ML20154P550
| ML20154P550 | |
| Person / Time | |
|---|---|
| Issue date: | 05/23/1988 |
| From: | Stello V NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | |
| References | |
| TASK-PII, TASK-SE NACNSRRC, SECY-88-141, NUDOCS 8806030375 | |
| Download: ML20154P550 (108) | |
Text
-
(M c
n ik 1
W gv /
MaY 23' 1988 SECY-88-141 (Informat. ion)
For:
The Commissioners From:
Victor Stello, Jr.
Executive Director for Operations
Subject:
HUMAN FACTORS INITIATIVES AND PLANS
Purpose:
To provide the Commissioners with a description of the staff's human factors initiatives and plans and with the staff's disposition of the recommendations provided in a National Academy of Sciences (NAS) February 1988 report entitled Human Factors Research and Nuclear Safety.
Background:
On January 28, 1987, the staff recommended termination of the agency's Human Factors Program. Plan (SECY-87-19, Human Factors Program Plan -- Progress Report for Fiscal Year 1986).
The Commission indicated acceptance of that recommendation in a February 27, 1987, memorandum from S. Chilk to V. Stello. At the same time, the Commission requested "... a brief description of agency human factors initiatives in progress or planned and an evaluation of the recommendations of the expected June 1987 NAS study on the need for additional human factors research...."
A March 23, 1987, memorandum from V. Stello to H. Denton and E. S. Beckjord directed the Office of Nuclear Regulatory Research (RES), with the assistance of the Office of Nuclear Reactor Regulation (NRR), to provide an information paper responding to the Corrnission's request by October 1,1987. The date for providing the information paper to the Commission was based on the scheduled receipt of NAS recommendations on human factors research by the end of August 1987. On August 17, 1987, the NAS informed the staff that it could not provide its report l
as scheduled, and the staff notified the Connission of this delay on September 21, 1987 (SECY-87-240, Human Factors Research l
Planning for FY 1988 and Beyond).
The staff committed to l
provide the information paper responding to the Corr:ission's request approximately two months following receipt of the NAS report.
The report was received on February 29, 1988.
Contact:
A. Rubin, RES 49-23546 22$6$36373 0
4 The Comissioners Discussion:
RES, NRR, the Office for Analysis and Evaluation of Operational Data (AE00), and the Office of Nuclear Material Safety and Safeguards (NMSS) all have ongoing and planned activities in the area of human factors.
The following paragraphs provide a brief description of those activities.
The Office of Nuclear Regulatory Research.
RES has developed an expanded program for human factors research in response to Commission guidance and to recommendations in a December 1986 NAS report entitled Revitalizing Nuclear Safety Research.
The Reliability and Human Factors Branch in the Division of Reactor and Plant Systems was formed to take the lead in this area.
To date, efforts by RES have included identification of human factors research needs in coordination with NRC user offices and development of a Human Factors Research Program Plan (HFRPP
-- See Enclosure 1). The objective of the HFRPP is to outline the structure and content of a Human Factors Research Program which meets the Comission's 1987 Policy and Planning Guidance commitment "...to explore methods to better understand the
~
causes of human error and to reduce its incidence."
Research areas within the HFRPP include-human performance and human reliability, the human-machine interface, procedures, l
personnel qualification and training, and organization and J
management.
Consistent with the NRC's research philosophy, projects in these research areas include both short and long term tasks.
Short term tasks will support timely regulatory decisions and user needs.
Long term tasks will allow the NRC to anticipate human factors problems of potential safety significance.
The Appencix to Enclosure 1 sumarizes ongoing projects and work planned to respond to user requests and NAS recomendations in each research area.
Formal user requests from NRR, AE00, NMSS as well as within RES have identifie:i a wide scope of human factors research needs.
l The HFRPP responds to these requests. More detail will-be provided in periodic revisions to the plan and its appendix as i
coordination among the offices continues.
1 In addition to the user requests, a February 29, 1988, report by the NAS entitled Human Reactors Rescarch and Nuclear Safety provided recomendations for human factors research.
This 1
report made two types of recommendations. The first type was designed "...to facilitate the initiation, planning, management, conduct, and use of human factors research." The second type was to identify "... specific research topics to be investigated by the NRC and the rest of the nuclear comunity."
The staff's responses to the NAS recommendations are provided as Enclosure 2.
The HFRPP is responsive to most of the NAS recomendations.
NAS recomendations which were not acted upon, as discussed in Encle:ure 2, will be considered in future revisions to the HFRPP.
4 4
The Comissioners The enclosed HFRPP represents the best estimates of the RES staff for human factors research products for the near term (i.e.,1988-1990).
It also encompasses recomendations for longer term research to support anticipated needs.
The Office of Nuclear Reactor Regulation.
NRR initiated significant programs in the area of human factors following the accident at Three Mile Island.
Those programs addressed control room design, safety parameter display systems, emergency operating procedures, training of plant personnel and operator examinations and licensing.
The emphasis of NRR's human factors programs has been on licensing issues. As the programs addressing licensing issues near completion, that emphasis is gradually shifting to plant operations and maintenance issues.
To support this shift, NRR has become involved in the evaluation of performance and the assessment risk associated with human factors issues. A summary of NRR's human factors programs related to licensing and operations and management issues is provided as Enclosure 3.
The Office for Analysis and Evaluation of Operational Data.
Several AEOD programs identify and evaluate human factors concerns related to the safe operation of nuclear power plants.
The programs include an ongoing review of operating experience using licensee event reports (LERs) as well as Incident Investigation and Diagnostic Evaluation Team Programs.
Information from LERs is also monitored a0 part of the Performance Indicator Programs. AE00's Technical Trair.ing Center (TTC)supportsNRR'shumanfactorslactivitiesby training staff members involved in emerge procedure reviews and operator licensing.qcy operating In the future, as time is available, the TTC simula. tors may be used for human factors research projects supported by RES.
In addition, the AE00 evaluation of nonreactor events includes human factors considerations.
Specific studies in this regard have involved medical misadministrations, radiography, and loss of integrity of sealed sources.
Current studies include an assessment of both human performance and equipment reliability for large irradiator events.
A summary of AE0D's human factors programs is provided as.
The Office of Nuclear Material Safety and Safeguards. Up to 4
now, the NRC human factors programs have concentrated on power reactors.
However, NMSS is taking steps to develop human factors programs for NMSS licensees.
4 4
The Commissioners-,
One of the Study Groups established in support of the Strategic Plan evaluated how human factors should be considered in regulating nuclear materials.
The Group concluded that human factors were important to materials safety, and that, because of the large numbers of different types of licenses involved, a study was needed to prioritize human factors needs for each type of licensee. The Group suggested a "matrix approach" to correlate human factors with various types of licensees.
HMSS will pursue development of a human factors program and establish priorities in coordination with RES.
NMSS has hired a fulltime human factors specialist who will work closely with management in developing the program. Although it will take time to review the wide variety of materials licensees and human factor issues, some examples of priority areas are:
medical misadministrations, radiography safety, and contingency planning.
a
[
s s
/ f..
w ifetorStello,,
Executive Dire tar i
for Operations
Enclosures:
As stated l
DISTRIBUTION:
Commissioners i
OGC J
OI OIA GPA REGIONAL OFFICES j
I i
a l
M b
6 ENCLOSURE 1 1
4
\\
O e
t 9
\\;
fs
\\
e
i
\\
- HUMAN FACTORS hESEARCH PROGRAM PLAN i
/
\\
Prepared By The Office of Nuclear Regulatory Research United States Nuclear Regulatory Comission
/
f
,/
Mrs/f 1988
)
a l
,r
/
)(
e I
/
/
s
TABLE OF CONTENTS PAGE
1.0 INTRODUCTION
I 1.1 Human Factors.............................................
I 1.2 Need for Human Factors Research..................................
4 3
2.0 OVERVIEW 0F THE HUMAN FACTORS RESEARCH PR0 GRAM........................
7 3,
l 2.1 Objective of the Plan............................................
7 b
2.2 Planning and Coordination........................................
2.3 Prioritization...................................................
9 r
2.4 Update of the Plan..............................................
10 2.6 Research Areas..................................................
10 2.5.1 Human Performance and Human Reliability Assessment...
11 2.5.2 The Human-Machi ne In terface...........................
12 2.5.3 Procedures...............................'.............
13 2.5.4 Quali fications and Training...........................
14 2.5.5 Organization and Management...........................
14 2.6 O t h e r A c t i v i t i e s................................................
16 3.0 RESOURCE REQUIREMENTS.......{.........................................
16 3.1 S t a f f i n g.......................................................
17
s..
F
- - i i i 1..
l' TABLE OF CONTENTS, CONTINUED v.
- PAGE
')
{
h.,
3.2 Budget......................................,,4,,,,,,,,,,,,,,,,,,_17 3.3 Schedule.........................................................-17 4,0 REFERENCES..............................................,............
18 APPENDIX -- Summary of Human Factors Research Projects:
1 9
')
,)q th e.
3 t'
)
I a
F 2
P wf-
- yg f3m
[ k
'ygg m'
,-*g t
gp w.p-p-em---
g79-
--atM-p
'yg s4FT@T F W (W"4 & rWW Mt 'rW 1-'y'w
$fwe V"Th "M f 1 M7 % 7'vMw = F e w W tr $-t W -$y --4rg m ee-peerwy*Pewy-> % vyW
- v-iw te> y
@r=t
HUMAN FACTORS RESEARCH PROGRAM PLAN OFFICE OF NUCLEAR REGULATORY RESEARCH UNITED STATES NUCLEAR REGULATORY COMMISSION
1.0 INTRODUCTION
The fundamental mission of the Nuclear Regulatory Commission _(NRC) "...is to regulate those who commercially use or produce nuclear material so that the public health and safety, the common defense and security, and the environment are protected."I Both the NRC and the industry it regulates recognize that humans can favorably or adversely affect the performance of systems which process or use nuclear materials.
This recognition has driven efforts to improve safety by reducing "human error." For example, the Three Mile Island accident led to NRC requirements addressii:g nuclear power plant staffing and qualifications, emergency operating procedures, control room design, and the safety parameter display system.2 The nuclear industry developed and imple-mented plans to resolve concerns in these areas based on the NRC requirements.
1.1 Human Factors.
NRC and industry efforts to reduce human error fall within the scope of "human factors " Human factors is'a system-oriented, technical discipline that integrates knowledge from a number of other disciplines (e.g., engineering, psychology, physiology, health, and' education).
Implicit in the discipline's system orientation is a broad 6
.-n--.
, +..
definition of system.
This definition includes the hardware, software, and people who function to satisfy system goals (e.g., for a nuclear power plant it would include not only operators but maintainers, managers, emergency response team members, and others).
The definition of system used here also includes other functional elements such as procedures and personnel selection and training programs.
Human factors attempt, through a process of evaluation and improvement, to assure that the humans within a system can perform their tasks in a manner that satisfies system goals.
The process of human factors evaluation and improvement involves a number of functions.
Those functions are:
1.
Identification of the performance requirements placed upon humans by the systems of which they are a part 2.
Identification of human performance capabilities 3.
Evaluation of the influence various factors-(i.e., performance shaping factors) have on' those capabilities 4.
Evaluation of a system's human performance requirements against known human performance capabilities to identify human factors concerns (i.e., situations where human performance requirements may exceed human performance capabilities)
.5.
Review of the significance of these concerns in relation to system goals (e.g., safety, availability, cost)
6.
Development and verification of alternative approaches to resolve human factors concerns considered to be significant 7.
Comparison of the impacts of the alternative approaches on the full set of system goals 8.
Implementation of a preferred alternative and monitoring tc see its impact on achievement of system goals As implied by Function 8, the process of human factors evaluation and improvement is frequently iterative.
For example, the process can be used throughout-the life of a nuclear power plant and may be particularly useful when there are changes in plant hardware (e.g., installation of techno-logica11y' advanced equipment that promises significant improvement in plant performance).
Human factors concerns may be addressed by proposing changes either to reduce human performance requirements (e.g., automate equipment or improve the human-machine interface) or tn improve human performance capabilities (e.g.,
improve personnel selection and training).
Reduction of human performance requirements is often the more viable and potentially effective approach; however, improving human performance capability is appropriate for resolution of-some human factors concerns.
There may often be several viable proposals for resolving a particular human factors concern.
In such a case, the i
alternatives should be compared carefully in terms of such issues as:
i
4 <
1.
Effectiveness (immediate and long term)
'2.
Impact on other parts of the system 3.
Ease of implementation 4.
Cost (initial and continuing)
Effective resolution of a human factors concern frequently requires coordinated changes in several areas (e.g., improving the human-machine interface may be combined with changes to training and procedures) in order to assure that the full set of system goals is met.
1.2 fleed for Human Factors Research.
There have already been significant safety benefits from applying human factors principles to the nucicar industry.
However, human factors concerns continue to be identified by events i
documented in Licensee Event Reports.
While many of these events are not highly significant from a safety standpoint (e.g., a missed or late surveillance test of equipment), a number of them (e.g., those involving 3
losses of safety system function ) do illustrate that human factors concerns have the potential to affect the public health and safety adversely.
Continued protection of the public health and safety requires a process, such as that outlined in Section 1.1, to identify, evaluate, and resolve significant human factors concerns.
-9
- - +,,
e-,--
,7c y
n
-g
5 Other sources also point to the need for continued attention to human factors concerns.
For example, a December 1986 report by the National Academy of Sciences (NAS)4 recommends a "... transfer of advanced technology for improving human reliability to the nuclear industry."
Introduction of some advanced technologies may be inevitable because replacements for currently installed instruments and controls are becoming unavailable.
The unavailability of replacement parts is leading the industry to difficult decisions about instrument and control replacements or upgrades. Other factors (e.g., regulatory requirements such as the one for the safety parameter display system) may also be expected to result in the introduction of advanced technology into existing plants.
Regardless of the reason for introducing more advanced technology, adequate information about how that technology affects human performance requirements and capabilities is needed.
That information can be used to confirm success in cases where the reason for introducing more advanced technologies is to resolve human factors concerns.
In other cases, that information can be used to provide guidance to assure that introducing advanced technology does not result in new, safety significant human factors concerns.
5 A February 1988 NAS report makes two types of reconmendations related to human factors research and nuclear safety.
Recomrendations of the first type are designed "... to facilitate the initiation, planning, management, conduct, and use of human factors research."5 Recommendations of the second type identify "... specific research topics to be investigated by the NRC and the l
f
rest of the nuclear community."5 In sum, the NAS's recommendations call for a continued strong commitment by the NRC to support a broadly defined human factors research program.
Results of an Electric Power Research Institute (EPRI) project provide further support for continued attention to human factors at nuclear power plants.
The EPRI project involved review of a sample of the summary reports submitted to satisfy the NRC's detailed control room design review requirements. Specific 6
results support the NAS's 1986 conclusion about the introduction of more advanced technology into existing plants.
They also identify the need for research which confirms the effectiveness of some current approaches to resolving human factors concerns (e.g., panel layout enhancement) and the need for research in anticipation of potential human factors concerns (e.g.,
criteria for integrating new technology into existing control rooms).
)
i The discussion aLove suggests that both the NRC and the nuclear industry need j
to enhance their abilities to identify human factors. concerns, to evaluate those concerns for significance, and to resolve concerns which can adversely impact system performance and thereby threaten the public health and safety.
Enhancement of current abilities to identify, evaluate, and resolve human factors concerns requires advances in knowledge about human performance capabilities and about the requirements and constraints placed upon those capabilities by systems that use or produce nuclear materials.
To that end, the Office of Nuclear Regulatory Rosearch (RES) has developed this Human E
e-
- I
,J-y
--r-,
- - y
Factors Research Program Plan (HFRPP).
Recommendations from the reports cited above, as well as requests for human factors research from within the NRC contributed to HFRPP development.
2.0 0VERVIEW 0F THE HUMAN FACTORS RESEARCH PROGRAM PLAN The HFRPP addresses both near and long term human factors research needs.
Consistent with the NRC's research ph11osophy,7 near term tasks will support timely regulatory decisions by providing independent expertise.
Long term tasks will allow the NRC to anticipate problems of potential safety signifi-cance by providing new or expanded knowledge.
The HFRPP also encompasses work to resolve human factors generic issues and to develop regulatory positions (e.g., rules,regulatoryguides,etc.)
2.1 Objective of the Plan.
In its Policy and Planning Guidance for 1987, the Co,mmission commits "... to explore methods to better understand the causes of human error and to point to methods to reduce its incidence."1 The objective of this plan is to outline the structure and content of a human factors research program designed to meet that commitment.
The plan identifies the i
major areas within which NRC sponsored human factors research is expected to be conducted.
An appendix to the plan summarizes ongoing and planned research in each of the areas. The plan also addresses issues such as program planning and coordination, task prioritization, plan update, and resource requirements.
i
2.2 Planning and Coordination.
RES's Division of Reactor and Plant Systems (DRPS) is responsible for planning and coordinating the HFRPP.
DRPS will prepare detailed plans for meeting human factors research needs identified by NRC Offices.
Current or expected users include the Off. ice of Nuclear Reactor Regulation (NRR), the Office for Analysis and Evaluation of Operational Data (AE0D),
a Office of Nuclear Material Safety and Safeguards (NMSS), other divisions of RES, and the Regions.
Research requests from users will be reviewed and any necessary clarification obtained. Available research products and ongoing research by the NRC and others (e.g., EPRI, the Institute for Nuclear Power Operations (INP0)), will then be reviewed for applicability.
Additional research projects will be identified if ongoing research or the results of research already planned or completed can not satisfy the need.
HFRPP. activities will be coordinated both inside'and outside the NRC.
Initial coordination within the NRC resulted in statements of human factors research O
9 10 12 needs from NRR, AE0D, NMSS
, and RES.11 A draft plan for severe accident management research also identifies human factors research needs.
HFRPP activities will be coordinated with those of the Severe Accident Management Research Plan to assure that those needs are met.
Ongoing coordination within the NRC will help to assure that modified or new research needs are addressed when identified and that usable research products are developed and transferred to user Offices.
Coordination outside the NRC (e.g.. with EPRI, INP0, the Halden Research Project, other international agencies, vendors, utilities, owners groups, national laboratories, professional organizations, consultants, and
colleges / universities) will provide important information about current and future developments.
Proposed new systems and system modifications, human performance data, and programmatic issues related to the application of human factors technology to the production and use.of nuclear materials will be of particular interest.
DRPS will seek to develop coordinated or cooperative human factors research efforts with organizations outside the NRC in accordance with recommendations in the December 1986 and February 1988 NAS reports.4,5 Prior to initiating most research tasks, a review of the applicability of research from other industries to the nuclear power industry will be performed by appropriate experts who are independent of the organization that may conduct the research.
2.3 Prioritization.
Identified human factors research needs will be prioritized to assure that those most crucial to the NRC's mission are satisfied first and that other needs are satisfied in a timely manner.
The prioritization process will address such issues as:
1.
Does the research have significant potential for reducing risk associated with human perforr.ance at facilities which process or use nuclear materials?
2.
Is the proposed research technically feasible?
3.
Is the schedule for completing the research timely (i.e., will licensing decisions be made before the rese'rch is completed and can a
be applied)?
~
4.
Is the research directly related to' resolving. identified human factors generic issues? What is the priority of these issues (e.g.,
high, medium,orlow)?
5.
Is there a specific user need for the research?
6.
Does the research cover areas that NRC reviews as part of its regulatory responsibilities?
7.
What related work is ongoing at NRC or outside the-agency?
8.
Do the potential benefits justify the costs?
2.4 Update of the Plan.
This plan wil.1 be revised periodically.
DRPS will continue to seek formal statements of user needs, as those needs. become apparent.
Plan updates will reflect modified or new user needs as well as the resolution of any current needs resulting from research that has been completed.
\\
2.5 Research Areas.
The HFRPP encompasses five research areas.
They are:
i 1.
Human Performance and Human Reliability Assessment 2.
The Human-Machine Interface 3.
Procedures
______.___c__
- 11 4.
Qualifications and Training 8.
Organization and Management Research tasks in these areas will enhance performance of.the NRC mission by providing the analytical and experimental bases for guidance, recommendations, or requirements which work to reduce the likelihood of human errors that could adversely affect public health and safety.
The broad issues and safety concerns addressed by these research areas are discussed below.
The process of evaluating human factors concerns and recommending i.1provements to resolve these concerns inclu' des issues which transcend the five research areas described above.
For example, a hypothetical human factors concern might be resolved by improving either the human-machine interface or by.
additional training.
In such a case, the question is not what human-machine interface or training improvements can resolve the specific concern.
- Rather, it is which of the proposed improvements is a better resolution given system goals and constraints. A more typical example might involve resolution of a human factors concern through coordinated improvements in'several areas (e.g.,
human-machine interface, procedures, and training).
The question in this case is not how well any single approach works.
Rather, it is the appropriate integration of approaches.
2.5.1,
'formance and Hum 3n Reliability Assessment.
Human factors concerns are identified by comparing human performance requirements with human performance capabilities.
Research in this area will provide needed m
information about human capabilities and the impact of various performance shaping factors (e.g., high temperature and humidity) on those capabilities.
-Research in this area will also, through appropriate comparisons, identify human factors concerns.
Human reliability assessment may be seen as a specialized area of human factors research.
The NRC's human reliability assessment effort has two objectives.
The first is to develop a data base of human error rates and to support methods for integrating human reliability assessments and segments of reliability evaluations into probabilistic risk assessments.
The second is to develop tools which allow use of data generated by human reliability and probabilistic risk assessments to address unresolved and generic safety issues.
Human reliability assessment can also support a comparison of alternative approaches to resolving safety significant human performance concerns.
2.5.2 The Human-Machine Interface. The human-machine interface is the if nk between humans and system hardware and software.
Displays and controls are the most obvious elements of the human-machine interface, but cther hardware and software are included (e.g., communications equipment). A safety concern may result when elements of the human-machine interface do not support an adequate level of human performance.
For example, human error rates for safety-related tasks may be unacceptably high when vital information is difficult to find, difficult to read, difficult to interpret, or misleading.
9
13 -
Research tasks in the human-machine interface area will address a broad range of issues related to whether th'at interface supports an. adequate level of human performance.
Examples of important human-machine issues are:
1.
Compatibility with human sensory, perceptual, and cognitive capabilities (e.g., vision, audition, and information processing) 2.
Compatibility with human physical characteristics (e.g., height, reach,andstrength) 3.
Compatibility with human physiological characteristics and capabilities (e.g., heat tolerance, oxygen requirements)
Human-machine interface research will support comparison of human performance requirements against human performance capabilities.
Research will also serve to confirm the effectiveness of human-machine interface modifications designed to resolve significant human factors safety concerns.
2.5.3 Procedures.
Procedures are a means to assure that tasks are performed in a manner that satisfies system requirements.
They typically provide information about:
1.
Conditions indicating the need to perforn a task 2.
The sequence of task activities n.
. en - --
w
~
-e.
y n
w
- 14 3.
Any constraints (e.g., time limits, warnings, and cautions) 4.
Feedback indicating cne effect of task performance 5.
Conditions indicating the need to branch to other procedures 6.
Conditions indicating that the task is complete Procedures provide an appropriate resolution to some human factors concerns, but inadequate or poorly presented procedures can increase human error rates.
Research in this area will serve to identify the need to implement or improve procedures designed to resolve human factors concerns.
Research will also serve to confirm the effectiveness of proposed procedures or improvements to procedures, l
2.5.4 Qualifications and Training.
Human performance is related to the innate abilities and acquired knowledge and skills that are brought to a task.
Thus, one approach to resolving human factor: concerns is to select personnel based on their qualifications with yespect to performance of tasks that are important to system performance. Another approach is to enhance knowledge and skills through training to improve task performance.
The rationale for both these approaches is that human performance requirements are more likely to be.
met by greater human performance capabilities.
The approaches are not mutually exclusive and,'in fact, are often used together.
Research in this
.n,
.=
I area will serve to identify the need to implement or improve qualifications and training of personnel who are parts of systems which use or produce nuclear materials.
Research will also serve to confirm the effectiveness of proposals to implement or modify qual.ifications and training requirements.
2.5.5 Organization and Management.
Human performance can be affected by a number of factors related to organizational design (e.g., staff size and composition, span of control) and management practices and decisions (e.g.,
shift scheduling, equipment selection).
Changes in organization or management can affect either human performance require'ments or human performance capabilities.
For example, changes in staff size or composition can affect individual workloads (human performance requirements) while changes in shift length can affect such factors as fatigue (human performance capabilities).
Research in this area will address the relationship between organization and management and human performance.
It will also serve to confirm the effectiveness of proposed changes to organization and management for resolving human factors concerns.
Research on the influence, of organization and management on human error rates will also provide tools by which plant safety performance can be monitored.
This rese' arch will support the development of programmatic performance indicators.
G
l 2.6 Other Activities. Conduct of the HFRPP will involve related activities which are not, themselves, classified as research.
For example, an NAS 0
report recommends that the NRC develop "... a bibliographic system for NRC-supported human factors reports." Activities of this type will be planned and conducted to the extent that they are necessary to assure the HFRPP's support of the NRC's mission.
3.0 RESOURCE REQUIREMENTS As part of its responsibility for planning and coordinating the HFRPP, DRPS will prepare recommendations to the Director, RES for the resources needed (i.e., staff and budget).
3.1 Staffing. Management of the overall program and of individual human factors research tasks is the responsibility of the Reliability and Human Factors Branch (RHFB), DRPS.
Because of the broad scope of human factors, the RHFB staff includes professionals with training and experience in several technical disciplines (e.g., psychology, engineering, computer systems, industrial organization).
Human factors research tasks will be carried out by national laboratories, colleges and universities, and private indust,ry.
Qualified sources will have experienced human factors staffs with multi-disciplinary capabilities (e.g., engineering, operations, and psychology).
e
.m.
v...
3.2. Budget.
Budget projections for the human factors research program for fiscal years (FYs) 1988-1992 are shown below:
Human Factors and Human Reliability Research Budget ($1,000s)
FY 1988 FY 1989 FY 1990 FY 1991 FY 1992 Human Performance and 1,365 1,705 Human Reliability Details Assessment of Human-Machine Interface 1,051 1,394 FY 1990 - 1992 Procedures 665 1,124 budgets Qualifications and 488 695 Training to be Organization and 661 1,169 determined lianagement TOTALS 4,230 6,086 7,550 6,650*
6,650*
- Budget decrease in FY91 and FY 92 is based on the assumption that currently identified human factors generic issues will be completed and no additional ones will be identified and performance indicator work will reduce.
3.3 Schedule.
Schedules for research tasks are documented and tracked in the Research Project Management Infora tion System.
Schedules for research related to generic issues and the resolution of those issues are documented and tracked in the Generic Issue Management Control System.
~
4.0 REFERENCES
1.
United States Nuclear Regulatory Comission.
NUREG-0885,' Issue 6, U.S. Nuclear Regulatory Commission Policy and Planning Guidance, 19-87.
Washington, D.C. 20555, September, 1987, 7
2.
United States Nuclear Regulatory Commission.
NUREG-0737, Clarification of TMI Action Plan Requirements. Washington, D.C.
20555, November 1980; Supplement 1, December,1982.
3.
United States Nuclear Regulatory Commission. AE0D/C505, Case Study Report on Loss of Safety System Function Events. Washington, D.C.
20555, December, 1985.
L 4.
National Research Council'.
Revitalizing Nuclear Safety Research.
National Academy Press, Washington, DC.
December, 1986.
5.
National Research Council. Human Factors Research and Nuclear Safety. National Academy Press, Washington, DC, February, 1988.
6.
Electric Power Research Institute.
NP-5795, Control Room Deficiencies, Remedial Options, and Future Human Factors Research Needs.
Electric Power Research Institute, Palo Alto, CA. In press.
i,
7.
Memorandum from E. S. Beckford to RES staff.
Subject:
"Nuclear Regulatory Research Philosophy," dated May 4, 1987.
8.
Memorandum from T. E. Murley to E. S. Beckjord..
Subject:
"Human Factors.ResearchProgram,"datedNoYember9,1987.
9..
Memorandum from E. L. Jordan to B. W. Sheron.
Subject:
"Human Factors Research and AE0D Needs," dated September 4, 1987.
- 10. Memorandum from H. L. Thompson to E. S. Beckjord.
Subject:
"Human Factors Research Program," dated March 23, 1988.
- 11. Memorandum from B. M. Morris to 8. W. Sheron.
Subject:
"Request for Research Needs in the Area of Human Factors and Human-Reliability,"
dated November 9, 1987.
12.
Office of Nuclear Regulatory Research, Division of Reactor and Plant Safety, Reliability and Human Factors Branch.
Draft Severe Accident Management Research Plan, dated November 3, 1987.
J
t
' l HUMAN FACTORS RESEARCH PROGRAM PLAN APPENDIX 1
I
SUMMARY
OF HUMAN FACTORS RESEARCH PROJECTS i
e
.1 1
m,ww.v,,N,,,,,,,,g,,m,-,,mwn..,
-,- e,m ?.n en --- -.,,,,
,.ne,
,.w'--w.,e,ne~,,,ne,,w,,
,n.
,,mw,,.,,,w-w,--+
upwgrg,---,7 e ~ -w w.e
, -r n, -
,~,,w,.
)
- A 11 -
TABLE OF CONTENTS PAGE
1.0 INTRODUCTION
.................................................... Al 2.0 SUMMARIES OF ONG0ING AND PLANNED RESEARCH'.......................
A2 2.1 Human Performance and Human Reliability Assessment......... A2-2.1.1 Ongoing Projects................................. A3 2.1.1.1 Acquisition of Human Performance Data.. A3 2.1.1.2 Data Base on Human Failure Rates....... A3 2.1.1.3 Cognitive Model to Predict Intended Human Action During Emergencies.............. A3 I
2.1.1.4 Integrate Human Reliability Assessments I
into the Probabilistic Risk Assessments A4 2.1.1.5 Maintenance Personnel Performance Simulator.............................. A4 2.1.1.6 Criteria for Safety-Related Operator e
Actions................................ A5 2.1.2 Planned Research................................. A6 2.1.2.1 Operator Performance Under Extreme l
Co n d i t i o n s............................. A 6 mer,,e,--,
---w--
g e--
1-me,,e m--
e.,~,
n g
y,
- A-111 -
PAGE 2.1.2.2 Severe Accident Management --
Habitability........................... A6 2.1.2.3 Performance Indicators for Cognitive Errors................................. A6 2.1.2.4.
Human Error Data Bank (Probabilistic Risk Assessment Methods)............... A7 2.1.2.5 Other Planned Research in the Area of Human P, r fo, aance and Human Reliability Assess. tert............................. A7 2.2 The Human-Machine Interfs e................................ A7 2.2.1 Ongoing Projecis................................. A8 2.2.1.1 Local Control Stations................. A8 2.2.1.2 Centrol Room Design Standard........... A8 2.2.1.3 Improved Control Rocm Instrumentation.. A9 2.2.1.4 Review Criteria for Human Factors Aspects of Advanced Controls and In s trumen ta tion........................ A10 2.2.2 Planned Research................................. All 2.2.2.1 Computer Classification................ All i
i l
4
/-
- A iv -
PAGE 2.2.2.2 Impact of Over Reliance on the Safety Parameter Display System (Co'nfirmatory Research Using Simulators)............. A12 2.2.2.3 Expert System Verification'and Validation Methodology............................ A12 2.2.2.4 Sealed Source Devices.................. A12 2.2.2.5 Improvements to Annunciator Systems.... A13 2.2.2.6 Impact of High Technology on Control Room Operators......................... A13 2.2.2.7 Other Planned Research in the Arealof the Human-Machine Interface............ A13 2.3 Procedures................................................. A14 2.3.1 C geing Projects................................. A14 2.3.1.1 Guidelines for Upgrading Other Procedures............................. A14 2.3.2 Planned Research................................. A14 2.3.2.1 Presentation of Emergency Operating Procedures (Confirmatory Research Using S i mu l a to rs )............................ A 14 2.3.2.2 Chernobyl Follow-Up -- Procedure Violctions............................. A15
+
4
-Av-PAGE
+
2.3.2.3 Materials and Fuel Cycle Procedures.... A15 2.3.2.4 Other Planned Research-in the Area of Procedures............................. A15 2.4 Quali fica tions and Traini ng................................ A16 2.4.1 Ongoing Projects................................. A16 2.4.2 Planned Research................................. A16 2.4.2.1 Training Effectiveness................. A16 2.4.2.2 Personnel Quali fications............... A17 2.4.2.3 Training for Rare.and Difficult Events................................. A17 2.4.2.4 Long Term Project on Training.......... A18 i
2.4.2.5 Qualifications and Training of l
Materials and Fuel Cycle Licensees..... A18 2.4.2.6 Other Planned Research in the Area of Quali fications.and Training............ A18 2.5 Organization and Management................................ A19 2.5.1 Ongoing Projects................................. A19 2.5.1.1 Shift Staffing......................... A19
, t
.A vi -
PAGE 2.5.1.2 Management /0rganization Influence on Huma n Erro r Ra te....................... A29 e
2.5.1.3 Management Impact.on Plant Safety...... A2]
~ -e 2.5.1.4 Programmatic Performance Indicators....-A21 2.5.2 Pl a n ned Re s ea rc h................................. A' -
c 2.5.2.1 Shift Scheduling and Overtime.......... A22 2.5.2.2 Shift Length (Confirmatory Research~
i Using Simulators)...................... A22 2.5.2.3 Operating Staff Size and-Composition (Confirmatory Research Using Simulators)............................ A23 2.5.2.4 Team Performance....................... A23 2.5.2.5 Shift Technical Advisor Role and Function............................... A23 2.5.2.6 Performance Based Regulation A24 2.5.2.7 Other Planned Research in the Arca of Organ.ization and Management............ A24 2.6 Other Activities........................................... A25 2.6.1 Ongoing Projects................................. A:5 4
2.6.1.1 Event Reporting........................ A25 6
t 7,,-,
...,,gw..,,,,-
l t
(v
. yc
('.
A vii.-
.i PAGE-t 2.6.2 Planned Research
...............................,..'d25 9
2.6.2.1 Human Factors and Materials Licensing ~'..'A25 2.6.2.2 Bibliographic System for NRC-Supporked Human Factors Reports.................. A26
3.0 REFERENCES
...................................................... A27 1
i
/<
. I, I'
4 6
[
,s'
'(
I 1-
\\
.l i
/.f I;
s
\\;
7' l
,j; i
4
,')
l' gb'
(
1 t
-T.
\\'(
s, 1
4 /-
'l
. f.
/
./
1 i
t t
)
j r/
J A
+
/
j i
2 4
. i i
t J
J_,__._
'3
- Al -
l i
SUMMARY
OF HUMAN FACTORS RESEARCH PROJECTS 1
1.0 INTRODUCTION
This appendix provides brief summaries of the ongoing projects and pianned research in cach of the five areas of the Human Factors Research Program Plan (HFRPP). Additional-project descriptions will be included in updates to this sa d
plan as they are developed.
This appendix also provides information on other ongoing and planned activities to support the HFRPP.
Topics discussed under planned research include needs identified by the Office of Nuclear Re.'ctor Research1 (NRR), the Office of Nuclear Material Safety and 2
3 Safeguards (NMSS), th'e Office for Analysis and Evaluation of Operational Data (AE00), and divisions of the Office of Nuclear Regulatory Research,5 (RES) as 4
well as recommendations by the National Academy of Sciences 6(NxS).
These needs and recommendations are not yet addressed by ongoing research projects i
but are being reviewed and prioritized.
The review and prioritization process I
is intended to develop an integrated approach for addressing human fac+ars concerns and, thereby, to optimize the use of research resources.
For example, identified needs within a single research area may be consolidated under one research project.
In other cases, review of a particular need ray indicate that the objectives of the requested research are unclear.
In such a,, /o case, RES will work with the office identifying the need to clarify it to a point where useful research is viable.
Because of limited resources, development of viable research projects to address every identified need may not be possible.
Review and prioritization of the needs may also indicate that some are not sufficiently related to safety to warrant further Nuclear Regulatory Commission (NRC) sponsored research.
- A2 -
j In addition to the programs summarized below, the HFRPP is expected to. benefit fron NRC participation in the Halden Research Project.
The NRC is particu-icely interested in Halden's efforts in the following areas:
h I,
1.
Computer-Based Procedures i
-/
2.
Verification and Validation of Software
/
3.
Disturbance Analysis Systems
/
4.
Human Factors Aspects of Advanced Instrumentation and Controls t
5.
Operator Training Methods Results of Halden's tasks in these areas are expected to provide useful information in the areas of nuclear powe,r plant human-machine interfaces, i
4 procedures, and qualifications and training.
2.0 SUMMARIES OF ONGOING AND PLANNED RESEARCH.
2.1 Human Performance and Human Reliability Assessnent.
Research projects in this area will identify basic human performance ca, abilities and the impact of performance shaping factors on those capabt; = ~ies.
e
- A3 -
Research in this area will also identify situations in which human performance requirements exceed human performance capabilities.
Finally, this area includes activities to estimate human error rates.
2.1.1 Ongoing Projects in the Area of Human Performance and Human Reliability Assessment.
2.1.1.1 Acquisition of Human Performance Data.
The goal of this project is to collect data for use in estimating human-error rates.
Data will be obtained from the commercial nuclear power industry and from other sources.
Data from sources outside the i'ndustry will be used when industry data are sparse or unobtainable and will serve primarily as "anchors" for estimating human error rates in nuclear power plants.
2.1.1.2 Data Base on Human Failure Rates.
The goal of this project is to develop a comprehensive data base system for storing and processing human error and hardware failure rate data.
This system, known as the Nuclear Computer Library for Assessing Reactor Reliability, will be the primary repository for data that were collected as part of AE0D's Integrated Risk Assessnent Data Acquisition Program.
2.1.1.3 Cognitive Model to Predict Intended Human Action During Emergencies.
The goal of this project is to develop and evaluate better methods for nedeling the cognitive behavior of nuclear power plant personnel in situations involving plant safety.
The project will concentrate on an artificial intelligence model of how utility personnel (including supervisors and
~
- A4 -
managers) form intentions to act in emergencies.
The success of this proaect will result in a model capable of providing input to human reliability and probabilistic risk assessments.
2.1.1.4 Integrate Human Reliability Assessments into Probabilistic Risk Assessments. The goal of this project is to integrate human reliability assessments into the probabilistic risk assessment process. This integration will result in more realistic assessments of the human's overall contribution to nuclear power plant safety.
The project will also permit improved analysis of data related to unresolved and generic safety issues.
2.1.1.5 Maintenance Personnel Performance Simulator.
Human error during maintenance can be a major contributor to nuclear power plant safety concerns.
7 For example, a 1984 staff study determined that maintenance and testing activities were involved in 70% of those events that resulted because actions were performed on equi l ment in the wrong unit (of multi-unit sites) or wrong p
train (within~a single unit). A more recent Licensee Event Report Compilation 0 (October 1987) showed that 20 of the 50 operational events involving human factors concerns were related to maintenance or surveillance.
In "esponse to this concern, the NRC sponsored development and validation of a computer model for evaluating maintenance practices and procedures at nuclear power plants.
The Maintenance Personnel Performance Simulation (MAPPS) model is capable of simulating corrective and preventive maintenance task trformed under both normal and abnomal plant conditions.
MAPPS can be used to i
- A5 -
evaluate existing maintenance practices and procedures as well as proposed improvements to those practices and procedures.
The current goals of this project are to:
1.
Collect data on the transferability and usefulness of MAPPS as a diagnostic and design tool to enhance maintenance practices'and.
procedures 2.
Analyze data from participating plants to identify common tasks which yield abnonnally high success and failure probabilities, and, to the degree possible, determine the causal factors a'ssociated with those abnormally high probabilities 2.1.1.6 Criteria for Safety-Related Operator Actions.
Current nuclear power plant designs require operator action in response to certain transients and 9
accidents.
NUREG-0471 identified a concern about whether operators would, under all conditions, have sufficient time to perform the actions required of them during transients and accidents.
Generic Safety Issue B-17 responds to this concern by supporting development of time criteria for safety-related operator actions. These criteria will allow one to determine whether operators have sufficient time to perform actions required of them during transients and accidents or whether automatic actuations are required.
l
- A6 -
2.1.2 Planned Research -in the Area of Human Performance and Human Reliability Assessment.
2.1.2.1 Operator Performance Under Extreme Conditions.
There is little information about human task performance under extreme conditions that might reasonably occur during or after certain plausible events (e.g., high temperaturc and humidity, reduced lighting, and conflicting control panel l
indications during station blackout).
NRR has requested research to obtain information on human cognitive and physical performance under extreme conditions.
2.1.2.2 Severe Accident Management -- Habitability.
Severe accidents might require nuclear power plant personnel to perform tasks under extreme 5
environmental conditions.
The Severe Accident Management Research Plan identifies the need for research to determine what environmental conditions, are to be expected during severe accidents and what protective measures will be required for personnel to perform under those conditions.
Research on this topic and on "Operator Performance Under Extreme Conditions" (2.1.2.1 above) are expected to be merged.
2.1.2.3 Performance Indicators for Cognitive Errors.
This project responds to an AE0D reouest for a performance indicator for cognitive error.
Research on this topic is expected to be based on results of the project to develop a i
"Cognitive Model to Predict Intended Human Action During Emergencies" (2.1.1.3above).
t b
.,--,.-c-.
m,,
~e-,-
- A7 -
2.1.2.4 Human Error Data Bank (Probabilistic Risk Assessment Methods).
This project responds to a request from RES's Division of Regulatory Applications.4 Plans to address this request are expected to be formulated as part of efforts to develop a "Data Base on Human Failure Rates" (2.1.1.2 above) and to "Integrate Human Reliability Assessments into Probabilistic Risk Assessments" (2.1.1.4above).
2.1.2.5 Other Planned Research in the Area of Human Performance and Human Reliability Assessment.
Research on two additional human performance and 1
human reliability assessment topics has been requested.
The topics are:
1.
Operator Performance Under the Stress of Emergency Operations 2.
Advanced Control Rooms -- Evaluation of Operator Performance in Existing Control Rooms 3.
Human factors in the transportation of spent fuel and other Type B packages Summaries of the proposed work will be provided following coordination between RES and the requesting offices to clarify the objectives of the requested research.
2.2..ae Human-Machine Interface.
Research projects in this area will support comparison of human performance requirements against human performance capabilities.
It will also serve to confirm the effectiveness of human-machine interface modifications designed to resolve significant human factors safety concerns.
- A8 -
2.2.1 Ongoing Projects in the Area of the Human-Machine Interface.
2.2.1.1 Local Control Stations.
Detailed control room design reviews at nuclear power plants have identified many displays and controls in the control room that exhibit human engineering discrepancies (i.e., features which are not consistent with accepted human factors principles).
These discrepancies indicate a poor human-machine interface, and they are being resolved when assessment indicates that they can contribute to safety significant operator errors.
Documented (e.g., D' avis Besse
, Rancho Seco11) and undocumented reviews of 10 local control stations (i.e., stations outside the control room) at nuclear power plants indicate that some displays and controls at these stations also suffer from human engineering discrepancies.
Generic Issue HF 5.1 addresses this concern.
Its resolution involves efforts to determine and quantify the risk to public health and safety resulting from poor human-machine interfaces at nuclear power plant local control stations.
Results are expected to establish the magnitude of the risk involved, the potential for risk reduction, and tho need for regulatory action to resolve human factors concerns at local control stations.
2.2.1.2 Control Room Design Standard.
This project responds to item (now 12 Generic Safety Issue) I.D.4 of NUREG-0660 The planned resolution of the item was NRC development of a regulatory guide based on evaluation of industry control room design standards.
RES review indicates that all operating plants and current applicants are conducting detailed control roorr design ' reviews in
- A9 -
12 13 response to NUREG-0660 item I.D.1.
_NUREG-0700 and acceptable alternatives are being used as control room design standards for this effort.
Applications for future light water reactors shall include, per 10 CFR 50.34(g)14, an 15 evaluation of the facility against the Standard Review Plan Section 18.1 13 of that document addresses control room design and references NUREG-0700 as appropriate guidance for control room design.
Generic Safety Issue I.D.4 has been resolved by industry compliance with NUREG-0660 item I.D.1 and 10 CFR 50.34(g)14 Therefore, the generic issue 12 has been closed.
Standards for incorporating advanced technologies into existing ~ control roons'and for advanced control rooms will be addressed by other projects within the HFRPP.
2.2.1.3 Improved Control Room Instrumentation.
Generic Issue HF 5.2 was originally conceived as a broad issue addressing:
1.
Computers and Computer Displays 2.
Evaluation of Operator Aid Systems 3.
]
j 4.
Safety Status Indication and Applications of Automation and a
Artificial Intelligence j
e i
1 f
I
- A10 -
A review of this issue resulted in a decision to retain only the activities related to annunciators under the generic issue.
The three other activities are being evaluated under a project entitled "Review Criteria for Human Factors Aspects of Advanced Control and Instrumentation" (See 2.2.1.4 below).
Research to resolve Generic Issue HF 5.2 will evaluate human factors concerns related to annunciators in nuclear power plant control rooms.
The first step will be to determine the risk to public health and safety resulting from human error in the use of information from control board annunciators through the use of probabilistic risk assessnent.
Results will establish the magnitude of the risk involved and the potential for risk reduction through annunciator system improvement. A subsequent value/ impact analysis will compare the cost of potential annunciator system improvements with potential risk reductions and will serve as an aid in determining whether regulatory actions are needed to address human factors concerns With annunciator systems.
If regulatory actions appear necessary, additional research may be needed (e.g.,
a program to evaluate alarm reduction techniques).
Results from other research (e.g.,
i Halden's work on advanced alarm systems or the Electric Power Research Institute's work on annunciator systems) may also be used in developing appropriate regulatory actions.
2.2.1.4 Review Criteria for Human Factors Aspects of Advanced Controls and Instrumentation..The initial effort on this project will identify specific human factors concerns related to "high technology" modifications of existing 1
2
. - All -
control rooms.
Subsequent efforts will prioritize any safety concerns which are identified and, as appropriate, develop regulatory actions to resolve the Concerns.
As part of the initial effort, the NRC is sponsoring a survey of the commercial nuclear industry's current and planned use of artificial intelligence, expert systems, and digital computers.
The survey will identify existing and proposed uses of these high technology systems and, for proposed systems, the schedule for their implementation. The survey will also address the proposed use of such high technology systems in the advanced control room designs being developed by nuclear steam supply system vendors. As noted above, safety concerns identified by the surveys will be prioritized.
Resources will then be allocated to resolving those concerns based on their priority.
2.2.2 Planned Research in the Area of the Human-Machine Interface.
2.2.2.1 Computer Classification.
This project will respond to an NRR user 1
need.
Its goal is to develop guidance and criteria for qualification of the various safety and non-safety applications of digital computer systems.
Computers are being used increasingly at nuclear power plants to provide l
information to operators.
In many cases the quality of the computer system is not appropriate to the way the information is being used.
The guidance u..d criteria developed by this project will apply to the design, test, verification and validation, maintenance, and reliability of computer systems.
t e
E
-A12 2.2.2.2 Impact of Over Reliance on the Safety Parameter Display System (Confirmatory Research Using Simulators).
This project will respond to an NRR 1
user need.
Staff evaluation of safety parameter display systems (SPDSs) implemented in nuclear power plants indicates that there is potential for over reliance on the SPDS during transients and accidents.
There is also potential for underuse of the SPDS and for use by inappropriate personnel.
This research project will examine the impact of SPDS misuse on safety.
2.2.2.3 Expert System Verification and Validation Methodology.
This project l
6 will respond to an NRR user need and to an NAS recommendation.
Its goal is to develop acceptance criteria and guidelines for the evaluation of expert systems. Artificial intelligence and expert systems are being introduced into nuclear power plants.
These computer programs cannot be verified and validated using methods now available.
This project will provide tools for evaluating expert system verification and validation methods.
4 1
2 2.2.2.4 Sealed Source Devices.
NMSS has requested that safety concerns involving sealed source devices be studied.
The use of sealed sources in l
radiographic devices leads to a substantial fraction of documented overexposures, and tile use of sealed sources in large irradiators has the potential for creating significant problems if mishandled.
Safety concerns
]-
with sealed sources have been traditionally linked to inadequate human-machine interfaces.
This project will identify, evaluate, and propose resolution of human-machine interface concerns with devices using sealed sources.
i r
- A13 f
2.2.2.5 Improvements to Annunciator Systems.
This project will respon to an d
l NRR request.- This request is expected to be addressed by the project on "Review Criteria for Human Factors Aspects of Advanced Controls and Instrumentation"(2.2.1.3above).
2.2.2.6 Impact of High Technology on Control Room Operators. Coordination.
1 between RES and NRR related to.a request for this research is expected to identify the need to address potential high technology modifications to existing control rooms other than artificial intelligence, expert systems and digital computers.
The research approach for these modifications is expected to be similar to that used in the project to develop "Review Criteria for Human Factors Aspects of Advanced Controls and Instrumentation" (See 2.2.1.4above).
2.2.2.7 Other Planned Research in the Area of the Human-Machine Interface.
Research on two additional human-machine interface topics has been requested.I' 4 The topics are:
1.
Advanced Control Rooms -- Design Standard 2.
Advanced Reactors -- Human-Machine Interface Summaries of the proposed work will be provided following coordination between RES and the requesting office to clarify.the need.
i i
l
- A14 -
2.3 Procedures.
Research in this area will serve to identify the need to implement or improve procedures designed to resolve human factors concerns.
Research will also serve to confirm the effectiveness of newly proposed procedures or improved procedures.
2.3.1 Ongoing Projects in the Area of Procedures.
2.3.1.1 Guidelines for Upgrading Other Procedures.
The goal of this project is to resolve Generic Issue HF 4.4 through development of technical guidelines for upgrade of normal and abnormal operating procedures similar to those 16 developed for emergency operating procedures.
Results of one study conducted under this program indicate a number of ways in which normal and abnormal procedures in nuclear power plants are discrepant from accepted human factors principles.
The NRC is currently funding a value/inpact assessment to determine which improvements to current industry programs for developing and implementing operating procedures will result in meaningful reduction in risk to public health and safety.
2.3.2 Planned Research in the Area of Procedures, i
i 2.3.2.1 Presentation of Energency Operating Procedures (Confirmatory l
Research Using Simulators). This project will respond'to an NRR request.
Operators and.others in the nuclear power industry have expressed concern about the size and complexity of Emergency Operating Procedures (E0Ps).
Most
R
- A15 -
E0Ps are written in text format. Others are written in flow / chart or a combination of text and flowchart format.
Computerized E0Ps may be developed
~
in the near future.
Little is known regarding the safety implications involved in the different methods of presenting E0Ps. -This research project will address the safety impact of issues related to presentation of E0Ps.
2.3.2.2 Chernobyl Follow-up -- Procedure Violations.
Procedure viciations at nuclear power plants are committed by licensed and auxiliary operators, plant technicians, maintenance personnel, and contractors. While the NRC believes that procedure violations are infrequent and rarely committed intentionally, the exact nature and extent of these violations and their consequences is unknown. This project will determine the extent and nature of procedure violations in nuclear power plants.
2.3.2.3 Materials and Fuel Cycle Procedures.
Holders of materials and fuel cycle licenses are required to have and to follow procedures for all operations associated with nuclear materials.
There has been no systematic 2
review of these procedures for human factors concerns.
NMSS has requested such an effort.
The project will identify, evaluate, and propose resolutions to safe'.y significant problems with procedures used by naterials and fuel cycle licensees.
2.3.2.4 Other Planned Research in the Area of Procedures.
Research on one
+
additional procedures topic has been requested.# That topic is:
Advanced Reactors -- Accident Management Procedures.
( -
,.-~,---*.,...-w,
,r
,y-rn--,-,re
,c w,
I A16 -
o A summary of the proposed work will be provided following coordination between RES and the requesting office to clarify the need.
i 2.4.
Qualifications and Training.
Research in this area will serve to identify the need to implement or imprc,ve aualification and training resolutions for human factors concerns.
Research will also serve to confirm the effectiveness of proposals to implement or modify qualification and training requirements.
2.4.1 Ongoing Projects in the Area of Oualifications and Training.
There are no ongoing research projects in this area.
2.4.2 P_lanned Research.in the Area of Qualifications and Training.
I 1
2.4.2.1 Training Effectiveness.
This project responds to an NRR request and an NAS recommendation.6 The objective of the research will be to develop guidelines for evaluating the adequacy of the various methods and devices used currently by the nuclear industry for acquiring and maintaining operational skill.
Although the industry has committed to the Institute for Nuclear Power Operation's (INPO) accreditation program, it has not thus far ' emonstrated d
that there is a correlation between individual training and performance.
This may be ciue to inconsistencies in the degree to which individual utilities 4
- A17 -
adhere to the INP0 guidelines. This project will explore the degree of success of the current industry efforts to improve training.
It will build on work done in both nuclear and non-nuclear areas to develop methods and criteria for measuring the effectiveness of the various training methods.
4 2.4.2.2 Personnel Qualifications.
This project responds to an NAS recommendation.6 Regulatory Guide 1.8, "Personnel Selection and Training,"
endorses ANSI Standard N18.1-1971. Together these documents provide holders of nuclear power plant licenses with guidance on acceptable personnel qualifications. The Regulatory Guide has not been updated with respect to qualifications of personnel other than licensed operators.
However, errors by other personnel can affect safety.
For example, maintenance personnel perform a large number of safety-related tasks of various levels of difficulty, at widely varying time intervals, and under stressful environmental conditions.
This project will identify the need for development of qualifications criteria for personnel other than operators at nuclear power plants.
2.4.2.3 Training for Rare and Difficult Events.
This project responds to an NAS recommendation.6 The issue to be addressed is whether the operator and other utility personnel are trained adequately to deal with rare and difficult events.
The project will consist of two phases.
Phase 1 will address the engineering skills required for dealing with the occurrence of initiating
. events and safety system unavailabilities.
Phase 2 will address the personality traits required for dealing with rare and difficult events, such as cognitive action, communication, and decision making.
F e
9
- A18 -
2.4.2.4 Long Term Program on Training.
This project responds to an fAS recommendation.6 It will develop a systematic framework for defining the goals of a long term NRC research project on training.
The experience gained from research on "Training Effectiveness", and on "Training for Rare and Difficult Events" (see 2.4.2.1 and 2.4.2.3 above) will assist in formulation of the plan for a long term effort.
Several questions will be addressed (e.g., what topics should be explored by the NRC, how joint NRC/ industry effort can be achieved, how related research and implementation efforts can be integrated).
Results this effort will aid in determining the scope of any long term research program on training by the NRC.
2.4.2.5 Qualifications and Training of Materials and Fuel Cycle Licensees.
This project responds to an NMSS request.
Qualifications and training are especially important for fuel cycle licensees, nuclear medicine licensees, radiographers, and irradiators.
The project will identify and evaluate concerns related to the qualifications and training of such personnel.
Regulatory activities to resolve safety significant concerns will be proposed.
2.4.2.6 Other Planned Research in the Area of Qualifications and Training.
Research on two additional qualifications and training topics has been requested by user offices or recommended by the NAS.6 They are:
5 1.
Severe Accident Management -- Training 2.
Degree Requirements for Senior Reactor Operators
- A19 -
Summaries of any work proposed on these topics will be provided followina coordination between RES and other offices to establish or clarify user needs.
2.5.
Organization and Management.
Research in this area will address the re at onship between organization and l
i management and human performance to resolve human factors concerns.
It will serve to confirm the effectiveness of proposed management and organizatio3 changes and to provide tools by which plant safety performance can be monitored.
l 2.5.1 Ongoing Projects in the Area of Organization and Management.
2.5.1.1 Shift Staffing. The goal of this project was to respond to Generic Issue HF 1.1 by determining the ap'propriate composition and minimum size of a shift crew for operation of a nuclear power plant. This goel was met by the 14 "Licensed Operator Staffing Rule" which is now part of 10 CFR 50.54 Two tasks remain.
They are revision of Regulatory Guide 1.114 "Guidance to j
j Operators at the Controls and to Senior Operators in the Control Room of a 10 Nuclear Power Unit"17 and of Standard Review Plan Section 13.1.2 "Operating Organization" to conform to the Licensed Operator Staffing Rule.
l i
J l
f f
- A20 -
I 2.5.1.2 Management / Organization Influence on Human Error Rate.
The goal of this project is to develop methods to identify and evaluate the influences of supervisory and management policies, practices, and behavior on the safety performance of nuclear power plant personnel.
Tasks to be undertaken in support of this goal include the following:
a.
Develop an organization model of.the site-specific operational units (operators, maintenance technicians, instrument and control technicians, etc.) that can impact nuclear power plant reliability.
Indicate the relationship between these operational units and other site-specificunits(e.g., security),corporateunits(e.g.,
purchasing) and non-utility units (e.g., contractors).
b.
Use the organization model to identify supervisory and management positions which may affect performance reliability, i
c.
Refine the organization model to pinpoint supervisory and manago.ent positions which may affect performance reliability.
J j
l 4
I j
d.
Develop methodologies which can be used to evaluate supervisory and
]
management factors affecting performance reliability, 1
i 1
I e.
Develop probability estimates of human reliability perfomance as a function of identified management and organizational factors.
o.
- A21 -
2.5.1.3 Management Impact on Plant Safety.
This project is based on the "Management /0rganization Influence on Human Error Rates" project (see 2.5.1.2 above) and has the same basic objectives.
Data from the "'ianagement/
Organization Influence on Human Error Rates" project will be used to develop a set of performance indicators associated with management actions that could affect plant safety.
Results of these projects will provide a method to determine trends in management effectiveness.
Such a method could be useful to the current performance indicator program and will complement planned research on "Programmatic Performance Indicators" (see 2.5.1.4 below)
Conversely, data collected by the program can be used to validate results of this program and the program on "Management /0rganization Influence on Human Error Rates."
2.5.1.4 Programmatic Performance Indicators.
The performance indicators provide an objective view of nuclear power plant operational performance and enhance the NRC's ability to recognize changes in plant safety.
This new project on Programatic Performance Indicators will enhance the NRC's ability to forecast poor or dec'iining performance, particularly with respect to human factors.
The goal is to develop indicators of the effectiveness of all plant programs affecting safety.
Initial activities will identify indicators of maintenance and training effectiveness, and determine the most appropriate methods to validate these indicators.
Long tem efforts will include the development of indicators of both the thoroughness and effectiveness of plant programs such as maintenance, operations, training, radiation control, quality I
assurance, and management.'
- A22 -
't 2.5.2 Planned Research in the Area of Organization and Management.
2.5.2.1 Shift Scheduling ar.d Overtime.
This project will respond to an NRR I
user need and to the Commission's direction to the staff to explore development of a performance indicator associated with overtime practices at Nuclear Power Plants.
Recent occurrences of operators sleeping on shift have identified the need to evaluate the safety significance of long work hours on 18 the'part of the operator. This project will build on previous NRR work and will provide a quantitative basis for addressing nuclear power plant operator performance problems resulting from fatigue due to overtime and shift scheduling practices in nuclear power plants.
NRC policies are based on non-nuclear experience and expert judgement.
This project will provide a basis for evaluating and updating the current policies based on overtine and shift scheduling at nuclear power plants.
l 2.5.2.2 Shift length (Confinnatory Research Using Simulators).
This project l
0 will respond to an NRR user need and to an NAS recommendation.
The research will focus on ways in which the state of knowledge outside the nuclear arena, especially in the area of vigilance, can be a'pplied to nuclear power plant j
operations.
As the NAS recommends, it will explore biological, social
'd i
environmental based solutions as opposed to hardware solutions.
Issues such as boredom, work load, social interactions, and sleep-wake cycle will be considered, i
- A23 -
2.5.2.3 Operating Staff Size and Composition (Confirmatory Research Using 1
Simulators).
This project will respond to an NRR user need.
Other personnel in addition to licensed operators (e.g., auxiliary operators, instrument and control technicians, radiation and chemistry technicians, health physics personnel, and maintenance personnel) play major roles during transients and accidents and during abnormal operations.
Errors by any of these personnel may result in unnecessary challenges to safety systems or to significant safety concerns.
The contribution of these personnel to risk has not been systematically reviewed. This project will address the contribution to risk of errors committed by personnel other than operators.
1 2.5.2.4 Team Performance. This project responds to an NRR user need. The NRC presently evaluates a licensee's team performance during reviews such as emergency operating procedure audits or inspections, training audits or inspections, requalification and initial examinations and sometimes during incident reviews. However, adequate evaluation criteria for team performance are presently not available.
The objective of this project is to develop such criteria.
The scope of the work includes: identifying team skills, developing team evaluation criteria, simulator and field testing of criteria, and correlating measures of team performance with measures of safety.
2.5.2.5 Shift Technical Advisor Role and Function. A program addressing the 0
Shift Technical kdvisor may be developed in response to an NAS recommendation.
A pending Commission decision on degree requirements for Senior Reactor I
i m
~.
, - - - -,. _,. _., - ~ -. -. - _ -. - - -. _.
- A24 -
c Operators may affect the role of Shift Technical Advisers.
Thus, the staff will evaluate the need for such a program after the Commission has decided the issue of degree requirements for Senior Reactor Operators.
2.5.2.6 Performance Based Regulation.
This project will respond to 19 6
Commission guidance and an NAS recommendation. The objective will be to assist implementation of the Coma.ission's intent "to shift its regulatory emphasis away from detailed requirements toward a more general, performance-based requirements."
2.5.2.7 Other Planned Research in the Area of Organization and Management.
Research on several additional organization and management topics has been requested by user offices, 5 or recommended by the NAS.
They are:
2 6
1.
Severe Accident Management -- Organization 2.
Organization and Management Influence on the Safety Performance'of Materials and Fuel Cycle Licensees 3.
Organizational Design and a Culture of Reliability 4.
Operational Decision Making 5.
Time'.y Recognition of Emergencies
]
6.
Characteristics of Managers Summaries of any work proposed on these topics will be provided following coordination between RES and other offices to establish or clarify user needs.
f
..,_,__....c
~,_,-._m,_m.,,_..,__ _,-..- --._.-~_. _ _. ~.
- A25 -
)
2.6 OTHER ACTIVITIES i
t Conduct of the HFRPP will involve related activities which are not, themselves, classified as research.
Such activities will be planned _and j
conducted to the extent that they are necessary to assure the HFRPP's support of the NRC's mission.
2.6.1 Ongoing Projects in Other Activities.
W 1
2.6.1.1 Event Reporting.
This project responds to an NRR request.
Present reporting systems identify human factors concerns which contribute to reportable events, but they often do not provide sufficiently detailed causal information.
This project is developing a structured protocol that can be used by NRC personnel to identify human factors concerns which were at the root of, or contributed to, reportable events.
i 2.6.2 Planned Research in Other Activities.
t 2.6.2.1 Human Factors and Materials Licensing. A Study Group supporting development of the NRC's Strategic Plan has concluded that human factors is an l
l important concern in naterials licensing.
That group also produced a matrix 20 of "best-judgments" about the most effective means for resolving human factors concerns related to the broad spectrum of license ~es in question.
NMSS has requested a detailed follow-up of the Study Group's activities related to 1
i
- A26 -
t human factors.
The objective of those activities is "... to develop a good agenda with suitable priorities to obtain more effective treatment of human factors in materials licensing."
i 2.6.2.2 Bibliographic System for NRC-Supported Human Factors Reports.
The O
NAS recommended that "mechanisms to' improve the dissemination of human
~
factors results throughout the industry be developed." As a first step, the panel recommended "... development of a bibliographic system for NRC-supported human factors reports."
In response, RES initiated a project to:
(1) identify the desired characteristics of a human factors state of knowledge system; (2) determine which of those characteristics can be satisfied by human i
factors information systems currently available to RES through the NRC Library, Public Document Room and Department of Energy Laboratories; (3) develop a specification for computer softwar'e to download information from currently available systems and to accommodate information from sources not
- j available to those systems, and (4) do a cost analysis for developing a computer software package responding to the characteristics identified under (1) and taking advantage of existing information system resources determined unde'r (2).
O G
- A27 -
3.0 REFERENCES
1.
Memorandum from T. E. Murley to E. S. Beckjord.
Subject:
"Human-Factors Research Program." dated November 9,1987.
%/
2.
flemorandum from H. L. Thompson to E. S. Beckjord.
Subject:
"Human Factors Research Program," dated March 23, 1988.
/
3.
Memorandum from E. L. Jordan to B. W. Sheron.
Subject:
"Human Factors Research and AE0D Needs," dated September 4,1987.
l 4.
Memorandum from B. M. Morris to B. W. Sheron.
Subject:
"Request-e for Research Needs in the Area of Human Factors and Human Reliability," dated November 9, 1987.
t 5.
Office of Nuclear Regulatory Research, Division of Reactor and Plant Safety Reliability and Human Factors Branch.
Draft Severe Accicert 4
Management Research Plan, dated November 3,1987.
4 6.
National Research Council. Human Factors Research and Nuclear Safety.
flational Academy Press, Washington, DC, February,1980.
i 7.
United States Nuclear Regulatory Conmission. AEOD/S401, Human Error in Events Involving Wrong Unit or Wrong Train. Washington, D.C. 2 CSS 5, i
j January. 1964.
b
^
y, (t t,.
A
_f_
ft
- A28 -
t tt c
.I :
1 3
' [
i i
\\
8.
United States Nuclear Regulatory Comission.
NUREG/CR-2000, Volume 6, Number 10, licensee Event Report (LER) Compilation for Month of
/ -
I t
October 1987. Washington, D.C. 20555, Noveinber,1987.
l
(
y
/'
/
i 9.
Uhited States Nuclear Regulatory Comission.
NUREG,0471, Generic Task Problem Descriptions (Categories 8, C, 0).
J Washington, D.C. 20555, June,1978.
j j j;
', g
,1 t
s'S 10.
United States Nuclear Regulatory Comission.
NUREG-1154, Loss of s
s v,
g1 Main and Auxiliary Feedwater Event'at the Davis-Besse Plant on June 9, 1985. Washington,,0 C'. 2'0555, July, 1985.
.a gi l
j s
. v 11.
United States Nuclear Regulatory Comission. (NUREG-1195, Loss g t
3 j.
Integrated Control System Power and Overcoo H y Trati/ tent at_ Rancho
/
i 26, 1985. p' Washington, D.C. 20555, February,1986.
- s Seco on December
/
s 3
e<
\\
12.
United States Nuclear Regulatory (Comission.. NUREG-0660, NRC Ac*. ion _PlanDevelopedasaResultofMSTb'l-2 Accident, t
Washington, D.C. 20555, May, 1980.
h i
4
.g 13.
Unit td States Nuclear Regulatory Comissio1. NUREG-0700, GuidelinesforControlRoomDeshnReviews. Washington,kC.20555,
,h 4
~
September, 1981.
i f
1 i
s 1
h p
1 Es g
- A29 -
- );
if 14.
Office of the Federal Register, National Archives and Records o
Administration.
Title 10 Code of Federal Regulations.
U.S.
f Government Printing Office, Washington, D.C. 20402, January,1987, a
15.
United States Nuclear Regulatory Commission.
NUREG-0800, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants. Washington, D.C. (1st Edition) November '.975, (2nd Edition) March 1980,~(3rd Edition) July 1981, 20555, September [.
1981.
16.
Battelle Human Affairs Research Center, Pacific Northwest Laboratory.
NUREG/CR-3968, Study of Operat'ing Procedures in Nuclear Power Plants:
Practices and Problems.- U.S. Nuclear Regulatory Commission, Washington, D.C. 20555, February, 1987.
17.
United States Nuclear Regulatory Commission, degulatory (aide 1.114, Revision 1, "Guidance on Being Operator at the Controls of a Nuclear Power Plant." Washington, D.C. 20555,' November, 1976.
18.
United States Nuclear Regulatory Commission. NUREG/CR-4248, Recommendations for NRC Policy on Shift Scheduling and Overtime and at Nuclear Power Plants.
Washington, D.C. 20555, July, 1985.
19.
United States Nuclear Regulatory Commission.
NUREG-C885, Issue 6 U.S. Nuclear Regulatory Commission Policy and Planning Guidance, 1987.
Washington, D.C. 20555, September, 1987.
,r v e
-n-
-,e-
,v.e e s 4--
a
- h30 -
=
l
- 20. Memorandum from R. M. Bernero to J. M. Taylor.
Subject:
"Study Group I Report," dated February 1, 1988.
e i
l
4 l
ENCLOSURE 2 l
l 1
STAFF USE OF THE NAS RECOMMENDATIONS FOR HUMAN FACTORS RESEARCH The purpose of this enclosure is to describe to the Commission the staff's use of National Academy of Sciences (NAS) study panel recommendations for human factors research.
The NAS panel, comprised.of experts in the areas of the behavioral and social sciences (human factors), nuclear power operations and nuclear physics, was brought together in response to a grant from the Commission (ref: SECY-85-353 dated November 6, 1985, and Commission concurrence dated December 19,1985).
The panel's charge was to identify study areas in current and recent human factors programs that may have received inadequate attention and to provide guidance for human factors research.that could enhance the safe operation of nn;1 ear power plants.
In February 1988, the study panel published its findings ana recommendations in a report titled, "Fuman Factors Research and Nuclear Safety." The report makes two types of recommendations:
(1) Recommendations 1 through 6 to.... "facilitate the initiation, planning, management, conduct and use of human factors research," and (2) Recommendation 7..."on specific research topics to be investigated by the NRC and the rest of the nuclear community" (p. 2).
The remainder of this enclosure presents the NAS recommendations, as they pertain to the NRC, and the staff's responses.
Additionally, NUMARC, EPRI and INFO were requested to review the NAS report for comments on its recommendations and the topics they may wish to pursue alone or through cooperative research with the NRC.
In summary, the NAS report recommends that the NRC facilitate its capability for carrying out h'uman factors research by:
(1) staffing and maintaining continuity of the program at the branch level, (2) adopting a systems-oriented or socio-technical perspective to the research, (S) uti.lizing independent peer 9
~
NAS Response 2
reviews to enhance the quality of research products, and (4) establishing improved mechanisms for transferring research results to user comunities by means of annual written reviews and a ' bibliographic search system.
Additionally, the report recomends that the staff conduct human factors research in the areas of improved methods and criteria for assessing the adequacy of:
(1) computer-based systems for processing and displaying plant status information including those which involve expert systems technologies; (2) personnel training, qualifications, staffing and shift scheduling; (3) individual, team, management and organizational performance, and (4) the regulatory environtrent.
In reviewing each of the NAS panel's recomendations, the staff determined whether or not NRC regulatory user needs corroborate the recomendation.
This enclosure describes the ongoing or planned research, or other means for addressing the recomendations, e.g., solicit the involvement of industry.
For the most part NRC had already ongoing or planned research that addresses the NAS's recommendations.
In some areas NRC participation in the Halden project will provide information that addresses a number of NAS recomendations.
This is especially true in the area of Human-System Interface Design.
As is shown in Table 1 and subsequent sections of this enclosure, the staff is implementing, or is planning to imp'xment a majority of the recommendations.
The column headed "Status" indicates whether the recomendation or the research is: "Ongoing", "Planned", or "No action planned at this time".
"0ngoing" denotes either that the policy question raised is answered by the current program or that the research question identified is being addressed as part of current initiatives.
"Planned" denotes that the recomended research is be planned but not yet underway.
"No action planned at this time" denotes that the program either partially addresses the recomendation or the question will be considered us the program is periodically updated in view of related research.
In the pages following Table 1 each of the seven NAS recommendations for undertaking human factors research is followed by the staff's rarrative response.
Where appropriate, NAS recomenaations are broken out into topic areas correspond'ing with the entries in Table 1.
i
NA3 Response 3
Table 1 Summary of NAS Recommendations and NRC Staff Responses Recommendation Status
- 1. Comitment to human factors research
- a. Staffing / financial Ongoing support
- b. Qualified manager Ongoing
- c. Branch level program Ongoing
- 2. Systems-oriented Ongoing approach
- 3. Peer reviews
- a. Research planning Ongoing
- b. Research reporting Ongoing-
- c. Plant access No action planned at this time.
j
- d. National research facility No action planned at this time.
- 4. Continuity Ongoing
- 5. Transfer of knowledge No action planned at this time.
- 6. Dissemination of results Ongoing 8
,_,-,-y-
- - -- ~- e o-
-%s
NAS Response 4
4 Table 1 (cont'd)
Summary of NAS Recorrendations and NRC Staff Responses Recommendation
-Status
- 7. Research agenda
- a. Man-machine interface design (1)Ccmputer-basedsystems Human-system inter-faces Planned Safety parameter display systems Planned Disturbance analysis s;-3tems Ongoing Function allocation Ongoing Operating procedures Ongoing Expert systems Ongoing (2) Software development Software psychology No action planned at this time.
Guidelines for soft-ware development Ongoing
- b. Personnel sub-system' (1) Training Effectiveness Ongoing New approaches Planned l
e
NAS Response 5
Table 1 (cont'd)
Summary of NAS Recommendations and NRC Staff Responses Recommendation Status (2) 0ualifications Degree requirement Ongoing Licensing exams Ongoing Tester qualifications Ongoing Team licensing Ongoing Licensing non-control room staff Planned (3) Staffing Shift Tech Advisor No action planned at this time..
Screening / Selection No action planned at this time.
Shift Scheduling Ongoing Vigilance Ongoing
- c. Human performance (1) Error mechanisms Ongoing (1) Computer models Computer simulation Ongoing Operatiors, maintenance and organizations Ongoing Use of non-nuclear research Ongoing Theoretical and empirically based Ongoing (3) Industry participation Ongoing (4) Operational experience data No action planned at this time.
d
NAS Response 6
j Table 1 (cont'd)
Sumary of NAS Recomendations and NRC Staff Responses Recomendation Status
- d. Management / organization (1) Impact of regulations No action planned at this time.
(2) Cesign and culture Ongoing (3) Decision making Ongoing (4) Emergencies Ongoing (5) Characteristics Ongoing
- e. Regulatory environment (1) Models of regulation No action planned at this time.
(2) Performance indicators Ongoing NAS Recomendation 1: Comitment to Human Factors Research "The NAS panel recomends that the NRC make a firm public comitment to applied behavioral and social science (human factors safety) research.
This would require a decision to increase staffing and financial support.
Without such a comitment, the public and the nuclear industry can reasonably assume that human factors is not regarded by the NRC to be a matter of importance, nor will the NRC be able to attract the highly competent staff it will need.
in addition, the human factors program should be directed by a qualified human factors specialist at the levei of branch head, not as a subdivision of the j
reliability branch.. Finally, the branch should be staffed by....a -team of multidisciplinary scientists."
NAS Response 7
NRC Staff Response The NRC made a comitment to human factors safety research in its April 12,1987, reorganization which reestablished a human factors capability within RES's Division of Reactor and Plant Systems / Reliability and Human Factors Branch (DRPS/RHFB).
In addition, the Comission's Policy and Planning Guidance (NUREG-0885, Issue 6, September 1987) directed RES "...to develop a progran for Comission review in the area of human factors / performance consistent with agency needs." That plan is presented in the previous enclosure.
It provides the framework for a strong, stable, and sustained program in human factors research.
Staffing and financial support have already increased. Human factors staffing in RES is currently four-times what it was prior to the April 1987 reorganization.
The staff consists of a multidisciplinary team of eight professionals whose backgrounds include academic training and experience in psychology, human factors engineering, industrial organization, operations research, computer systems / control systems, mechanical engineering, and nuclear engineering.
The staff possesses a total of four Ph.Ds and ten Masters Degrees, with each of the areas above covered with at least one Masters Degreed person.
The staff has a cobbined total of over 170 years of professional experience with more than 100 staff years devoted to human factors.
The branch manager is qualified both by education in operations research and organizational perfonnance and by experience in directing NRC programs that evaluated the human factor.
Current funding for human factors research is also above previous levels, that is, from a peak of about 2.2% of RES's total budget in FY 1983 -- to about 3%
of the total RES budget in FY 1988.
The Five Year Plan comitnent for funding i
human factors research is currently more than $4M annually.
j HAS Recomendation 2:
Adoptina.a Systems-Oriented Approach "The NAS panel recommends that the NRC's research program maintain a broad
NAS Response 8
perspective"... "Further, the panel firmly believes that research that recognizes a systems approach, in which the ' system' is broadly defined, has great potential ~for delivering results that yield useful recommendations for safety improvement....by a systems approach, we mean a way of looking at a nuclear power plant not as composed of components whose properties can be examined in isolation, but rather as a collection of components including human ccmponents, each of whose properties affects and is affected by the other dynamically from moment to moment, so that to predict the performance of any component requires that one consider the state of, in general, many others."
NRC Staff Response This NAS recommendation is consistent with the approach outlined in the Human Factors Research Program Plan (previous enclosure) because the totality of human factors can be grouped several ways.
General correspondence between the NAS topic areas and the topic areas of the Human Factors Research Program Plan is shown below:
NAS Topic Area NRC Human Factors Research Program Recommendations Topic Areas Human-System Interface Design - Human-Machine Interface
- Procedures Personnel Subsystem ----------- Qualifications and Training Human Performance ------------- Human Performance and Human Reliability Assessment Management and Organization --- Organization and Management Regulatory Environment -------- Organization and Management in the area of programmatic performance indicators Within the NRC's researt'1 program plan there are other topic areas that approach human factors systematically because these other topic areas transcend the topic areas listed above in both columns.
Specifically, research l
into maintenance, performance indicators, and accident management transcend l
l l
1 J
NAS Response 9
both individus) human factors tapics and hardware.
This program approaches safety research fully cognizant of the interactive nature of the individual topics and the fact that they are part of a collective entity with its own emergent. properties.
!!AS Recontendation 3:
Peer Review and Enharced Access to Nuclear Power Research Facilities and i)ersonnel "The panel reconnends that the NRC involve a diverse group of knowledgeable researchers in planning, conducting, and evaluating its research program.
In addition, peer review of proposals and draft reports by behavioral science experts is needed to ensure the quality of sponsored research.
One of the barriers to effective human factors research has to do with providing behavioral science researchers access to realistic settings, to facilities such as simulators, and to people such as experienced operators.
While the panel recognizes the practical difficulties involved, we strongly urge the NRC and the nuclear industry to take significant steps that enhance researchers' access to these facilities and people. One step to achieve this goal would be to create a national research facility for the study of human factors in nuclear power systems."
NRC Staff Response.
The staff appreciates the usefulness of peer reviews at scrre stages of regulatory research.
Human factors research currently benefits tron frequent informal and formal inputs from diverse sources.
Those sources use the mechanisms listed belcw to identify research needs, to redirect, and to evaluate research projects /results against those needs.
Staff and management reviews within RES using RES in-house expertise in the behavioral sciences and other relevant disciplines Reviews of RES projects and recontrendations by personnel with diverse disciplines from other hRC offices N
NAS Respo.nse 10 Technical reviews'of project proposals, draft reports and final reports prior to and during conduct of the work, using outside consultants Reviews by the ACRS, including the ACRS Subcomittee on Human Factors Reviews of research program areas by an independent advisory comittee called the Nuclear Safety Research Review Comittee.
This comittee is made up of cutside experts and reports to the Director, RES.
Publication of research results in refereed journals We consider access to realistic settings and experienced personnel to be important to the success of human factors research.
Currently, we do not have an ongoing or planned research activity directed soiely at achieving access to plant personnel during operations.
However, the work on Maintenance Personnel Performance Simulation (MAPPS), involves utilit'y personnel to assure that useful products are' developed.
Regarding a national research facility for human factors, RES currently provides funding to the Halden Research Project in Norway.
Halden has appropriate facilities and a multidisciplinary staff capable of doing human factors-related research.
Projects to address NRC human factors research needs, especially in the area of advanced displays and expert systems, are being developed with Halden.
Other means of jointly funded and cooperative research are being explored with EPRI.
Because of EPRI's access to nuclear facilities and personnel, joint NRC-EPRI research projects might serve in lieu of research done at a national research facility.
Similar coordination of research projects may also be explored with INPO.
Therefore, the staff has no current plans to establish a national research facility solely for the study of human factors.
NAS Recomendation 4:
Continuity in the Research Program The NAS panel recomrends that..."a NRC human factors research program operate 1
NAS Response H
coherently for an extended period rather than change in response to each new, incediate, external demand.
Since effective research is cumulative, continuity is as important as level of expenditure."
NRC Staff Response Tne staff is revitalizing the human factors program in RES.
It has developed a plan to carryout this program (see previous enclosure).
The research budget for human factors, as projected in the Five Year Plan, will remain stable.
Further, NRR, AE00, and NMSS also supported the need for continuity of human factors research through their user need letters.
Finally, the Human Factors Research Program Plan presented in the previous enclosure emphasizes the need for coordination between the RES.and user offices to assure existing research needs are satisfied and new needs are identified in an orderly manner.
The 1
plan will be updated periodically to reflect revised and new user needs as well as the resolution of current needs resulting from completed research.
NAS Recomrnendation 5:
Transfer of Knowledge l
"The panel recourends that the NRC take the greatest possible advantage of existing research in the behavioral and social sciences.by increasing the transfer of knowledge to the nuclear industry.
To this end, the panel recomends that the NRC publish an annual review of the human factors research relevant to the nuclear power industry."
NRC Staff Response The staff understands that transferring knowledge gained from research to the utilities is necessary before safety can be enhanced.
Results of research are made available to the utilities and the public through publication of NUREG reports, technical articles, and participation in international conferer.ces.
RES has, for a nurrber of years, presented the results of its research in sessions of the annual NRC Water Reactor Satety Information Peeting.
Water
NAS Response 12 Reactor Safety Meetings are widely attended by the industry and other e
interested parties.
Some projects, such as the one to develop MAPPS, cchieve direct transfer of knowledge to participating personnel from the nuclear utilities.
The agency also publishes NUREG-0304, Regulatory and Technical Reports Abstracts Index Journal.
This document is published quarterly and annually, and contains abstracts of all NRC published reports including those on human factors.
The feasibility and advisability of publishing additional reviews will continue to be examined with industry groups such as NUMARC, EPRI and INP0.
NAS Recommendation 6:
Dissemination of Human Factors Research by the Nuclear Industry The NAS panel recommends that "mechanisms to improve the dissemination of human factors results throughout the industry be developed"... by the NRC and industry.
"One element is to use or develop a bibliographical search service"
... of relevant human factors publications by other government agencies, DOE laboratories, industry and academia.
"As a first step the panel recommends the development of a bibliographic system for NRC-supported human factors reports."
NRC Staff Response Dissemination of human factors reseatch results has the potential to be important to continued improvements in nuclear safety.
As described in our response to Reconmendation 5, above, RES disseminates the results of its research projects publicly.
Additionally, the staff has initiated a project to explore the possibility of enhancing the dissemination of hunan factors research results.
That project will: (1) identify the desired characteristics of a human factors state of knowledge system; (2) determine which of those characteristics can be satisfied by human factors information systems currently available to RES through the NRC Library, Public Document Rocms and DOE Laboratories; (3)- develop a specification for computer software to download information from currently available systems and to accommodate information from sources not available to those systems, and (4) do a cost analysis for
NAS Response 13 developing a computer software package responding to the characteristics identified under (1) and taki'ng advantage of existing information system resourcesdeterminedunder(2).
RES is also coordinating with EPRI on the dissemination of its related research projects.
NAS Recomendation 7:
A Human Factors Research Agenda The NAS panel recommends research in five major areas (man-machine interface, personnel subsystem, human performance, organization and management, regulatory environment).
It followed three criteria
..."in detennining higher priority topics within the areas."
First, research issues having a critical impact on safety and thus must be addressed imediately.
Second, research issues whose resolution will lead to better evaluations of plant safety.
Third, short-term research issues whose resolution will serve as essential building blocks for longer-term research.
The NAS panel states that..."in all cases research should be aimed at management, maintenance, and other ancillary workers, as well as control room operators."
- a. Human-System Interf ace Design i
1.
Computer-Based Information and Display Systems NAS Recomendation Regarding Human-System Interfaces The NAS panel recomends research on methods and criteria for assessing improved human-system performance and for evaluating the adequacy of new human-system interfaces and decision aids.
Developing these rr.easuring techniques is the highest imediate priority.
NRC Staff Response RES is planning a research activity to address reccmendations on both the methods and criteria issues.
The key elements of this activity are:
(1) review the pertinent literature to identify existing human-computer interface
NAS Response 14 models which may be relevant, (2) conduct a workshop to evaluate each model and select one to servs as an integrating framework for methods and criteria development,and(3)developandtestcandidatemethodsandcriteriafor assessing human-systems performance and the adequacy of new human-system interface and decision aids.
NAS Reccanend'ation Regarding Safety Parameter Display Systems The NAS panel recommends research on improved methods and criteria for evaluating the adequacy of existing and proposed Safety Parameter Display System (SPDS) designs.
NRC Staff Response Useful experiences with computer-based interfaces resulting from the SPDS have been summarized in NUREG-1286, Supplement Number 1, March 1988, SER Related to the Restart of Rancho Seco Nuclear Generating Statiosi, Unit 1.
That experience is being used to guide ongoing and planned research directed toward methods and criteria for evaluating human-computer interfaces.
2.
Automation and Computer-Based Human Performance Aids NAS Recommendation Regarding Disturbance Analysis Systems The NAS panel recommends research on disturbance analysis systems as a means of providing useful diagnostic ano emergency management information to human problem solvers in nuclear power plants.
NRC Staff Response RES is currently supporting research on disturbance analysis systems as a participating member of the OECD Halden Reactor Project.
One study in FY 1988 involves an evaluation of a rule-based expert system for diagnosing faults in a nuclear power plant.
Knowledge gained frem this stuoy will be
NAS Response 15 useful to the NRC in formulating guidelines to the industry ori expert systems.
The NRC, however, does not do developmental research on new systems.
This is the rightful responsibility of industry.
Additionally, RES is planning a research activity whose objective is to establish criteria for assessing the adequacy of human factors aspects of advanced controls and instrumentation.
The existing human engineering guidelines for nuclear power plant control rooms are ostensibly for today's controls, displays, and information concepts.
While these guidelines may be adequate for the current generation of plants, they may be inadequate for advanced designs and for major revisions to existing control rooms.
Major revisions to control rooms will include digital computers as well as new technologies such as artificial intelligence and computer graphics. This study will evaluate current practices and technologies available, and will develop a regulatory position on such systems.
NAS Recorreendation Regarding Function Allocation The NAS panel reconinends research on better methods and criteria for allocating functions between the human and automated systems in nuclear pcwer plants.
More specifically, it recommends research on the merits ano methods of dynamic function allocation between the human and system.
Also, the NAS p'anel recorreends research to determine the appropriate role of the human in future plants, especially with regard to higher-level decisionmaking and supervisory functions.
l hRC Staff Response RES is currently funding research as part of the OECD Halden Project to investigate the potential role of integrated surveillance and control systems f or control roo.ns.
One of the goals of this study is to evaluate a fully computerized control room.
The staff anticipates that a better understanding of human execution of a function (e.g., manual control) versus automated functions (e.g., automated centrol system) will errerge from this Halder work.
NAS Response 16 These results may also prove useful in evaluating the period of time after a reactor trip before a need for operator action is ma'ndatory.
NAS Recomendation Regarding Operating Procedures The NAS panel recomends research on operating procedures because they are inherently a part of the human-system interface.
The report j
I discusses the issues associated with computer-based Emergency Operation Procedures (EOPs).
It states:
"at present, there is no coherent theory for the design of E0Ps, and this requires research to develop such a theory."
NRC Staff Response The NRC is supporting research in the use of computer-based procedures as a participating member of the OECD Halden Reactor Project.
One of the ongoing studies at Halden is a controlled evaluation of a computerized procedure l
system.
The objective of this research is to identify ano assess problems with present day manual procedure systems.
Some of these problems are the very l
issues discussed in the NAS report.
The current project at Halden will j
l l
l evaluate operator acceptance for the functions performed and the man-machine interface of the computerized procedure system being evaluated.
As a participating member of the Halden Project, the NRC will help direct and will receive the results of these tests.
These results will be used to help NRC evaluate computerized procedure systems and what further research is needed in this area.
NAS Recomendation Regarding Expert Systems The NAS panel recomends a review and evaluation of literature on expert systems and artificial intelligence simulation of cognition.
The nature of validation as applied to such systems should also be examined.
Research should concentrate on the validation and assessment of such systems, leaving their development to industry.
NAS Response 17
\\
NRC Response The NRC is a participating member of the OECD Halden Project which is conducting several studies on computer-based man-machine interfaces, including expert systems, for use in nuclear power plants.
These studies involve control room operators to conduct man-in-the-loop tests of these interfaces.
The NRC staff anticipates that the knowledge obtained from these studies will help to establish methods to specify and evaluate new interfaces and decision aids.
3.
Human Factors in Sof tware Development NAS Recommendation Regarding Software Psychology The NAS panel recomends a review of the literature on software psychology. The purpose of the review is to identify potential applications of such a psychology in the development of nuclear power plant ccrrputer software programs.
NRC Staff Response The NRC regulates the design and use of computer software that performs safety-related functions.
One example of this type of software is the computer program in a computer-based reactor trip system.
In special circumstances, the NRC also regulates software that perfo ms non-safety related fur.ctions.
One example of this type of software is tr.e ccmputer program for the Safety Parameter Display System (SPDS).
However, the scope of regulation for non-safety-related applications is much less than for safety-related applications.
NRC does not regulate the softe re in the plant process computer.
Because of other high priority issues ad limited resources, the staff does not plan to begin research on software psychology.
The staff feels that this topic could be more effectively addressed by the industry since the industry designs and develops computer software.
We will pursue with industry (e.g., EPRI) the need to conduct this research.
i 1
I
NAS Response 18 NAS Recommendation Regarding Human Factors Guidelines in Software
. Development The NAS panel recommends research on guidelines for incorporating human factors concerns into the design of computer software systems.
NRC Staff Response At the present time RES does not have an ongoing or planned research activity in this area since there are publications that address this issue.
For example, EPRI NP-3701, Computer-Generated Display System Guidelines, Vol.
1: Display Design, and Vol. 2, Developing an Evaluation Plan, is one example of human factors guidelines for computer-based displays.
Another example is ESD-TR-86-278, "Guidelines for Designing User Interface Software," by Sidney L.
Smith and Jane N. Moser of MITRE.
This latter report is for the United States Air Force.
It contains 944 guidelines for designing software to support the user interface to computer-based information systems.
The staff believes these and similar publications provide some guidance to designers for incorporating human factors concerns into software development.
b.
Personnel Subsystem NAS Recommendation Regarding Training Effectiveness and New Training Approaches The NAS report recognizes that the availability of properly trained and qualified utility persennel is essential for effective and efficient job performance. Also, it acknowledges that the NPC, as well as the industry, has made advances tcward improving training and qualifications of nuclear power plant personnel.
It concludes, however, that additional research is needed on a number of topics.
Specifically the NAS panel recommends:
- 1. Research on the effectiveness of existing training techniques and tools for acquiring and maintaining operator skills.
Q r
y
---e e
e
NAS Response 19
- 2. Research on alternate approaches to training utilizing current advancements of training techniques such as "exploration training," "embedded training for rare and difficult events," and "problem-solving training.
More specifically, research is recomended to determine their usefulness and adaptability to a nuclear pcwer plant context, especially for managing abnormal events in the complex social environment with many people involved and sharing responsibilities for managing rare events.
- 3. Longitudinal research, in conjunction with the industry, un all aspects of training as a central element of the human factors research program. This research should focus on the types of skills required in the nuclear power plant context and on the systems approach to training pursued in the industry accreditation program.
NRC Staff Response Regarding Recomendation 1, the staff is initiating research curing FY 1988 to develop guidelines for evaluating the adequacy of various training techniques and devices currently used by the industry for acquiring and maintaining operator skills.
Although the industry is comitted to INP0's accreditation program, thus far it has not demonstrated that there is a correlation between individual training techniques and performance.
This may be due to inconsistencies in the degree to which individual utilities adhere to the INP0 guidelines.
The study being initiated by RES, will explore the degree of success of the current industry efforts to improve training. This in itself will be difficult since no objective measures of performance in the context of nuclear power plants have been established.
This difficulty notwithstanding, there is considerable activity on training technologies dealing with the "what" and "how" to measure training effectiveness. The Journal of Human Factors Society in its special
~
issue en training (Vol. 27, ho 3) focuses specifically en methods for measuring training effectiveness.
By building on work already done in this area outside the nuclear industry, the staff expects to be able to develop criteria for
NAS Response 20 measuring the success of the various training approaches or training devices being pursu'ed by industry.
Finally, regarding training simulators, the NRC has imposed on industry a requirement to install whole-task, plant-referenced simulators by 1991.
The staff intends to defer consideration of research on training simulator effectiveness in the short-term and recomends that incustry take the lead to determine the effectiveness and benefits of part-task versus full-task simulators for acquiring and maintaining operator skills, f
Regarding Recomendation 2, as the the panel points out, the industry f
should take the lead in striving for excellence, and hence advancement of training should be one of industry's greatest concerns, especially since the adaptation of better training can have economic as well as safety benefits.
However, since the potential safety significance of operator readiness to deal with complex physical as well as human-factor phenomena during rare events has been demonstrated (e.g., Chernobyl), the staff will initiate research in FY
' 1989 on general criteria for evaluating training adequacy.
Special attention l
l will be given to skill achievement ano sustainment training for responding to j
i rare events. This effort will be pursued as a potential joint industry and NRC I
effort.
Regarding Recomendation 3, the staff agrees with the NAS panel that longitudinal research on training should be undertaken jointly between the NRC and industry.
Therefore, the staff is planning to extend its research on all aspects of training, including cooperative studies with industry, into the 1990s.
NAS Recomendation Regarding Qualifications (Degree Requirements, Testing and Licensing)
The NAS panel recomends research on porsonnel qualifications in the following areas:
1.
advisability of requiring a baccalaureate degree for control room staff, and if a degree is required, the recomended course content.
NAS Response 21 2.
reliability and validity of licensing examinations, and on improvements in the process of administering these examinations.
3.
qualifications of the regulatory staff and contractors engaged in licensing NPP operators.
4.
factors that affect team performance.
5.
qualifications of personnel other than licensed operators at nuclear power plants, especially maintenance personnel.
NRC Staff Response Generally the staff agrees with the NAS recommendations listed above.
More specifically, with regard to Recommendation 1, the NRC (RES and NRR) has conducted numerous studies to determine the academic knowledge needed by an operator and to determine in which types of degrees that knowledge can be obtained (NUREG/CR-4051 Assessment of Job-Related Educational Qualifications for Nuclear Fower Plant Operators and NUREG/CR,-4411 Assessment of Specialized Educational Programs for Licenseo Nuclear Reactor Operators).
The extent to which college engineering curricula cover job-related academic knowledge has been assessed.
Therefore, the staff does not perceive a present need for research on knowledge and course content for operators at this time.
With regard to Reconrendation 2, NRR has completed, or has ongoing work in the areas of testing and licensing of operators.
For example, l
o a new pilot requalification exam is being tested and implemented which is responsive to industry concerns and is responsive to the need for crew level evaluation, o NUREG-1122 Examiners Handbook, provides test specifications which are based on operator knowledge and abilities (KAs) which in turn are based t
NAS Response 22-1 on empirical studies.
In addition, examiners now develop tests that test KAs in proportion to the level of rated importance of the KAs.
o NRR is also developing a performance-based simulator examination.
NRR has constructed performance-based rating scales that represent the skills required of operators.
In addition, with regard to the simulator exam, NUREG-1291 BWR and PWR Off-Normal Event Cescriptions presents BWR and PWR simulator scenarios that test the important knowledge and skills of operators.
o NRR is currently studying the development of a "Fundamentals" exam which would be standardized and centralized.
The exam would cover the fundamentals of nuclear power and would be given nationwide in several locations each year.
o NRR has contracted with several consultants to evaluate the adequacy of current licensing examinations.
Recommendations from these evaluations have resulted in changes to the examination.
With regard to Recommendation 3, NRR has an ongoing program to develop a new "Initial Examiners Training Program:" This training program has two parts, (1) training en relevant technical NPP issues and (2) training on techniques for giving examinations.
With regard to Recommendation 4, RES has initiated research in the area of team performance.
This research is directed toward developing criteria to evaluate team performance.
With regard to Recommendation 5, RES will initiate research in FY 1989 on the ocalifications required of NPP personnel other than licensed operators, especially in the area of maintenance.
l t - - - - ' -
+,v v
r w
y r
r, - - - - -.
r-eaa-g -
r, w
NAS Response 23 o
NAS Recomendation Regarding Staffing (Shift Technical Advisor, Screening and Selection, Shift Scheduling and Vigilance)
The NAS panel recomends res'earch on:
1.
how to improve the effectiveness of the Shift Technical Acvisor (STA),
especially the STA's role vis-a-vis other members of the control room crew.
2.
potential screening and selection baseo on psycho-physiological profiles.
problem-solving skills, and standardized performance tests.
3.
shift scheduling and vigilance addressing the issues of work schedules, overtime, boredom and work rotation.
NRC Staff Response j
i With regard to Recomendation 1 on the STA, NRR has already completed substantial work in this area.
NRR first reviewed the practices in selected foreign countries for providing engineering expertise on shift and the results were published in NUREG/CR-2952, Eng'ineering Expertise on Shift: The Foreign Experience.
NRR also had a contractor examine different approaches that have -
been used by U.S. utilities in implementing the STA requirement.
Next, NRR had a contractor conduct interviews with utility personnel which led to the publication of NUREG/CR-3396, Eyperience with the Shift Technical Advisor Position.
Lastly NRR published NUREG/CR-3785, Alternative Aoproaches to Providing Engineering Expertise on Shift.
After the degree requirements issue
' l for SR0s is resolved, the hRC staff will review the need for research in the area of the STA.
It is expected that the STA role will be easier to define after decisions on qualifications of the SRO have been established.
With regard to Recomendation 2 for research on personnel screening and selection, the staff has not done research on these topics.
The staff believes that it is more appropriate for industry to undertake research in this area and we will make this kncwn to industry.
4
NAS Response 24 With regard to Recommendation 3, the staff is initiating research during FY 1988 on shift scheduling and vigilance.
This research builds on previous work
/
in this area (NUREG/CR-4248, Recommendations for NRC Policy on Shift Scheduling and Overtime at Nuclear Power Plants, ongoing work at EPRI and at individual plants).
It also takes cognizance of recent problems of operator alertness on shift.
FY 1988 research on shift scheduling will focus en:
o confirmatory research on shift schedule and overtime data collection and dnalysis.
Nuclear power plant data on the effects of overtime, shift scheculing, shift rotation, and staffing will be collected and analyzed to assess the adequacy of the current technical basis for existing NRC policies.
The results of this research will be used as a basis for updating policy revisions if appropriate.
\\
o shif t duration, shift station, and operator alertness.
This will be exploratory research recognizing that new approaches such as "informed innovation" (NAS Report, p. 70) may be required to address each of the above issues.
More.specifically, this research will focus on ways in which the state of knowledge outside the nuclear arena, especially in the area of vigilance, can be applied to nuclear power plant operations. It will explore, as the report suggests, biological, social and environmental based solutions as opposed to harcware solutions.
Issues I
such as boreocm, work-lead, social-interactions, and sleep-wake cycle will be considered.
- c. _ Human Performance NAS Recomrendations:
The NAS report recognizes that human performance (human error) can play a major role in the malfunctioning of nuclear power plants.
The report concludes, however, that there is too much uncertainty in measuring human error rates, since current quantification methods rely primarily on expert judgement
NAS Response 25 Y
and on very limited operational experience'fata.
Finally, the report concludes that the greatest limitation of these~ methods and data is in the area of cognition.
\\\\
The report recomends that research be directed toward a r'Nre fundamental understanding of the nature (mechanisms) and causal factors of human
,1 performance (including error), and subsequently toward more quantitative,ered
[ j ',
N theoretically based models and methods which are not only capable of producing credible snapshots of performance, but which are also capable of predicting performance in rarely occurring events and in response to system design
- it changes.
More specifically, the report makes the following recomendations regarding performance measurement and performance data.
/
1.
Develop a better understanding of the mechanisms of hunan error so that circumstances and conditions under which such errors cccur can be predicted 3
beforehand, rather than merely on a subjective probability basis.
2.
Develop perfory::ance measurement models arid methods, udng computer simulation techniqms where possible, which are qualitative and quantitttive, and are theuryt tiriven and empiric 611y based.
These mcdels should capture both '
behavioral ano cognitive aspects of performance, should be predictive as well as' descriptive, should be capable of identifying and diagnosing causal factors, s't.culd be capable of analyzing individt.al, team and organizational performance, andshouldbeapplicableacrossperformagcesituationsandplints.'The v
i empirical base should include operating experience, laboratory experiments, and controlled studies on plant simulators.
l 3.
In developing these models and methods, fccus on operations, maintenance and organizational aspects of human performance at nuclear power plants.
\\
l
\\
4 In Ueveloping these models and methods, build on work done outside the nuclear industry (e.g., military, aerospace),
r, i
/
l
[
[' '
J
\\
I i
\\_,
,j 1p'
NAS Response 26 5.
Comence research on these models, methods and. data acquisition mechanismi; imediately, and focus on both theoretical and empirical aspects of plant operations, maintenance and organizational performance measurement.
6.
Pursue an industry-wide effort in.this area with a single organization, preferably the NRC, integrating the theoretical, analytical'and empirical results.
,b 7.
Develop mechanisms for acquiring valid operating experience data and unencumbered (r.ot confounded by instructor interference) training simulator data to insure greater certitude in modeling performance and its asaciated causal factors.
NRC Staff Response:
Pertormance measuring techniques which rely primarily on expert.judgemer;.-
are only an interim solution.
With regard to Recommendation 1, we are examining the feasibility of REC engaging in such as effort. While we agree with the general thrust of the recomendation, we currently do not sponsor controlleo laboratory studies of the type that are needed to fully address the issue due to a lack of laboratory facilities and to a lack of an appropiate human subject pool.
We have, howaver, developed computer-based sirc.ulations wh'ich we believe can provide us with insights into probable mtchanisms of human error on both cognitive and behavioral levels.
With regard to Recommendations 2 through 5, we have developed or are in the process of developing performance measurement rrodels and methods which generally are:
(1) theory driven and empirically based, (2) qualitative and quantitative, (3) predictive as well as descriptive, (4) capable of analyzir.g behavioral and cognitive aspects of performance and their causal factors, (4) focused on operations and maintenance activities, (5) capable of analyzing individual, team and organizational performance, and (6) cpplicable across.
_NAS Response 27 0
performance situations ano plants.
The majority of models and trethod,s emerging from this research are computer-based, and build on knowledge developed in other industries and academia.
For example, we have developed and tested a Maintenance Personnel Performance Simulation (MAPPS) corrputer model based on stochastic techniques cited in the NAS report.
MAPPS logic takes advantage of thirty years of modeling experience gained in the military and the aviation industry.
It is capable of simulating the activities of traintenance mechanics, electricians, instrumentation and control technicians and supervisors in groups of two to eight.
It allows the analyst to assess human performance using a series of qualitative and quantitative indexes including success probability.
It allcws the analyst to preview the potential effects, on perfonnance, of man-man and man-machine retrofits.
Finally, it allows the analyst to investigate on a limited basis the effects of operations and maintenance interfaces on organizational perforr'ance.
We have developed and are testing a Cognitive Environment Simulaticn (CES) for analyzing intention formation (decisionmaking) aspects of human performance.
CES 1s computer-based and currently resides at the bestinghcuse Research and Development Center (our contractor) facility.
CES employs the latest artificial intelligence / expert systems processing techniques.
CES allows the analyst to investigate dacisionmaking performance and its causal f actors, and to preview alternative' cognitive strategies resulting frcm man-tran and man-machine retrofits.
Finally, CES allows the analyst to investigate individual and team decisionrraking behavior.
CES is currently unoergcEng operability testing.
Unlike MAPPS, CES is deterministic and does not quantify human performance on a probabilistic basis.
Therefore, as part of the CES project, a Cognitive Reliability Aralysis Technique (CREATE) which converts CES outputs to error probability statements has been developed and is being tested.
We also have research ongoing to cevelop indicators of individual, team and organizational performance.
These indicators are being developed based f
NAS Response 28 primarily on existing plant data sources (e.g., Licensee Eyent Reports [LERs],
Nuclear Power Reli6bility Data System [NPRDS], In-Plant Reliability Data System
[IPRDS]).
Finally, we have recently initiated research using a theoretical and empirical model for analyzing organizational and management perfortnance influences on safe operations.
Subsequent methods development will use the results of the research, if successful, as a basis for integrating management factors into the probabilistic risk assessment (PRA) process, and as a basis for ceveloping new indicators of organization and management performance.
With regard to Recomendation 6, there is presently no industry-wide research program directed toward human performance models and measurement trethods.
It should be noted, however, that much of our research in this area es industry groups, the utilities and other research and regulatory agewies.
For example, we are just completing a human reliability analysis (HRA) methods evaluation in conjunction with EPRI and the Comission of European Comunities (CEC).
We are also pursuin'g a cooperative agreement with the CEC, and U.S. and European utilities, to employ the 1%PPS computer code at an analytic and design tool for helping resolve maintenance related issues cf interest to the utilities. 'This study is part of the MAPPS technology transfer prccess.
Finally, we have and will continue to use independent review groups composed of a broad spectrum of industry representatives to assist us in each of the modeling and rrethods programs described above.
With regard to Recomendation 7, we consider the paucity of operational experience data and training simulator data to deter human performance assessment.
Currently, we do not have ongoing or planned a research activity directed solely toward collection cf training simulator data.
Howeyer, we have requested that the Halden Pro,)ect provide to us training simulator performance data it collects as part of its ongoing research acti"ities, for input to our IIRA/PRA cata bank, known as NUCLARR, and located at the Icaho fir'.ional Ergineering Laboratory. We are also looking to the NRC train:ng simulators located at Chattanooga, TN.as potential sources of performance data.
e n
NAS Response 29 Developmental research was ccmpleted on a third party reporting system concept of the type currently being employed by other U.S. regulatory agencies and the milita ry.
We have also initiated research with the George Mason University to investigate the feasibility of a method for systematically equating, on both behavioral and psychological levels, human tasks performed outside the nuclear industry (military, aviation, ground transportation), on which considerable operating experience data exist, with tasks performed by operators and maintainers of nuclear power plants.
Such a method would allow us to employ r.cn-nuclear industry oata as surrogate measures of performance, especic11y for doing probabilistic assessments.
In summary, the research staff feels that it has ongoing and planned research fully responding to the NAS recomendations except for:
(1) mechanisms of human error, and (2) an inoustry-wide prograni.
In the case of mechanisms of human error we will continue to study the feasibility and propriety of establishing a long-tenn basic research program, within the NRC and/or in conjunction with industry, to more fully address this issue.
In the case of the industry-wide program, the staff will continue to examine ways in which the NRC, industry and academia can work more closely to achieve ord validate performance models and methoos of mutual interest.
i d.
Management and Orgsnization The NAS report recognizes that management decisions impact many aspects of plant safety, both directly and indirectly.
The primary concern voiced in the NAS report in the area of organization and management is the extent to which the organization is able, quickly and efficiently, to prevent, detect, and react to any threats to overall system safety.
The NAS panel therefore recomer.cs research on the folicwing topics, j
1.
The impact of regulations on the practice of management Research in this area should answer cuestions such as "are there organizational ccrditions that produce a sense of regulatory overload?"
NAS Response 30 2.
Organizational Design and a Culture of Reliability Research in this area should answer questions cor.cerning the effects on performance of such factors as span of control, vertical height, horizontal breadth, functional specialization, interunit coordination, and the linkages between staff and line.
Research should answer how these variables affect safety.
3.
Operational Decision Making Research in this area should answer questions concerning the effects on performance of communications between groups such as management and operators, or corporate and plant management.
4.
Timely recognition of emergencies Research in this area should answer questions concerning the effects on performance of organizational designs to deal with emergencies since organization designs for normal operations may not be effective.
5.
Characteristics of Managers Research in this area should answer questions concerning the effects on their performar ce of supervisor and manager demographics such as interpersonal skills, academic training, corporate training and on-the-job experience.
fjRCStaffResponse The staff, recognizes that managenent and organizational factors attending the ope' ration of nuclear power plants appear important.
In recognition of this, the NRC has funded several research and technical advisor;/ service projects in this area over the past six years.
The statf also recognizes that solution to some of the issues surrounding effective tranagement and organization climates at nuclear pcwer plants will require a long-term effort.
Finally the staff recognizes that research of this type vill require cooperation among the NRC, EPRI and INPO who also have research anc development in% rests in this area.
Regarding Recorrirendation 1, the staff considers regulatory philosophy to be a policy as well as a research issue.
v 3
e,
NAS Response 31 Regarding Recommendations,2 through 5, the staff has an cngoing research project whose results will be used as one basis for developing additional research in the management and organization area.
This project builds on results of work in this area completed during the past six years within RES and NRR.
It is directed toward methods for identifying and evaluating the influences of maragement on plant performance from both human factors (qualitative) and human reliability (probabilistic) perspectives.
It involves tasks leading to identification of:
(1) a oynamic model (characterization) of the plant organization; (2) supervisory and management functions and roles critical to performance; (3) profiles of supervisory and management influences; (4) data required for assessing those functions and roles, and means for acquiring such data; and (5) an algorithm for integrating those measures into reliability and risk assessments.
Objectives 1 through 4 respond to the the NAS recommendation for human factors research in the management and organization area.
In addition, Objective 5 includes an approach to applying the results of the research to probabilistic risk assessments, e.
The Regulatory Environment NAS Recommencation Regarding Models of Regulation The NAS panel recommends research on:
1.
the possibility of assigning regulatory activities to different organizations and transferring some regulatory functions from the regulatory bodies to the utilities themselves.
2.
a closer examinction of the effects of regulation.
3.
human factors practices within the NRC itself.
NRC Response:
With regard to Reconn.endations 1 and 2, mechanisms for ensuring that regulatory 4
i
+- -
nw---m
-~
,m
NAS R:sponse 32 activities enhance, safety without placing undue burden on the industry do exist (e.g., the regulatory analysis process, review by the Committee to Review Generic Requirements, and the "Backfit Rule").
Efforts to allow the industry to play a larger role ir, regulation (e.g., in the area of training) aise exist.
With regard to Recomendation 3, the staf f believes that increased NRC management and staff awareness of the human factors discipline and its potential role in achieving the Agency's mission is important.
However, the staff does not plan any research on methods for increasing human factors awareness at this time.
NAS Recomendation Regarding Plant Performance Indicators "The fundamental purpose of performance indicators," the NAS panel states, "is to readily monitor and assess individual plant perfonnance and take actions when appropriate".
- e report recognizes that the NRC has made progress in this area. It points out, however, that the current indicators are based on publicly available data which are limited and do not allow monitoring of all important facets of safety, in particular, monitoring indicators of diagnostic value. The report concludes that additional work is needed.
The use of incentive-based indicators monitored by the public utilities is also discussed in the NAS report. It point,s out that we do not exactly know how such indicators affect safety; some positive as well as negative aspects are mentioned, it recomends research in the area which will allow us to, first, evaluate safety effects and second. improve safety by developing regulatory mechanisms based on combined safety and utility incentives.
Specific MAS recomendations on performance indicators can be sumarized as tollows:
- 1. Pursue research on better performance indicators based on data not currently available to the NRC staff.
This should include data on "inputs" to piant operations, "throughputs" and "outputs." Also, data should reflect
~1%
NAS Response 33 performance at different levels, that is, at the individual operator level, at the team level, and at the plant level.
Finally, performance indicators and the data upon which they are based should be continually updateo and validated as part of the research effort.
"Plant performance" is a variable depending on many time-dependent parameters, (such as plant aging, regulatory changes and technologychanges),affectingthevalidityofanindicator.
- 2. Pursue research on the effectiveness of incentive programs currently in use by the public utilities, especially in terms of their safety significance Focus research on the interactions of three variables which may provide a basis for structuring regulatory mechanisms for improving both safety and performance:
(a) incentive-plan characteristics, (b) utility responses, and (c)plantperformance.
NRC Response:
The NRC plans for research on performance indicators are consistent with the NAS recomendations. The NAS report correctly states that publicly available data are not sufficient to support monitoring all variables important to safety.
The NRC is developing performance indicators, especially at the plant i
level, beyono those already in the program.
The current set of indica, tors is based en the developmental work by an interoffice task group during FY 1986-87.
Ongoing research and development activities on performance indicators can be sumarized as follows:
o Risk-based inoicator work is focuseo on "outputs" and some aspects of "throughputs" of plert safety performance, e.g., f requency of ever.ts,
equipment availability. As part of this effort, an indicator of safety system unavailability has been ic'entified (and theoretically tested utilizing PRA methods) as a measure of safety (System pnavailability Indicators, BNL draft report, Sept. 1987). As a result of this work the NRC is reviewing alternative ways to collect the neeoed data. The
'NAS Response 34 development of risk-based indicators is. continuing and is aimed at improved methods of monitoring safety by analyzin'g plant performance
- outputs, c' Efforts to develop programmatic performance. indicators are focused on internal parameters of plant performance such as "inpLts" and "throughputs." In FY 1987, work was limiteo to analyzing publicly i
available data and determining future data needs.
As a-result of this
~
work, a set of measures was oetermined based on mon 1toring 'causes of reportable events' such as operator errors,' personnel errorc, maintenance problems and others.(Progrannatic Performance-Indicators, PNL Draft ".eport, March 1988).
Furthermore, the study illustrated the limitations of readily available data for indicate:s suggested by I
experts as true safety indicators (such as percentage of contaminated areas). In FY 1988 work is ongoing to analyze data beyond those that are publicly available in order to identify leading indicators, that is, indicators that are intended to act as early signals of changing trends in plant performance.
This is not a trivial task because it deals with the whole spectrum of plant operations. The questions of what "inputs-throughputs-outputs" to measure and how, have yet to be E
a resolved.
For this purpose, a structured framework describing relationships between functional areas of plant performance and safety is being pursued. It is expected that such an approach will alicw the determination of a coherent set of effective measures.
i In addition to these developmental efforts, AE00 is continuing to work on program improvements.
Issues such as:.the monitoring and interpretation the current set of indicators, the implementation of new indicators, the impact of
~
ino1cator application to plant. safety, the time-cependency of plant-8 performance, indicator update, and real-time validation, are being analyzed and j
evaluated by the AE0D iniplementaItion activities and the results are fed back to l
research programs.
I A
2
NAS Response 35 NRC's current efforts on performance indicators are quite extensive. The main focus is on measures of performance at the plant level which is what the NRC. regulates.
Individual or team performance indicators would discriminate plant performance in a very detailed level. -Finally, we will include in the indicator validation part of our research the indicator-update, in order to incorporate the effects on plant performance from external factors such as regulatory changes, technology changes and incentive plans.
With regard to Recommendaticn 2, the NRC currently tracks, using NUREG-1256 Incentive Regulation of Nuclear Power Plants by State Public Utility Commissions, the economic performance incentive plans used by the utilities, and studies them on a generic basis.
Although we have some concerns for the safety implications that these plans might have in the long-run, (by shifting the emphasis of plant operations from safety to incentive-plan goals), the NRC has not yet performed detailed analyses of the effects of these incentive-plan goals on safety. However, AE00 has recently initiated a program to monitor the performance of plants, as measured through performance indicators, that are under the incentive regulations of public utilities ccmmissions.
This is a first step towards determining the impact of economic incentives on safety performarce.
b I
e O'
C:
I'k a
6 ENCLOSURE 3 t
9 i
?
l 4_,
_-w.-
'r+
- " i "'
g
.o1v,-r7--**"*'
" " ~ " " ' " " '
NRR'S HUMAN FACTORS PROGRAMS 1.
Licensing Issues Human factors was established as a significant area of endeavor within NPR after the accident at Three Mile Island.
Since that time, programs and criteria have been established addressing:
- 1) control room design reviews, 2) safety parameter display systems, 3) emergency operating procedures, 4) plant personnel training, and 5) operating examinations and licensing.
Generally, the focus of these programs involved licensing issues. The status of these ongoing programs is summarized below.
C0tlTROLiqMDESIGNREVIEWS On December 17, 1982 the NRC issued generic letter No. 82-33, "Supplement 1 to NUREG-0737--Requirements for Emergency Response Capability," to all nuclear power plant operating reactor licensees and holders of construction permits.
One of the elements of tais generic letter was for each licensee and applicant to conduct a detailed control room design review (DCRDR).
The objective of the DCRDR was to improve the ability of nuclear power plant operators to prevent accidents or cope with accidents by improving the information provided to them.
This design review was to identify any modifications of control room configuration that would contribute to a significant reduction of risk and to enhancement in the safety of operation.
As of the end of fiscal year 1987, NRR had issued DCRDR safety evaluation reports (SERs) for approximately 88 nuclear pcwer units.
Of this number, 16 have been closed out, having all control room improvements implemented. Another 30 units have proposed satisfactory improvements but have not yet completed implementation.
Staff review efforts associated with DCRCRs has been reouced in FY1988 due to higher priority NUREG-0737 activities.
The staff is considering alternate approaches to bring closure to the DCRDR process.
SAFETY PARAMETER DISPLAY SYSTEMS One cf the recommendations of the Three Mile Island Unit 2 Lessons Learned Task Force was that a minimum set of plant parameters, descriptive of plant processes, should be concisely displayed in the control room to provide plant operators with a simple, integrated measure of plant safety status. The safety parameter display system (SPDS) was another element contained in Supplement 1 to NUREG-0737.
Its stated purpose was to aid control room operators in rapidly and reliably determining the safety status of the plant during an emergency.
Since rapid deployment of the SPDS was a design goal, the NRC staff review was established to minimize interference.
Thus the staff undertook post-implementation reviews of the in-place SPDSs.
By April 1986, the staff had issued SERs for about half'the nuclear power plants.
These SERs were based on documentation reviews and meetings with licensee staff.
Because there was a concern that the basic requirements for the SPDS were not being met, six SPDS onsite progress evaluations were conducted.
It was concluded that many SPDSs may not be achieving the gcal of aiding control
room operators in rapidly and reliably determining the safety status of the plant during an emergency. Additional visits to re/iew SPDSs, claimed to be operational, have further confirmed that very few of the systems are fully meeting the requirements.
The staff is currently developing a revised approach to determining the acceptability of the SPDS, EMERGENCY OPERATING PROCEDURES Another element of Supplement 1 to NUREG-0737 was the upgrade of emergency operating procedures (EOPs).
The primary intent was to ensure that E0Ps be human factored and function oriented to improve human reliability and the ability to mitigate the consequences cf a broad range of initiating events and subsequent multiple failures or operator errors, without the need to first diagnose specific events.
Licensees were required to reanalyze transients and accidents and prepare technical guidelines. The required analysis was to identify operator tasks, and information and' control needs.
In addition, the analysis was to serve as the basis for integrating upgraded E0Ps and the DCRDR and verifying the SPDS design.
It was further reouired that licensees submit Procedure Generation Packages (PGPs) which would be reviewed by the staff. On an audit basis at selected factilities, the staff would review upgraded E0Ps developed using the PGP.
Currently, all licensees have submitted their PGPs for review by the staff, and SERs have been issued for approximately one third of those. Nineteen implementation audits and inspections were conducted to evaluate the edequacy of the PGP implementation, with disappointing results. Consequently, the staff has instituted an accelerated inspection progran that focuses on the adequacy of the E0Ps themselves.
As part of this effort, it is anticipated that about 28 E0P inspections will be conducted during FY1988, with special emphasis on BWR11 ark I plants.
Review of PGPs has been minimized due to redirection of resources to the inspection program.
PLANT PERSONNEL TRAINING On February 7,1985 the Commission adopted the Policy Statement on Training and Qualification (50 FR 11147).
The Policy Statement endorsed the INP0-managed Training Accreditation 7rogram and the industry, through NUMARC, commitment to have 10 programs at 61 sites ready for accreditation by December 31,~1986.
In approving the Policy Statement, the Commission deferred rulemaking on training and qualification for two years in recognition of industry efforts and directed the staff to independently evaluate implementation of improvement programs.
NUMARC met its commitrent and the staff has proposed that the NRC continue to endorse the accreditation program.
As of December 31, 1987 INP0 had conducted 134 teen visits covering 605 programs.
They also cerducted 16 team visits covering 72 programs at 10 new facilities. A total of E75 programs have been accredited at 67 sites.
NRC staff has observed 19 visits and conducted eight post-accreditation reviews.
)
l 1
o i OPERATOR EXAMINATI0flS AND LICENSING Section 107 of the Atomic Energy Act of 1954, as amended (42 U.S.C. 2137),
reo.uires the hRC to prescribe uniform conditions for licensing individuals as operators of production and utilization facilities, to determine the qualifications of these individuals and to issue licenses to such individuals.
Regulations to implement these requirements are set out in Part 55 of Title 10 Chapter 1 of the Code of Federal Regulations. These regulations were updated in March 1987 to meet NRC responsibilities under Section 306 of the Nuclear Waste Policy Act of 1982.
Reactor operator and senior reactor operator licensing written examinations and operating tests are. scheduled and administered through the NRC Regional Offices.
Regional office personnel also conduct cperator requalification examinations and requalification program evaluations at all licensed facilities.
Recently, a methodology for the evaluation of certified simulation facilities was issued. This methodology will be used to ensure that simulators are adequately designed for use in the conduct of operating tests.
Program oversight of the operator licensing function is maintained at headquarters in the Division of Licensee Performance and Quality Evaluation.
Responsibilities include regional support and oversight, examination development and simulation facility evaluation.
II.
Operations and Maintenance Issues The focus of the efforts cf NRR in the area of human factors has been gradually shifting from licensing issues to an emphasis'en plant operations and maintenance. Industry representatives have estimated that roughly 50 percent of significant operating events involve human performance issues. Therefore, it is no longer sufficient to focus on hardware as the sole determinant of plant safety.
Rather, the impact of plant personnel on safe plant operations and maintenance must be embraced as an integral factor in the Agency's decision-making processes and determinations of corrective actions.
To support this shift in emphasis, NRR has become involved in the evaluation of performance and the assessment of risk associated with human factors issues.
These' efforts are described below.
LICENSEE MANAGEMENT AND ORGANIZATION The Division of Licensee Performance and Quality Evaluation has provided support to the human factors evaluation of licensee management and organizatier.
through its active participation in team investigations.
Such participation has occurred in the evaluation of management effectiveness at the Davis Besse, Peach Bottom, and Turkey Point plants.
In addition to these evaluations, human factors evaluations have been an integral part of the diagnostic inspecticns conducted at the McGuire and Dresden plants.
9
l 4-HUMAN RELIABILITY AND RISK ASSESSMENT The Division of Radiation Protection and Emergency Preparedness has a project underway with the Brookhaven National Laboratory (BNL) to estimate the sensitivity of risk parameters to human error rates.
The objective of this effort is to identify and characterize risk significant human. actions or categories of actions using PRAs from two plants.
As part of the effort heran factors specialists will be utilized to address the adequacy of the data base and make an assessment of the adequacy of the modeling of human errors in the PRAs.
i This project will be integrated with a related PES-sponsored prograr and is intended to result in the development of a method for estimating the contribution of management. influences to' risk in operating plants.
I i
I I
r J
L w
-r g
u.-
-m-
,yn
.~.-,-,+,----ag--
y,,,m_,y
,,..w-.,
.,,-...,-n.m.n.,-
+ - -ewsw-~-,w ww o -
,,m---
i Y
l J
i ENCLOSURE 4 4
e 0FFICE FOR ANALYSIS AND EVALUATION OF OPERATIONAL DATA ACTIVITIES IN HUMAN FACTORS The Office for Analysis and Evaluation of Operational Data currently has a number of activities that involve evaluation of human factor considerations on the safe operation of nuclear power plants and nonreactor activities and radioisotope uses licensed by the NRC.
Human factor issues, such as training, procedures and personnel actions have been incorporated, or are the specific subject of our ongoing review of operating experience.
The office has firm commitments for some near term activities for performance indicators that involve human factors; some of these activities are being performed jointly with the Office of Research.
In addition, the Technical Training Center of AE00 has the capability to support additional potential future work by the Office of Research.
I.
Ongoing and Continuing Activities Both divisions of AECD have programs that involve assessing the impact of human factors on the safe operation of nuclear power plants.
A.
Division of Safety Programs One of the major sources of information on the operation of nuclear power plants is the licensee event reports.
Each report is reviewed by several engineers to assess the significance of the event.
In addition, each report is encoded for computer retrieval, with emphasis on coding of causes of the events and corrective actions.
When a series of events are identified as stemming from a common cause, an engineering evaluation or a special study of the events may be prepared if the problem appears to be generic.
In the past, AE00 has frequently assessed human performance and human factors considerations in its studies of operational experience; for example, one of the AEOD studies concerned the contribution of labeling to operator errors involving the wrong unit, train, component or channel.
The causes of events (cause codes) and corrective actions are identified through expert evaluations of the licensee event reports for monitoring in the Performance Indicator programs..Three of the cause codes, namely:
Licensed operator error, other personnel error, and administrative control problems directly involve human factor issues.
Similarly, several of the corrective actions, such as training, discipline, and management changes involve human factors.
Events reported for nonreactor operations are also reviewed to determine the significance of the event, with the reports encoded for computer retrieval.
Most studies of nonreactor events devote a substantial portion of the study to the human factors aspects of the events.
Past studies have included medical misadministration, radiography 9
l 0
-2 overexposures, and ruptures of sealed well logging sources.
Sirce manual cperations dominate these type of events, human factors were found to be l
a substantial contributor to the events.
These human factors included lack of quality assurance plans, absence of procedures, and~ failure to follow procedures. A current study of events at large irradiators includes an assessment of such factors as the contribution of management to the occurrence of events at these facilities.
B.
Division of Operational Assessment D0A incorporates human factors issues in both the Incident Investigation Team (IIT) and the Diagnostic Evaluation Team (DET) programs.
The IIT manual specifies that the team investigating a reactor event should include experts in human factors as well as operations (licensed operator).
The findings by these team members are included in a separate section of the team report.
These provisions assure that the human aspects of the event are given adequate attention by the team.
The DET goals include identifying actions and involvement of licensee management and staff in safe plant operation.
To make this identifica-tion, both corporate and plant management effectiveness (strength and weaknesses) are evaluated.
In addition, the DET attempts to identify problems of an organizational climate that could contribute to performance problems. The DETs have included human factors experts from the NRC staff, as well as management consultants.
The Technical Training Center is currently providing training to NRC staff who inspect licensee emergency operating procedures and to operator license examiners who examine license candidates.
II. Possible Future Activities The Technical Training Center now has three reactor simulators. Although their first priority must be dedicated to training NRC staff, they cculd also be used in support of programs by the Office of Research.
These programs might include: assessment of operating team performance; studies of staffing composition; evaluation of operator performance under stress; studies of shift duration and operator vigilance; studies of optimum reliance on the Safety Parameters Display System; and use of the simulators in support of the Cognitive Environment Simulation program.
I Several of the Technical Training Center instructors held Senior Reactor Licenses.
These individuals are an agency resource that might be used in conducting sc'fre of the research programs that use the NRC siraulators.
l
)
_.