ML20246G120
| ML20246G120 | |
| Person / Time | |
|---|---|
| Issue date: | 04/20/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-2620, NUDOCS 8905150192 | |
| Download: ML20246G120 (16) | |
Text
-__
s CERTIFIED COPY n., e
-. > - q r,,
DATE ISSUED: April 20, 1989 h
~~0 h hb c
fDB GON
SUMMARY
/ MINUTES ACRS HUMAN FACTORS SUBCOMMITTEE MEETING, THURSDAY, JANUARY 26, 1989 Room p 422, 7920 Norfolk Avenue Bethesda, Maryland The meeting was convened at 8:30 am in addition to Dr. Forrest J.
Remick, the following ACRS members were in attendance: Mr. Carlyle Michelson, Mr. James Carroll, and Mr. Charles J. Wylie.
Herman Alderman was the cognizant ACRS staff member for this meeting.
Dr. Remick noted that the purpose of the meeting was a discussion and review of the Draft Revision 1 of The Human Factors Regulatory Research Program Plan.
Dr. Remick said that the ACRS is and has been a strong supporter of human factors research. He said he had some personal comments on the draft.
He noted the draft is not complete, and requires some good. editing.
He noted inconsistencies in the table of contents.
He noted the wording from section to section is differ-ent.
He said he found parts 1 and 2 be very weak.
He notes part 3 o
was better than the other parts.
He said the interrelationship of the e
8'.
various projects was not always clear.
He stated that he didn't think co b
the draft was ready to go to the Commission.
a:0 Mr. Michelson asked why no consideration had been given to En$$
/
looking at how resident inspectors perform and how they interact in a
, human,fa tors sens.e with the plant that they are assigned to.
r, DESIGNATED ORIGINAL Qg,am y,,
.Cj Certified By
I-1 Minutes / Human Factors
~ 2
- ubcom.Mtg. Jan. 26, 1989 S
Mr Wylie noted that he assumed as he read the draft that it was a preliminary approach in laying out a process by which to attack human d
factors research rather than spelling out in detail every aspect of human factors research th6t is going to be accomplished.
1 Mr. Frank Coffman, Branch Chief, Human Factors Research made the opening presentation.
He noted that the subcommittee had been briefed on the first version of the human factors research plan on March 28, 1988 and the full committee was briefed on April 7,1988.
He noted the purpose for.the research is a multi-disciplinary endeavor and it relies heavily upon the behavioral sciences.
He said it involves a variety of engineering disciplines.
Mr. Coffman noted that their research takes two different forms.
One is timely support of regulatory decisions and the other is techni-cal assistance which is more foundational but still has potential applications.
Dr. Remick remarked that the research program plan should point out that there could be human factors efforts under technical assis-tance program in other offices.
Mr. Coffman noted the plan is intended for those who contribute to or make decisions that impact human factors regulatory research, p
Dr. Remick said the research report should show the intended audience.
l
- I Minutes / Human Factors
'3 Subcom.Mtg. Jan. 26,1989 Both Dr. Remick and Mr. Michelson pointed out that it wasn't clear that the Research program plan was Research's program plan and not a plan proposed by the entire Nuclear Regulatory Commission.
Mr. Coffman responded that they had responded to the staff's requirements memorandum. The SRM asked to provide a description of the resehrch plan and the staff interpreted that to'mean research.
Mr. Coffman said the research supports the entire agency.
Dr. Remick reiterated the need to include human factors activ-ities done under technical assistance with research handled by the office of Research.
Mr. Carroll noted the staff should make clear what research is done on other than nuclear power plants.
I Dr. Dennis Serig, NMSS said the NMSS human factors program is just starting.
He is working on development of a human factors action plan for NMSS.
He said he is trying to determine the research needs and coordinate those with the Office of Research to establish a research program. Two areas he has been working on medical research and industrial radiography, i
Mr. Coffman discussed the general approach to regulatory re-search. The regulatory research is coordinated with other offices.
In addition there is interaction with NUMARC, INPO, and EPRI. The staff also takes the Human Factors Society's report and the National
_...________________._____._____________________J
.. Minut'es/ Human Factors 4
Subcom.Mtg. Jan. 26,.1989 r
Academy of Sci _ence's report to be input.
He noted the emphasis was on regulatory users needs.
He said that the regulatory issues are both
)
risk significant and a regulatory need.,
l Mr. Michelson pointed out the differences between regulatory needs and significant risks.
He said you may have a regulatory need without a significant risk.
Dr. Persinsky noted that the Office of Research identifies areas where research is to be done based upon probabilistic risk analyses.
Mr. Michelson asked what was meant by technical resolution?.
Mr Coffman cited the example of degree requirements in the control room.
He said the technical resolution would be if you could establish objectively with repeatable experiments that in fact there is a measurable difference in performance because of a degree.
Mr. Michelson said he thought that would be a technical basis for resolution not a technical resolution.
l Mr Coffman discussed the model to illustrate the role of person-nel in a nuclear system. The role of the model is to illustrate that
[by] changes _to one subsystem can affect the other. parts of the model.
l L
He said you can't maximize the output of any one component and still optimize the output of the system.
flinutes/ Human Factors 5
Subcom.Mtg. Jan. 26, 1989 Dr. Remick noted the model doesn't add anything to the report.
Mr Coffman. discussed the relationship of the general model to the program elements.
Five program elements were listed (1) personnel performance measurement (2) The personnel subsystem (3) Human - System Interface (4) organization and management (5) Reliability-assessment.
Each program element had a small graphic display to illustrate the concept.
He said the slides illustrated how the progrem has.been divided into its elements. The organization and management issue was por-trayed by the symbols of two figures.
Dr Remick said he didn't understand what was portrayed by the two' figures.
He noted he didn't think it contributed to what you are trying to do.
Mr. Coffman discussed the program elements.
o The personnel subsystem is involved with training, quali-
- fication, staffing, shift scheduling, overtime, and vigilance o
The human system interface involves the work on displays and controls, the work on advanced instrumentation and controls, including expert systems and procedures o
The organization and management program element is develop-i ing a technology to guide the evaluation of organization, management and climate l
o The personnel performance measurement element focuses on the I
1
Minutes / Human' Factors 6
Subcom.Mtg.'Jan. 26, 1989 4
root cause. analysis determination and to provide the means for storing and collecting and analyzing data.
o The reliability assessment program element is an attempt to-integrate both human and hardware reliability for the purpose of looking at the overall system performance.
In response to a question.as to the use of personnel subsystem, the response was that is human-factors terminology.
Dr. Ryan ex-
- plained when'they talk of personnel subsystems, it is a man-man consideration.
Mr. Michelson asked about the progress in developing models-for management in attempts to get some kind of quantification or some way of judging good managed organization.versus poor.
Dr. Ryan responded that Brookhaven, three universities and a number of people are working on this issue. He noted that they have-been-trying to extrapolate the data from other venues into the nuclear area.
The program deals with:
o Candidate approaches o
Instruments for gathering status information on the orga-nization and on managers o
Development of indicators of team performance and overall
=
= _ - _ _ _ _ _ _ _ _ _ - _ _ _ _ - _ _ _ _ - _ _ - _ _ _ _
L
-Minutes / Human Factors 7
Subcom.Mtg. Jan. 26, 1989 safety performance o
Attempts to develop information to support regulatory:
decitions.
.Dr. Remick asked if the models the staff. is developing would be useful at both the licensing stage and the operating stage?
l Dr. Ryan responded that they plan to start developing a data base. Using the data base t'ey then will evaluate plants both at the licensing stage and operating plants.
It in hoped that the model will be useful for both cases.
Dr Ryan noted that they had been directed to try out their model at a non-licensed facility and they were going to try it out at EBR-2.
Dr. Remick noted that a facility like EBR-2 is largely driven by research needs in contrast to the generation of electricity.
Dr. Remick suggested the staff try out the management model at a successfully operated fossil plant.
Dr.-Ryan agreed that would be a good approach.
He said if the EBR-Z review does not work, serious consideration to a review of a fossil plant should be considered.
l
Minutes / Human Factors 8
Subcom.Mtg. Jan.- 26, 1989 Dr. Ryan discussed the steps in the model development. The first step is.to define the research issue. That is the issue identifica-tion step. The next step is to do.the necessary research to develop the technology or data. The next step is to bench test or field test it. The next step is.to determine practicality.- How much does it cost, what resources are needed. The acceptability is then de-termined. Will the intended users accept it. The model is reviewed to see if it is useful. To what degree does the prototype respond to the original issue.
The' lost step is to have people selected from the user community actually use the prototype with assistance of the research people.
Dr. Ryan discussed the ongoing work on organizational effective-ness and ways for distinguishing between individual and organizational performance.
He noted that they had developed NOMAN, nuclear orga-nization and management analysis model. This is a descriptive technique for evaluating nuclear organizations and management during normal operations.
Both Mr. Michelson and Mr. Carroll expressed interest in seeing a paper describing NOMAN.
[ Note - Dr Ryan told me that a paper describ-ing NOMAN will be coming out shortly and he assured me that ACRS will be on the distribution list.]
Dr. Ryan said that right now NOMAN is for normal operation.
He noted that they are starting some work to extend the model to an accident situation.
He remarked that they were interested in two
. Minutes / Human Factors
.9 Subcom. Meg. Jan. 26, 1989 periods of the accident situation. They were interested in the period when the emergency configuration stabilizes and in the transition period between going into an emergency configuration and when the situation stabilizes.
He noted that it is suspected that during'the transition that the crucial things may happen.
Dr. Ryan discussed reliability assessment.
He noted this program has been going on since 1982. The focus of this program is probabi-listic analysis.
It is used to support reliability and risk assess-ment. This program has a multi-disciplinary research element. The work is to consider both the human and hardware sides of the work.
The aim is to get a better estimate of the relative contribution of the human and hardware combination thereof to reliability and risk.
Dr. Remick asked if they look at human action where the human action helps to stop a transient or accident?
Dr. Ryan said they concern themselves with errors that are made without consideration to recovery and they also consider recovery in some cases.
Dr. Ryan noted work by the National Academy of Sciences. They have a panel to study the mechanisms of human error.
Dr. Ryan mentioned work done by Ed Fleshman of George Mason University. The people at George Mason University are developing a L._____________
__ _ _ _ _ _.______ _ _ _---- _ -__ _ _ _ _ _ __ _ _ o
Minutes / Human Factors 10 Subcom.Mtg. Jan. 26,-1989
. set of criteria to equate tasks performance in the nuclear industry with other industries for which data exists.
He noted they had presented a paper at the Light Water Reactor conference.
Dr. Remick indicated he'would be interested in seeing the paper.
Dr. J. Persinsky - personnel performance measurement element.
He noted'the purpose-is to find the root cause of human error.
He said that they are looking at what parts of the system contribute to human error. ' He pointed out the environmental concerns of physical distraction, heat, temperature, noise.
He noted there are problems with performance that are people related. One of these problems is organization and management.
All this data will be developed into some sort of data base and it would serve as the raw data for reliability analysis.
Dr. Persinsky noted the existence of the investigation protocol.
A draft human performance investigation protocol has been developed for IIT investigation.
It has not been used as yet.
y Dr. Persinsky mentioned the HPES which is operated by INP0. He noted that they believe the training of people who use the system is very important.
He said the staff would like to get involved in developing a training program for them and eventually turn it into an inspection module so that people who do incident investigation can use 1
L-___.-_-__________.
Minutes / Human Factors 21 Subcom.Mtg. Jan. 26, 1989 L
-it on a routine basis and provide the staff with a standardized form 3
![ ',
of this data.
He also noted the staff would like to be able to look at the feasibility of pulling all the data that exists under this system into a single system so that it could be accessed to look for the numbers and causes of human error.
Dr. Persinsky discussed the personnel subsystem.
He noted that the staff wants to understand the factors that affect peoples' behav -
ior and to develop the regulatory guidance to reduce the influence of those factors.
Mr. Michelson noted that there isn't much'of a data base of human response during accidents.
He asked how do you correct 'for that sort of thing?
Mr. Coffman responded that they are working with the people who are doing the accident management research. They are trying to define strategies for managing severe accidents. Those strategies set task requirements on the individuals.
Dr. Persinsky noted that the personnel subsystem would focus on people, their training, their qualifications, level of stress, their capabilities and limitations, how they take actions with the machine, l
and the influence of the environment on the people. This would l
i
Minutes / Human Factors.
12-Subcom.M29. Jan. 26, 1989 include the development of response to stress produced by disturbance, noise, and earthquakes.
He remarked that most of this work in the past has been done on licensed operators. The staff wants to expand this work to the qualification and training to other people in the nuclear industry as well as the other industries that are regulated by NMSS.
Dr. Remick noted that security was not listed.
He asked if they didn't play any important functions?
Dr. Persinsky said they could add security to their. list.
Dr. Persinsky said the major tcpic areas are training, staffing, qualifications, team performance and shift scheduling.
He noted that there was some effort in training prior to 1985.
This was ended primarily due to the effort in conjunction with INP0 on the accreditation program. He noted that there isn't a good n'easure of how effective that training is.
Dr. Persinsky noted the human system interaction element which involves advanced control room persons, artificial intelligence and other types of job performance issues.
Regarding the staffing qualification, Dr Persinsky noted the existence of a rule that requires a speicified number of licensed
Minutes / Human Factors 13 Subcom.Mtg. Jan. 26, 1989 operators in the control room and that is the only staffing require-ment that exists.
In the area of qualification the staff has done I
some work concerning licensed operators and STAS.
Dr. Persinsky mentioned evaluation changes to the_ licensed operator exam process as part of the qualification element.
He also noted the degree rule for licensed senior operators as part of the qualification element.
Dr. Persinsky discussed the team performance element. He noted that he thought this started out from a licensed operator exam and expanded in terms of how do you evaluate the team that is performing.
He noted.the staff has been asked to develop some evaluation criteria.
He said they are looking at ways to improve team performance in emergency situations. He remarked that this element also includes shift scheduling, shift rotation, and vigilance.
Dr. Persinsky noted research being done at Circadian Institute for Physiology in Boston. They are looking at a study of 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> shift versus an 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> shift. He also noted a study of PNL that is attempting to derive data in terms of accident data and correlate it where possible with the working hours of individuals, who were involved in that accident.
Dr. Persinsky discussed the human - system interface.
He said this element looks at the interface between the hardware and the individual operating the hardware. This program looks at control
Minutes / Human Factors 14 Subcom.Mtg. Jan. 26, 1989 rooms, job performance aids including things like artificial intelli-gence and expert systems, and procedures.
He noted that information that is brought in from other sources such as technical specifications would be included in the element.
Dr. Persinsky noted the human factors controls and instruments-tion issue involves a-survey to find out the status of controls.and instrumentation and working on some prioritization in that area.
He noted that the human - computer interaction involves the development of guidelines for future use.
Dr. Persinsky mentioned the procedures element. This is a question of whether or not to upgrade on symptom based or some other
' form for procedure other than E0P's. Should the staff look at proce-
~
dures from other than a regulatory standpoint?
Mr. Coffman made the last presentation.
He noted the Human Factors staff has 12 senior professionals plus two secretaries. He noted that about 50 percent of their experience is with psychology and human factors.
l Dr. Remick recommended to the staff that the human factors research program plan was not ready to present to the full Committee and advised against it; however, the decision was up to the staff.
The staff decided to delay the presentation until a later date.
The meeting adjourned at 4:05 pm.
1 1
Minutes / Human Factors 15 Subcom.Mtg. Jan. 26, 1989 NOTE:
A transcript of th; meeting is available at the NRC Public Document Room, Gelman Bldg. 2120 "L" Street, NW.,
Washington, D.C.
Telephone (202) 634-3383 or can be purchased from Heritage Peporting Corporation,1220 L Street, NW., Washington, D.C.
20005, Telephone (202) 628-4888.
i i
lL----
i ACTIONS, AGREEMENTS,'AND COMMITMENTS, ACRS HUMAN FACTORS SUBCOMMITTEE MEETING JANUARY 29, 1989 1.-
Dr. Remick expressed interest in seeing a paper presented at the-Light Water Reactor Conference by George Mason University on developing a set of criteria to equate tasks performance.in.the nuclear industry with other industries for which data exists.
2.
Mr. Michelson and Mr. Carroll expressed interest in seeing a paper describing NOMAN.
Dr.- Ryan said that a paper describing NOMAN will be coming out shortly and ACRS will be on the distribution list.
l I
~,
1