aid through illust rative examples. The NRC is interested in this activity because human error has been identified as an important contributor to significant events across multiple technologies and industries including medical application of byproduct materials.

10.2 Research Approach This work builds on an earlier user need, provided by NMSS to RES in 2003, to develop HRA capability specific to materials and waste applications. This earlier work was conducted in two

phases:

• Phase 1 work consisted of feasibility studies for developing NMSS capability in HRA. The feasibility study for materials applications addressed both medical and industrial applications.

• Phase 2 work focused on the recommendations from the feasibility study, namely, the development of job aids (e.g., HRA-informed decisionmaking aids) and associated training for NRC staff on HRA-informed issues in human performance in medical applications.

In this earlier work, the final products of the Phase 2 work, a prototype HRA-informed job aid (i.e., a database of risk-relevant human performance issues and historical errors, related to treatment steps) and associated training materials for medical applications (gamma-knife based), were presented to FSME staff and delivered to the NRC in December 2008. In all three cases, the products delivered to FSME were the starting point for the new development.

10.3 Status

RES published NUREG-2170, "A Risk-Informed Approach to Understanding Human Error in Radiation Therapy" in 2017.

11.0 Fire Human Reliability Analysis Methods Development

11.1 Object i ve The overall objective of this effort is to develop fire HRA methods beyond those currently in NUREG/CR-6850 (EPRI TR-1011989), "EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities," and to develop an HRA methodology and approach suitable for use in a fire

PRA. The intent of the fire HRA guidance developed through this effort is to support plants making the transition to 10 CFR 50.48(c) and NRC reviewers evaluating the adequacy of submittals from licensees making that transition.

Current research is focused on expanding initial guidance that has been developed in this project, "EPRI/NRC-RES Fire Human Reliability Analysis Guidelines-Final Report." NUREG-1921/EPRI 1023001, to address main control room abandonment (MCRA) scenarios. Such expanded guidance is expected to improve realism for fire HRA/PRA. Also, the importance of MCRA scenarios is expected to have increased because recent research has resulted in an increase in the frequency of main control room (MCR) board fires. In addition, this research will address issues related to command and control, and local actions, which are important to other PRA hazards, as well as fire PRA.

11.2 Research Approach NRC has worked collaboratively with the Electric Power Research Institute (EPRI) to develop a methodology and associated guidance for performing HRA in support of fire PRA. In July 2012, the NRC and EPRI jointly issued NUREG-1921 (EPRI 1023001), "EPRI/NRC-RES Fire Human Reliability Analysis Guidelines-Final Report." NUREG-1921 identified several issues or areas requiring further research. One of those areas is treatment of scenarios requiring operators to abandon the MCR. To address this need, NRC-RES and EPRI began working collaboratively in early 2015 to develop additional guidance for both loss of habitability (LOH) and loss of control (LOC) scenarios that result in MCRA. This guidance builds upon that already provided in the joint EPRI/NRC-RES Fire Human Reliability Anal ysis Guidelines and interactions between NRC and industry in the Frequently Asked Questions (FAQ) process. The updated guidance will be in the form of a NUREG report (or reports) and issued as a supplement(s) to NUREG-1921.

11.3 Status NRC and EPRI have begun development of additional HRA guidance for MCR abandonment scenarios. Currently, the first research product, EPRI/NRC-RES Fire Human Reliability Analysis Guidelines: Qualitative Analysis for Main Control Room Abandonment Scenarios (NUREG-1921, Supplement 1/EPRI 3002009215, is scheduled for publication in 2017. Follow-on work to develop HRA quantification guidance has begun and is expected to be published in 2018, including peer review and testing. NRC will continue to assist the Office of Nuclear Reactor Regulation (NRR) with the development of responses to NFPA 805 FAQs regarding HRA and will provide expert consulting as needed as NRR performs reviews of licensee submittals as well as support for other future activities that require fire HRA expertise.

12.0 Conclusions The NRC is staying abreast of the rapidly changing te chnology in the nuclear industry so that it can ensure that the human and organizational contributions to safety remain sufficient. In the area of control room human factors, the R ES program leads in advanced technologies, new concepts of operation, adaptive automation, and computerized procedures. In other human and organizational factors areas, the staff is looking into non-destructive examination techniques, safety culture, and fitness for duty. Finally, in human reliability analysis, the staff is working to improve the consistency and effectiveness of HRA methods through benchmarking and method development activities, and to expand the use of HRA through new and novel domains such as fires, flooding, severe accidents, spent fuel pools, fuel handling operations, and by-product and medical applications. The NRC continues to impr ove the state of science in human reliability analysis and human and organizational factors.