ML20155C946
| ML20155C946 | |
| Person / Time | |
|---|---|
| Issue date: | 04/11/1986 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1507, NUDOCS 8604170190 | |
| Download: ML20155C946 (110) | |
Text
O&ggg UNITED STATES cC NUCLEAR REGULATORY COMMISSION IN THE MATTER OF:
DOCKET NO:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 312TH GENERAL MEETING l
OV LOCATION:
WASHINGTON, D. C.
PAGES: 1 - 54 FRIDAY, APRIL 11, 1986 DATE:
}
30 X0% 30m0Ve I0m ACRSOEice Weaam os. h W g 41 g o 860411 ACE-FEDERAL REPORTERS, INC.
OfficialReporters I
444 North Capitol Street
\\
Washington, D.C. 20001 0
(2023347 3700 NATIONWIDE COVERACE
r CR26461.0 REE/sjg I
UNITED STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION 3
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 312TH GENERAL MEETING 4
5 Nuclear Regulatory Commission 6
Room 1046 1717 H Street, N.W.
7 Washington, D.
C.
8 Friday, April 11, 1986 9
The 312th General Meeting reconvened at 1:40 p.m.
10 11 ACRS MEMBERS PRESENT:
l l
MR. DAVID A.
WARD MR. JESSE C.
EBERSOLE l
13 DR. MAX W.
CARBON j
i MR. HAROLD ETHERINGTON i
14l l
DR. WILLIAM KERR b
DR. HAROLD W.
LEWIS i
15 DR. CARSON MARK l
MR. CARLYLE MICHELCON 16 l DR. DADE W.
MOELLER ll DR. DAVID OKRENT I!
MR. GLENN A.
REED 17 DR. FORREST J.
REMICK DR. CHESTER P.
SIESS 18 MR. CHARLES J. WYLIE i
19 O,
i 20 j 21 22 l
i 23 4
Am Feder:0 Reporters. Inc.
25 i
(~}
PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS FRIDAY, APRIL 11, 1986 The contents of this stenographic transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards (ACRS), as reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above date.
No member of the ACRS Staff and no participant at
(/
this meeting accepts any responsibility for errors or inaccuracies of statement or data contained in this transcript.
o -
.26461.0 2
REE I
i p
1 PROCEEDINGS E
2 MR. WARD:
The first topic on the agenda this 3
afternoon is a report on the Human Factors Subcommittee.
4 There will be a couple parts.
The Human Factors 5
Subcommittee had a two-day meeting March 19 and 20.
6 Attending were members Moeller, Michelson, Wylie and myself 7
and ACRS Consultant Gimmy.
8 This meeting had three purposes.
The subject of 9
the first day was an information session on one called the 10 state of the art in applying automatic monitoring and 11 control functions in the operation of nuclear power plants.
12 We heard from the representatives from Staff and several
[)
13 representatives from the industry.
The second topic was a v
14 review of the NRC's Human Factors program plan by the Staff.
15 I will summarize what we learned on the first 16 and third topics and then a representative of the Staff 17 will review the present Human Factors Progress Plan'for you.
18 At the present time, I do not see a necessity for an ACRS 19 report on any of these topics at this month's meeting.
The 20 third topic concerned the implementation of emergency 21 operating procedures in actual plants.
22 Dan Jones f rom the Staf f will review the status 23 of the Human Factors program plan, the accomplishments for 24 last year and the plans for the coming year.
25 On the first day of the meeting we heard 11
/~N V
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6 4 6
26461.0 3
REE i
1 presentations on the state of the art in automating 2
monitoring and control functions.
In an introductory 3
presentation from the Staff it was stated that the early 4
attitude of both the industry and the NRC toward computers 5
was largely influenced by experience with big mainframe 6
computers and with the unreliability of this type of 7
computer in a process setting.
8 However, in recent years, both the NRC and the 9
industry have become aware that computers specifically 10 designed for monitoring and controlling processes can be 11 considerably more reliable than the earlier experiences 12 indicated, and in fact may be extremely useful tools for
(,,)
13 aiding operators in managing both routine and emergency v
14 operations in nuclear power plants.
15 It is interesting that the SPDS, Safety 16 Parameter and Display System, which has been mandated as an 17 addition to each plant as a post-TMI backfit seems to have 18 served as sort of an introduction for the nuclear power 19 industry to the use of modern process computers.
As we 20 heard yesterday, however, that is not an unmixed story, 21 that some utilities are having problems while others are 22 successful.
23 There was not really a specific requirement by 24 the NRC that the SPDS should be computer based, but the 25 characteristics desired for the SPDS and the availability
(-
V ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-334 6 %
L.
26461.0 4
REE
's
\\~l 1
of process computers, this has led every licensee to use 2
some form of digital computer as tha heart of the SPDS 3
system.
4 However, despite the introduction to the 5
computer age offered by the SPDS, the state of applications 6
of computers in nuclear power plants is'not very advanced.
l 7
We heard a report f rom the Staf f people who had meetings 8
with NASA at the Kennedy Space Flight C' enter.
Originally 9
NASA made very little use of process computers in their l
i 10 ground control systems.
It has gradually, however, through 11 the years, made more and more use until now the ground l
12 control systems at Kennedy Space Center are largely, what
(
r' i I
(wJ you might call, automated and make heavy use of process 13 i
14 computers with, they believe, considerable success.
15 They also reported that in NASA's design of a 16 future space station, they have set aside 10 percent of the 17 total development budget for the development of software 18 and related automatic control systems.
19 We also heard a report from Leo Beltracchi of 20 the Staff who some time ago visited the Savannah River 21 Plant and was briefed there by our consultant, Mr. Gimmy, 22 on the application of process computers in the monitoring 23 and control of the Savannah River reactors.
This approach 24 was initiated about 18 years ago and has gone through 25 several generations of improvements.
First there was only CT k/
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800 336-tM6
26461.0 5
l REE O
1 monitoring, then some close-loop control of the process was 2
added.
The next step was adding scram system safety l
l l
3 circuits based on computers, a backup scram system, and 4
most recently addition of an alarm diagnosis system which I
l l
l l
5 is a form of an expert system.
Application of computers in 6
today's nuclear power plant control is about at the stage l
7 where the Savannah River plant reactors were in about 1970.
l 8
We also heard a report on methodology that was 9
developed for so-called function allocation.
This is a l
l 10 systematic approach for deciding which monitoring and I
11 control functions will be assigned to the human operator l
l 12 and which will be assigned to computers or other automatic I
O 13 c "tr 1 erste e-1 1
14 We also heard a report from Professor Barclay j
15 Jones from the University of Illinois on researcN he is 16 doing in what he calls the process of embedded training.
l l
17 By " embedded training," he means combining the training l
18 functions for a nuclear power plant operator with his
)
1 19 normal operational functions by providing them with a l
20 control system for the reactor that is part simulator.
To l
l 21 most of us, this seemed to be like a radical, perhaps even l
l l
22 dangerous concept, but Professor Jones is looking at it l
l l
23 conceptually and he sees this as a possibility for fully 24 automated control rooms of a future generation.
i i
l 25 We also heard a report from Westinghouse on the O
l l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage fun 336-(M4 l
l 26461.0 6
REE O
1 approach they are taking on advanced control room design
{
2 which is going to extensive use of plant computers and
)
3 automation of the many process functions.
Their approach
)
4 is to move in very deliberate steps and their intent is not 5
use for the full potential of the technology for near-term l
6 designs.
However, the French, with similar plants, seem to l
7 be taking rather radical steps toward quite complete 1
8 automation of the control rooms.
l l
9 We also heard a report from Rom Duffey from EPRI 10 who described the work that is going on at EPRI.
This 1
11 program is all directed toward application to the present l
12 pooulation of plants and not for future plant designs.
He O
13 he seve taet the eti1ittee ere verv ee99 reive e the l
14 research they are doing.
In fact, each element of the l
l 15 research program has what he calls a host utility which is
(
16 particularly involved in nuclear development work.
{
l 17 Finally, we heard a couple of good tutorials l
18 really from Mr. Bill Bertch and Mr. George Niederauer of l
l 19 Energy, Incorporated.
They described in practical fashion l
20 the application that artificial intelligence and 21 particularly expert systems will have in nuclear power 22 plant operation.
They do not expect artificial 23 intelligence to ever give close-loop automatic control of a 24 process.
For example, of the nuclear power plant process.
25 Instead, it would automate certain observation, l O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-33MM6
h=e -264'61.0~
7 REE
-[3-1 interpretation and decision analysis in nuclear power plant 2
operation,. leaving the final decision and actions to the 3
human operator.
4 Next, we heard reports describing emergency 5-operating procedures developed for use at two specific I
1 6-
-plants.
1 7
The first was on behalf of the Hatch Plant of l
I 8
the Georgia Power Company.
They described and demonstrated 9
a flow chart format.
This is a very comprehensive and
-10 detailed approach to emergency operating procedures which 11 uses'large logic diagrams, or flow charts, to provide the 12 initial response to any major plant upsets.
The purpose of
()
13 the procedures under the flow chart was described as being 14 to bound the problem and stop the spread of degradation of 15 the plant equipment or systems or for the process.
Each 16 flow chart ultimately leads through a logic path to a 17~
preferred recovery procedure in what is called end path
-18 manuals.
19 The second type of procedure we heard about was 20 the so-called single-column instruction format.
These 21 procedures were described to us and demonstrated as applied 22 at.the Browns Ferry Plant of the Tennessee Valley Authority.
23 As you have heard before, the NRC has required 24 that each plant develop for itself some sort of emergency
- 25 operating procedures which are symptom oriented rather than (9
\\/
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationmide Coverage 800 336-6646
26461.0 8
REE
,~
1 event oriented.
The NRC just provided general guidelines 2
and specifies a process for development of the procedures.
3 The result is that there is a very wide variety l
4 and spectrum of styles and types of procedures in use at I
5 the many operating plants in the country.
The two that we 6
heard described may not be at the extreme ends of the 7
comprehensive versus simplicity spectrum, but they are
)
8 probably representative of something close to the ends of I
l 9
the spectrum.
I I
10 So far as which type of procedure is superior, I l
I
{
l 11 don't know, and I don't think anyone in the NRC Staff has l
i 1
l 12 taken a particular position on this.
Certainly, each of l
l s
i x -)
13 the utilities believes that their system of procedures is i
j l
14 adequate.
Whether they think their own system is, in i
l l
15 either case, the best approach that can be taken, I don't j
r 16 know.
l 17 The Staff approach to emergency operating
{
l 18 procedures has been to insist that each licensee develop a 19 process for developing its own EDPs and that the NRC Staff l
20 then would audit and monitor that process with the 21 expectation that if the process was going well, good 22 emergency operating procedures would somehow appear in the 23 plants.
24 The NRC seems to have reached the decision that 25 this approach has failed.
They have monitored the process
()
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 80M364M6
W 26461.0' 9
+
-REE'-
1
.and oftentimes found indications that there was a 2
-satisfactory process, but when they audited actual
'3,
_ procedures, they found major problems with the 4
all-important end product.
Apparently the Staff recognizes-5 the-need for more extensive auditing program of the actual 6
l final product, the emergency operating procedures.
7-MR. KERR:
Excuse me.
Did they ever audit one B
8 where'they_didn't find major problems?
9 MR. WARD:
I think so, yes.
10 MR. SIESS:
Two.
11 DMR. WARD:
I think I had mentioned to Jesse 12 yesterday that --
[~)
13 MR. EBERSOLE:
I wrote a letter.
N_-
14 MR. WARD:
There seems to be kind of a parallel 15 situation, the approach that the Staff took to encourage 16 utilities to develop EOPs and the approach that was being 17 taken to develop the SPDS.
And kind of by accident we 18
. heard about each of them in the last couple days.
There 19 seems to be -- perhaps the results have been somewhat 20 similar, uneven.
In some cases, the SPDS or the EOP in the 21-plants seem to be good and solid and work.
And in other 22 cases, they seem to be rather poor.
23 It may be that the committee would like to 24 comment on that.
25 MR. EBERSOLE:
That may be in your hand shortly, O
ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6M6
r y,
.a 126461.0 10-
~ REE
'd 1
if you want to talk about'it.
~
2 MR. REED:
I' might point out that certainly 3
there are emergency operating procedures from-the~beginning 4
of time,-and I remember algeneration of those for Yankee 5
'Rowe in about 1959.
In fact, if you get with one vendor, I 6
think you will find the tendency that.first a reaction, 7
Westinghouse, Yankee Rowe, those procedures went to-8-
proliferate.
It ~ is nice to copy somebody else's original 9
creation work.
10 So there were emergency operating procedures and 11 we should understand that.
After Three Mile Island and 12 because of those procedures, things didn't work out.
There
/'N
_(_)
13 was a great move on to get symptom oriented, more 14 standardized procedures.
And-there were groups formed and 15 I am a little surprised to hear you almost indicate that 16 every utility comes'out with something different.
17 But I know of one group of nine people from 18 Westinghouse that worked together.
Westinghouse first 19 created the generic EOPs.
These people worked together to 20 break them down and to get them-the best style and the most 2;1 -
concise and written.
I would have to believe, based on 22 what I saw, that a number of these emergency operating 23 procedures are pretty good.
24 MR. EBERSOLE:
The first generation, though, of 25 them, Glenn, they stopped abruptly when the system failed.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
pp,
[4
- -E26461.0'.
~ _. --
~~~
11 REE.
I'/
1
~ When system A ' failed, they_said, start system B.
You 2
always had an alternate path.
t 3.
MR. REED:
I don' t know that that is --
4 MR. EBERSOLE:
TMI-2 was the catalyst,.which 5
produced.the evolution beyond start system B.
Tell me what 6-
'I.do when I have both systems out? 'That was necessary to 7
have_TMI to get that extension of the EOPs.
8 MR. REED:
Well --
9 MR. WARD:
There were vendor oriented owners' 10 groups formed that developed the TPGs, the. technical 11 process guidelines for procedures, but then each utility
_12
.had to take those and make some actual plant operating
()
13 procedures.
14 MR. REED:
They had to be translated.
15 MR. WARD:
I think that is what has been done 16 unevenly.
17 Before we went to, when we met with the GPR, one 18 lof the things I did was -- this may be a little superficial 19
-- but I sampled procedures from eight different plants and 20 I kind of came to the same conclusion, that there was a 21 great variety.
Some seemed to be pretty good and the 22 others, as far as I could tell, maybe weren't too good.
23 MR. REED:
There.has got to be a variety.
Each 24 plant is its own version and it is specific.
i 25 MR. WARD:
That is right.
They are not l
l l
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Natsonaide Coserage 800-33&6M6
26461.0 12 REE 1
standardized, which is maybe -- if they are all good, who 2
cares if they are itandardized.
3 MR. EBERSOLE:
I think there is a natural I
4 process that goes on here.
You send out general policies 5
and general guidelines to the field and then you watch the 6
spread of quality that occurs as the people institute those.
[
7 Then you find somewhere in that field an adequate level and 8
then define it as a baseline for standardization.
If you 9
don't know how to do it yourself, you bleed the field to 10 find out how they will do it and what the range of quality 11 will be.
That is the only way you can use the field at f
12 large to do what you ought to be able to do yourself.
J
.s (v) 13 MR. REED:
Well, after Westinghouse developed 14 generic procedures, they put real people into those things, 15 like SRO shift supervisors were involved in those specific 16 interpretations.
17 MR. EBERSOLE:
There is another problem which is
)
18 the TVA problems.
The designers put this stuff out and l
19 told them how to do it.
That disrupted the operators' 20 hands.
21 MR. SIESS:
Was there any correlation between 22 the ones that did it poorly and the management?
23 f1R. WARD:
That is a good question.
24 MR. SIESS:
I mean, did they do other things 25 poorly?
Was this at a lower level of competence?
L.]
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationside Coserage 800-336-6M6
q
. ~ -
' ^
126461.0-13 1
REE-1 MR. WARD:
You know,lwe didn't really get that 2
from the Staff.
We didn't ask that question.
I don't know 3 ~
- that they could have answered it.
4 MR. SIESS:
What about the ones that did the 5-
'SPDS bad compared to the ones that did EOPs bad?-
Did we 6
correlate on that-basis?
7 MR. WARD:
I bet you 10 bucks we could.
8 MR. SIESS:
Another list of A, B,
C, D.
9 MR. REED:
You have to keep in mind that SPDSs 10 should have been developed more by engineering l
'll organizations than the utilities.
And what they were able 12-to by -- in a hurry, and the procedures, that is
(
13 site-specifics, should have been developed by real 14 operating personnel.
So --
l 15 MR. EBERSOLE:
There is a view, though, that
[
16 says there well may be very competent' operators who use i
17 this instrumentation so well that they won't watch the SPDS.
18 It is intrinsically unreliable.
It is just an adornment'in 19 there unless somebody really pays attention to the design.
20 So they may have identified in the first place as being 21 those dadburn things that the NRC stuck in their room and 22 they don't want it there.
l 23 I would honor that opinion.
I 24 MR. REED:
You might say it is a fifth set of i
25 readouts, but it is supposed to be certain key parameters.
' O I
l ACE-FEDERAL REPORTERS, INC.
I 202-347 3700 Nationwide Coverage 804336-6M6
'26461.0 14 REE s
/
1 That was supposed to be the thing that would cause people 2
to use it.
3 MR. EBERSOLE:
It is supposed to give him a 4
quick glimpse in an emergency before he goes to the real 5
panels.
It is just for a transient quick use.
6 MR. SIESS:
It is just the opposite of the GE 7
cor.'rol room that has all these fancy display screens that 3
aren't reliable and then they have that thing up there with 9
the gauges on it sitting in front of the control panel that 10 they are supposed to rely on.
11 MR. EBERSOLE:
That is right.
12 MR. SIESS:
Remember that?
We saw it at Clinton.
()
13 MR. WARD:
Before we go to Dan Jones and the 14 program, I guess Forrest was there for one day of the i
15 meeting.
16 MR. REMICK:
One day, yes.
17 MR. WARD:
Charlie, you were there.
I don't 18 know if anybody has any comments they would like to make 19 about the other aspects of the meeting.
20 MR. REMICK:
I found some of the NASA 21 inf ormation of interest.
They sure see their use of 22 computers, although it is far advanced to the nuclear 23 industry, they still see that as developing.
It was kind 24 of interesting to see how they started out.
In my notes I 25 have the numbers, but basically one person had g
L]
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage luh3%(646
r.
d6461.0 15 REE y.
\\'
~1 responsibility for -- maybe his responsibility'was to open 2
that valve.
In the whole startup he'might have-had 3
something like four or_five functions.
Then they went over 4
to that by use of computers that he had kind of a. system 5
that he could be. responsible for.
6 It was interesting to hear how NASA has 7
developed.
We think of them as being highly computerized.
8
- 7. hey ara.
But yet they have a long way to go compared to 9
what they think they will eventually have.
10 For example, apparently the rockets themselves 11 and so forth are not self-contained from the standpoint. of 12 doing internal testing and so forth.
You basically plug
(
).
13 them in on the ground.
So they have gone a step-by-step, 14 much like I see Savannah River has gone, kind of a step at 15 a time.
16 MR. OKRENT:
I think you mentioned that there 17 were two speakers from Energy, Incorporated.
I would ~ like 18 to get whatever material they supplied.
I would like to-19 get copies of it all, please.
20 MR. WARD:'
Anything else?
21 Let's ask Dan Jones, then, to give his report on 22 the human factors program plan.
23 MR. JONES:
When I started on this human factors 24 program plan, roughly three years ago, it was said that 25 this was going to be a three-year project.
Well, the three ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
~"26461i0 16
~
-REE-r
.p.
II years is up.
The plan has changed considerably.
- Remember, 2
_in the original plan there were seven program elements.
3
~One of_these is human reliability.
I will not talk-about 4
it today because of'some other reasons.
But it is doing 1
. - 5.
. well, too.
6 staffing and qualifications area, let me review 7
first what we did in 1985, last year.
I would like to' walk 8
you through what we are planning to do in 1986.
9 In -general terms, we had about 33 TMI action 10 items,.all.but six of these have been completed.
Four of 11 those six remaining are essentially complete and the other 12 two are long-term.
The human factors society also had some
. I) 13 51 recommendations; four, five of those recommendations _we 14 had some concern with.
Three of them we said don't apply 15 at all.
And of the remaining 46 that we did accept and 16 work on, essentially all of them have been implemented at 17 this time.
So-we have done quite a lot in the past three 18 years.
19 We decided there wasn't any need to license 20 additional personnel besides licensing the. operators.
We 21 looked at the operational qualifications or educational 22 qualifications for operational personnel in two studies.
23 One of these for NUREG-4501 was an assessment of job L
j 24 related qualifications for nuclear power plant operations, t
l 25 And the second assessment was of specialized educational i
l l
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6M6 i
1
^
~26461.0
17 1REE 1
3
'N 1
programs for licensed. operators.
2 So.what we did in both of these was, first, take 3
the INPO job task analysis, decide what parts of that task-4 analysis.were educationally related, job knowledge 5
requirements, and then compare that with educational 6
qualifications of people involved; and it turns out that as 7
was pointed out this morning by Mr. Reed, about half of 8
everything that an operator has to know is plant specific.
9 So there are certain general background materials.
The 10 other half is plant specific.
11 As you talk about the -- this item, the 12 assessment of the specialized educational programs, this
()
13 program, this study we took a look at nuclear engineering 14 courses, other engineering courses, specialized training 15 courses and associate two-year college degrees.
Using i
.16 educational experts and this list of job knowledge, content 17 that we worked out with INPO, we tried to look, to 18 determine what kind of educational levels and what kind of 19 expected knowledges could be obtained by students that went 20 through these three kinds of programs.
21 It turns out that the nuclear engineering, about 22 half of all he needs to know is taught in a nuclear 23 engineering course.
Other engineering courses, it runs 24 about 30 to 40 percent.
The associate degrees run 25 to 30 25 percent.
And the specialized programs run about the same ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nation *ide Coverage 800 336-6646
h
'26461'0-18 REE
, ri)
\\-
1 level.
So that roughly halffto two-thirds of the knowledge f
2.
requirements ~of an operator or SRO or RO have to be plant 3
specifically trained by the plant.
4
.MR.
EBERSOLE:
My question was at TMI-2, the 5
operators did pretty well what they were told to do.
That 6
is, they used'within the range they had been taught 7
everything^they could.
But they hadn't been taught.
They 8
hadn't been taught physically the plant.
They didn't know 9
that looking over here, if they saw something that wasn't 10
- over there.-
11 This says to me you can. beat yourself to death 12 teaching operators.
You have got to' teach the super
< ( ).
13 structure before you teach the bottom structure.
What are 14 you doing about that?
I haven' t heard anything.
15 MR. JONES:
Well, this is part of the program 16 that is being accomplished with INPO in their job 17 accreditation and training accreditation program, and the 18 emphasis that NUMARC is putting upon training of operators.
19 We think it is done pretty well.
20 MR. EBERSOLE:
I am talking about the training 21 of the trainers.
22 MR. JONES:
I would like to come to that in a 23 moment.
I am not going to give you a full answer because I 24 don't have one.
25 MR. REMICK:
You gave certain percentages.
Was C:)
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33MM6
^26461'.0 19 REE-A l 1
that.the percentage of the engineering knowledge operators 2
should have that were taught at the BS degrees or associate 3
degrees?
4 MR. JONES:
Of their engineering knowledge, 5
academic knowledge that we are talking in this case.
6 MR. MICHELSON:
One of the questions that.comes 7
up from time to time is this question of how well does an 8
operator understand accidents beyond a design basis; and 9
also, of course, how well should he.
It is thought that at 10 that point you are depending more on education and less on
~11 specific training, since this is something that has
'12 happened now that hadn' t been considered in the normal
()
13 training process.
So what is your view or feeling about 14 the adequacy of the operator training presently, at least, 15 relative to his ability to cope with beyond-design-basis 16 eventc?
17 MR. JONES:
I am not sure that we know what -- I i
18 am not sure that I know, anyway, what the 19 beyond-design-basis event is going to look like.
20 MR. MICHELSON:
Clearly, that is the case.
21 MR. JONES:
That is a bureaucratic dodge of your L
22 question.
23 MR. MICHELSON:
I think that is a crank answer.
I 24 Has the Staf f given any thought, though, as to l
25 what would constitute a training program appropriate to l
l ACE FEDERAL REPORTERS, INC.
t i
f 202-347 3700 Nationwide Coverage 800 346M6
b 264J1.0-20 REE D,
preparing an individual to face these hypothetical
'/
1 2
beyond-design-basis accidents?
Or is it even in -- is it 3
even in your thought process to consider it.
4 MR. JONES:
Yes, it is in our thought process to 5
consider it.
-It is one of the items that we are looking at.
6 MR. MICHELSON:
So it is in the future now?
l-7 MR. JONES:
It is in the future.
8 MR. MICHELSON:
And we will hear about it maybe l
9 in a year or so?
10 MR. JONES:
Or more.
11 MR. REED:
An operator is supposed to operate by 12 his emergency operating procedures.
.()
13 MR. MICHELSON:
I understand, but things have l'
14 happened that weren't envisioned when emergency operating 15 procedures were written.
16 MR.' REED:
So he doesn't have anything written.
l l
17 MR. MICHELSON:
Yes.
And I just wondered, how 18 do you prepare a person, if you even can, for such a I
19 situation.
I wanted to get his thoughts on that.
20 MR. JONES:
I have some personal ideas, but --
21 MR. MICHELSON:
I am not necessarily looking for 22 policy.
I was looking for your --
l 23
-MR.
JONES:
I think that there are ways that f
24 these things could be trained.
i l
25 MR. MICHELSON:
It might be we need to put it on l
l ACE. FEDERAL REPORTERS, lNC, l
202 347-3700 Nationwide Coverage IWn336-6M6
v
~
26461.0 21 REE D
1 the subject some time for a human factors meeting, just to 2
get the Staff to tell us their thoughts on such a matter, 3
because it keeps coming up around this table.
4 MR. REMICK:
There is a thought I would like to 5
make, there is specified training for STAS which includes 6
this type of thing.
And most, if not all plants, SROs 7
usually sit in on the same courses as the STAS do.
I can' t 8
tell you specifically what it is.
9 MR. MICHELSON:
I would like to hear some time 10 what is the philosophical training approach to preparing 11 people for that situation.
If it is being done for STAS, 12 then I think it would be worthwhile to hear about it.
f >')
13 MR. REMICK:
It would be good to have a couple s
14 utilities in to tell us what they are doing with that area.
15 MR. MICHELSON:
I am thinking of accident 16 litigation beyond the normally postulated accidents.
17 MR. JONES:
He have obviously worked on reg 18 guide 1.78 which is part of a rulemaking package.
We are 19 concerned with crew performance in a simulator, supervisor, 20 SRO and STA.
Research did a study using the instrumented 21 simulator down at TVA.
It turns out that in the crews that 22 were tested there, there essentially was no difference 23 between performance between crews that were supervised by 24 an SRO or by an STA.
And this was on BWRs.
He found 25 similar results on PWRs.
It really isn' t isn't surprising (m
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmcrage 800-336-(M6
26461.'O 22
'REE-lD N I.
I when you consider the training and level of these people.
2 MR. MICHELSON:
I am a little confused here.
3.
.Are you dealing now with measuring error rates and so forth
.?
4 for standard exercises?
5 MR. JONES:
Yes.
6:
MR. MICHELSON:
These are mostly standardized?
7 MR. JONES:
Yes.
8 MR. MICHELSON:
You are saying for that kind of 9
an atmosphere there is not much difference?
10 MR. JONES:
Yes.
11 MR. MICHELSON:' Thank you.
This wouldn't 12 reflect necessarily on how they might handle a
() beyond-design-basis situation at all.
14 MR. JONES:
No, it would not.
15 Also, we looked at the limits and conditions of 16 shift work.
The NRC has fairly liberal requirements for 17 overtime and shift work.
And so we went,.did a thorough 18 review of the industry and the scientific literature to try 19 to determine basically what, if anything, we should do 20 about working on shif t work, putting limits or changes on 21 our current requirements or simply reinforcing what we L
22 currently have.
He also looked at the problem of 12-hour 23 shifts.
24 Some of the utilities have gone to 12-hour j
25 shifts.
At this point in time we have no guidance on l
l l
l ACE. FEDERAL REPORTERS,1NC.
202-347 3700 Nationwide Coverage 800 33MM6
(
26461.0
-23 REE
-1 12-hour shifts.
We-don't have any objection to the concept.
2.
It is just that there may be, we want to make sure P
3 information or guidance on 12-hour shifts matches the 8-hour 24 shift.
So we looked at this.
That will be one of the 5-things we will come out with'this year.
6 As part of the.-- one of the TMI' action items 7
was a. requirement to get feedback from operators.
We had a 8
couple of. operator sessions in which operators were invited 9'
in and discussed their problems with the Staff and with 10 other individuals in person.
But that is a rather 11 expensive way to go, so we went out with a mail survey and 12 used classical and psychological sampling techniques and
{)
13 took a look at what operators thought.
We found that this i
14 was a very effective, reasonable way to get information at l
15 a much lower cost than inviting individuals in for a big 16 meeting.
17 Incidentally, we also looked at some of the 18 current attitudes towards such things as STAS, shift work, 19 staffing levels, and we -found that most of the operators, 20 for example, they didn't like shift work as far as shift 21 work per so was concerned.
But they liked the additional 22 pay that came with unusual shifts, overtime, that. kind of 23 stuff.
24 As far as STAS were concerned, most of them 25 thought the STA was a waste of time.
ACE FEDERAL REPORTERS, INC.
202 347-3700 '
Nationwide Coverage 800-3E6M6
t IV
-2645170
~
1 24
-REE:
L N
1 Interestingly enough --
2 MR. MICHELSON:
Most of who?
L 3
MR. JONES:
Most of the operators thought that
[
r 4
STAS were a. waste of time.
However, if you took.a;little f
5 deeper analysis of this and looked at the educational level I
'6'
~ of the operators, those operators who were reasonably well 7
educated, college experience or even college degrees, L
8 tended-to recognize the value of the STA more than'those l
9 who did'not, whatever that means.
l l
10 MR. REED:
Modern management techniques for the 11 last 10 years have been using surveys with companies and l
12 taking results of them confidentially and giving their t
(
).
13 conclusions back to top management.
I think a thread runs 14 through all of this survey business.
You have to watch for 15-
.a dissident with a loud voice who is always griping about 16 something.
You have to weed out somehow that that's very 17 prejudicial and emotional and doesn't relate really to the l
18 truth.
So I don't know how you do that.
j; 19 MR. KERR:
Take the stuff that you disagree with l.
20 and throw it away.
l L
21 MR. WARD:
I would like the individual opinions L
22 on what Glenn just said.
23 MR. JONES:
On this particular survey, there 24 were two things.
One we asked open-ended questions and we 25 got a good deal of comments.
We also had some multiple l
)
l ACE. FEDERAL REPORTERS, INC.
202 347 3700 Natonwide Coverage 800-336-6646
l L
26461.'0I
+*
~
'25 REE I
A j-1 choice type things.
The interesting thing to me was the 2
. percentage of responses that we get.
A good survey in.
~3 classical survey techniques is 30, 35, 40 percent.
We got
'4 almost 95 to 99 percent return on our. survey.
5 MR. MICHELSON:
How large a sample?
6 MR. JONES:
531.
7 MR. MICHELSON:
You should filter out the cranks --
8 MR. JONES:
That ought to filter out the cranks.
9 If the sample is big enough and the return is 10 large enough, it usually takes care of the occasional i
1 11 dissident.
12 Moving on to training, the Commission published
()
13 a policy statement on training and qualifications.
Along 14 with this went a memorandum of agreement with INPO on the 15 whole accreditation program.
We published a --
16 MR. OKRENT:
May I interrupt?
17 MR. JONES:
Surely.
18 MR. OKRENT:
I don' t know what other colleges do, 19 but at UCLA they have a system whereby the end of each 20 quarter, the -- before the final exam, the students are 21 asked to provide a rating of the professor.
One of the 22 leading luminaries at UCLA, a professor, says, and I am 23 inclined to agree, this is really more a rating of the 24 students than of the professor, because in my opinion, the 25 students may not kaow really what constitutes a good course.
i 1
i ACE. FEDERAL REPORTERS, INC.
f 202-347 3700 Nationwide Coverage 800-336-6M6
l
~26461.0 26 l
REE r~N l
1 They may like a professor who has in fact given less than j
1 2
the full amount, made it all very clear and easy and so 3
forth, as against the professor who has been very demanding 4
and so forth.
So one has to be somewhat cautious in what 5
implications you think arise from your surveys of operators.
6 They may or may not on all subjects be aware of what is the 7
best.
8 MR. JONES:
That is true.
9 MR. OKRENT:
Okay.
10 MR. JONES:
No doubt about it.
11 MR. ZIEMANN:
I am Dennis Ziemann.
Dan, I would 12 like to ask a question, I don' t remember how old that l
,,(,)
13 survey was.
I do remember those results as you represented 14 them.
But is that not fairly old?
And the reason I ask 15 the question, as a result of that, it seems to me as though 16 I know that there has been an interest, we have had an 17 interest in the industry as well in integrating the STA as 18 l a part of the shift crew.
And I think having done that, I 19 think it is being done, they have become better accepted by 20 the rest of the crew.
l I
21 MR. JONES:
This survey was initiated about 18 l
22 months ago.
So -- at least 18 months ago.
And please 1
23 don't hold me to a date.
I know it is at least that far 24 back.
It may even be two years.
We received the earlier i
25 results on this last winter, early winter.
l
\\J
\\
l ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage
- n33MM6
26461.0 27 REE 1
So we get into the training business; one of the 2
TMI items was to look at plant skills.
He published a 3
NUREG on how to conduct team training or team skills 4
training for the benefit of guidance of the industry.
It 5
is not a requirements document.
It is simply a NUREG that 6
says, if you want to do skills training for teams, this is 7
a good idea for doing it.
8 We looked at instructions.1 skills evaluation.
9 This is the question that was raised:
What do you do about 10 the capability of the instructor?
This NUREG was designed 11 to enable the NRC Staff to evaluate the ability of the 12 instructor as an instructor, not on the subject matter but n
(~,))
13 his ability as an instructional technician, if you will, or
{
(
14 college professor.
15 And then we developed, part of our training l
16 evaluation program, training review criteria and procedures 17 for the NRC Staff for us to use when we went out and 18 evaluated the training of utility programs.
19 MR. REED:
I would like to hit that STA thing 1
20 again.
Let me make a point in why, perhaps, some STAS are 21 not appreciated.
22 You touch a bit on the issue that the lower l
23 level non-college people perhaps didn't appreciate the
{
1 24 STAS' level.
I have been concerned ever since the NRC came 1
25 out with its requirements for an STA as f ar as formal l
(~)
{
\\>
- I 1
ace-FEDERAL. REPORTERS, INC.
202-347 3700 Nationwide Cmerage 800-33MM6
]
I
+
26461.0 28 REE-l 1
. degree requirements are concerned.
I always felt that one 2
.of these days we will have aging shift supervisors who have 3
served for 10 years, let's say, have been SROs'with fine 4
records.
I feel those people ought to be'qualifiable as
'S STAS.
What is being done here is cutting off a career path.
6' I really think somebody ought to reflect on highly 7
qualified, very good record, old, aging shift supervisors.
8 MR. JONES:
A point to consider.
L 9
If we look at licensing examinations, I have to 10 apologize, in your handout that you have there,.this'is the-j 11 last two items in the handout, not where it normally is.
12 It got misfiled in the -- I am sorry.
That is in the next
()
13 section.
I take it back.
This is okay.
14 What we got in the examination process was the 15 rulemaking on the licensing rule.
And this was a major 16 item this year concerning the regulatory guide, personnel 17 qualifications, medical evaluations and simulators.
This 18 rule is part of the 306 package that originally came out 19 several years ago and is actually going to be published 20 very shortly.
i 21 MR. REMICK:
What is the status.
Is it still 22 before the Commission?
We reviewed it four or six months 23 ago.
24 tiR. JONES:
Do you know what the status is on f
25 exams?
ACE-FEDERAL REPORTERS, INC.
(
202 347-3700 Nationwide Coverage 80lb336 6M6
r 26461.0 29 REE 7
1 MR. ZIEMANN:
Yes.
The degree rule?
2 MR. REMICK:
No.
Part 55?
3 MR. JONES:
The exam rule?
s 4
MR. ZIEHANN:
It has not yet been sent to the 5
Commission.
6 MR. JONES:
It is due there in April.
7 MR. REMICK:
Can we expect that that will bo 8
different than what we saw?
9 MR. JONES:
No, sir.
It is basically the same 10 thing that you have seen before.
I don' t see much reason --
11 The next item we have been working on is to 12 improve the whole simulator examination procedures.
He lG) 13 have been developing a series of scenarios which can be 14 used by examiners on simulators to develop reasonable and 15 reliable simulator examinations.
16 Then, of course, there is the whole business of 17 the simulator facility exam.
As you know, this rule 18 requires in it that operators be examined as a team and not 19 just individuals and that this examination be conducted on 20 an appropriate simulator or other facility.
And thus we 21 have to have some sort of criteria to evaluate what that 22 simulator is.
That is the process for that.
23 As far as the exam content is concerned --
24 HR. OKRCNT:
Will that be a publicly available 25 set of scenarios?
10 v
ace FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage MX)-33MM6
26461.0 30 REE
'l 1
MR. JONES:
Of the scenarios, yes, sir.
2 MR. OKRENT:
So you could take a course like 3
students do for boning up for admission to law school or 4
medical school.
5 MR. JONES:
Absolutely.
6 We have a PWR knowledge catalog which has been 7
published, NUREG-1122, which is an evaluation of the entire i
8 job task analysis done by the INPO analysis and then 9
compiled in sections, depending uport knowledge of 10 particular sections of systems itself and so forth, so that 11 you can relate the content of what that task analysis is to 12 the various subsystems and equipments and the functions
()
13 that the operator has to do as part of his operational job.
14 Then combine that with the examiner's handbook which has l
15 been published which is an instruction to the examiner as 16 to how to write exams and how to sample and get the proper 17 content in an examination.
18 Go to the PWR Knowledge catalog and get the 19 information out of that, follow the procedures in the exam 20 handbook and we hope we have a better examination.
21 Certainly more job related than some in the past might have 22
- been, i
23 The third item in this is to take all the 24 examination questions that we have records of, put them in l
25 a computerized examination question bank which is located (s
v l
ACE-FEDERAL REPORTERS, INC.
t 202-347 3700 Nationwide Coverage sk346M6
3=
26461.0 31 LREE-
,0
'\\/
1.
fat the Idaho National Engineering Laboratories and 2
determine how good these questions are, track them, get 3
information on their performance, their validity and 4.
reliability as examination questions, and make those 5
examination questions, some 30,000 of them, available to 6
.the public, if they want them.
7 And, of course, as new examination questions are 8
written, they will also be added to the bank, and as we 9
find exam questions that don't do a good job, they will be 10 ptil i e out.
This process of maintaining a computerized l
11 examination question bank, we think, will be a better way l
12 of improving our examination content and process and of
()
13 being given more valid reliable exams.
14 Each of the regions, regional offices now have 15 computer access to that exam bank.
He would hope that if 16 an examiner is going to prepare an examination, he will sit 17 down at his computer terminal, decide what particular 18 subject matter he is interested in, query the exam bank as 19 to what kinds of questions' pertaining to a particular 20 system or function there are, and decide which questions 21 are appropriate for him to use or modify or change; and if 22 he decides to right write a new examination question or 23 change one that is in there, he may do this independently.
24 Once the question is standardized and worked up, it will go 25 into the examination also.
.O ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 1100-336 %M
26461.
32
-REE x-1 And once again, if overybody knows'all'30,000 2
questions and the answers to them,-we are happy.
3 MR. WARD:
Dan, how do you know people aren't I
4 using the same old 200 questions out of that 30,000?
5-MR.-JONES:
That becomes a function of our-audit
-6 procedures at the Staff level to make sure that that 7
doesn't' happen.
I am sorry I can't give you the exact 8
details of how this works.
Maybe you can, Denny?
l 9
MR. ZIEMANN:
No.
f 10 MR. JONES:
What happens is an examination is j
l
-11 written by an examiner and that exam is then reviewed by 12 his supervisor and other examinors,, and even by our staff
(
)
13 at headquarters if necessary, to make auro that that l
14 doesn't happen.
It is a quality control audit process.
15 MR. MICHELSON:
In the discussions we had.with l
16 the commissioners this morning, there was an inference made 17 that PWR questions show up on BWRs and vice versa.
How l
18 could that possibly happen if it goes through soveral steps j
19 of examination review process before it over gets to the 20 site?
I I
21 MR. JONES:
I think that is unlikely that it has q
22 happened in recent months.
23 MR. MICHELSON:
It was cited.
Unless I don't 24 rocollect, I thought it was cited that people wore askinq 25 boiling questions on pressurized water or vice versa.
Acn FEDERAL REPORTERS, INC, 202-347-3700 Nationwide C;werage
, m>336-6646
i l
26461.0 33 REE 1
l O 1
MR. ZIEMANN:
I don't know the answer to that 2
question.
It does not mean it hasn't happened.
3 MR. MICHELSON:
I heard it first this morning 4
myself.
l 5
MR. ZIEMANN:
Another part of the quality 6
control that I think is very effective is during the time 7
the license, the examination is being administered, the 8
applicant or the licensee has the opportunity to review the 9
questions also.
l 10 MR. JONES:
At the utility?
)
l 11 t1R. ZIEMANN:
Yes, and to comment on those that i
12 may --
)
I I O 13 ria ri'cac'so":
"s t"
t "er <e eter <"e t
14 exam was administered?
l 15 MR. ZIEMANN:
They actually do it during the l
l 16 examination.
l 17 MR. MICHELSON:
So they will point out to you l
l 18 that this is an unfair question because it is the wrong 19 kind of reactor?
20 MR. ZIEMANN:
Correct.
And their comments then 21 are considered before the exams are graded.
22 MR. MICHELSON:
Maybe I didn' t misunderstand 23 what I heard.
Did you hear that?
24 MR. REMICK:
I didn't hear it but --
25 MR. REED:
It was as you said it.
O ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nation *ide Coserage mb33MM6
26461.0 34 l
REE I
\\
l fir. JONES:
I will make note of it and that is 2
one of the things I am going to talk about as soon as I get I
3 back.
4 MR. MICHELSON:
I had a higher level of l
l 5
appreciation for the process than the subcommittee.
{
6 MR. REED:
When you get into this human realm, 7
people tend to pick up those extremes.
I think we have to; l
8 I think that operators deal in extremes.
9 MR. JONES:
I do know t' hat it is absolutely not I
10 true that we never change the questions, just the ans.wers, j
11 (Laughter.)
I 12 MR. REMICK:
I assume you still give the utility I
h 13 an opportunity to review the answers if they differ with 14 them at that time.
1 15 MR. JONES:
Yes.
l I
16 MR. 11ICHELSON:
But it is a pretty closely 17 controlled process.
This is not just an exam that the l
1 18 examiner dreams up, it is out of a question bank, filtered l
l l
19 to some extent already.
And then it is read and reexamined l
1 20 by the supervisor and further examined by the industry.
l 21 MR. REf1ICK:
You ought to hear from the guy who 1
22 writes these exams what he thinks about all these levels of
)
23 review.
I 1
24 f1R. JONES:
They don't like it.
l I
25 MR. OKRENT:
There is a written answer to every ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationside Coverage 800-336-6646 1
~
n.
i; :
'26461.0 35 REE i
1
-question?
2 MR. REMICK:- Yes.
3 MR.' JONES:
As far as the procedures sections 4
are concerned, we looked at the emergency operating
.5-procedure upgrade and post-implementation audits.
That was 6
discussed just before I came on here a few minutes ago.*
I 7
The standard review plans for operator.and 8
, maintenance procedures was finished up and published.. He 9
developed an EOP inspection module for use by I&E 10
~
inspectors and we also looked at. methods for 'ev~iluating 11 alternative techniques and formats.
-12 Now, which kinds of procedures are most
(
13 effective?
It turns out it doesn't make much difference, 14 as long as the procedures themselves are well designed and 15 properly engineered.
In other words, the type of format 16 that is used is not nearly as important as the fact that l
17 the procedure is a well writt'en, content-valid procedure.
]
'f 18 MR. MOELLER:
Back on the questions.
You may L
19 have discussed it and I missed it, but do you give or don't 20 you have questions that.you find are -- that have several 21 interpretations; and do you allow for that?
Or I guess you 22 later try to correct that?
23 MR. JONES:
We try to correct for it so that 24 they are unambiguous.
The better designed the question, 25-the less ambiguous it is up to some point.
(I q
1 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 i _.
26461.0 36 REE A
.f 1-MR. REED:
It occurs to me, I remember when you 2
were talking about the question bank, but it occurs to me 3
when you'have gotten into these numbers, this may be by 4
intent, that nobody out there-is going to waste his time
.5 studying to answer questions.
What the training manager 6
must now do, he must train with a program that covers the 7
median material and forget the questions.
That ma'y be by 8
intent.
l l
9 MR. JONES:
That is by intent as far as we are
]
{.
10 concerned. ' Understand --
t 11 MR. REED:
Then when you are through and people'
[
12 f ail -- a lot of people f ail at a f acility -- right away 13 the NRC people know, then it is the program that is 14 deficient, unless there is some sloppiness in the 15 instructing or grading or in the attitude or motivation of~
16 the candidates.
i 1
(
17 MR. JONES:
There isn't any doubt that the.
l 18 examination process is certainly a reasonable. check on the l
19 ability of the training program to accomplish the mission.
l-20 Wo looked at the man-machine interface area.
We l-L-
21 identified deficiencies.
And there are significant h
[
22 variations from a human engineering point of view as to the 23
-- as to how local control stations are used, operated and l
24 laid out.
This will be one.of our major items to look at l
25 in 1986.
O ACE FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coverage 800-336WA6
c:
^
26461.0 37 REE m
1 Annunciators, we did a study conducted on
.2 computerized annunciator systems, what kind of techniques 3
should be used, what the NRC should-look at as far.as
~4 regulation in this area is concerned.
5 We also looked at alarm reduction techniques at 6
Sandia National Laboratories.
7 MR. REED:.One thing that bothers me on that top 8
item, the deficiencies in the local control stations.
It 9
seems to-me that you.couldn't look at that and past a fair 10, judgment'without reflecting on the history of how those 11 local ' control stations got started.
.They got started like
~
~
l2 SPDSs got started:
with saying, okay, if the control room
(
)'
13 has a fire and you lose the habitability of the control 14 room, you have to have some simple place to go to take care 15.
of --
16.
MR. WARD:
I don' t think he means just that, do 17 you?
18 MR. MICHELSON:
This is motor control centers.
19-MR. JONES:
All kinds of local controls.
- Valves, 20 local control -- whatever it takes.
21 MR. REED:
It is very easy 10 years later, like 22 you can be two years late on SPDS, to say that there are 23 deficiencies.
But the rules of the road were a certain way, 24 in history.
25 MR. JONES:
What we are talking about here is ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646 E-.1 1
r 26461.0 38 REE J
1 gross deficiencies are obvious sometimes, where in the 2
control room, you turn the valve to right, you go down to 3
the local control station, the valve goes to the left.
4 This sort of thing.
5 MR. REED:
But you are not talking ultimate 6
backup controls.
~
7 MR. JONES:
No.
8 We looked at a whole series of man-Aachine 9
interface concepts for advanced techniques.
I and L 10 conducted a simula' tor eva'luation of CRT displays, various 11 types of CRT displays.
Basically speaking, picture 12 displays are a much better way of displaying in forma t ion.
()
13 We did some experimental assessments of expert systems 14 using the test facility at Idaho National Laboratories.
15 And it turns out, interestingly enough, *: t the fairly 16 simple expert system that we tried out wasn't as good as a 17 good smart operator.
All of which says if you are going to 18 use expert systems, you have to make sure they are 19 reasonably expert because you gain a great deal by using a 20 computer system if the operator himself is well trained.
21 He looked at decision-making under stress to see 22 if this made some differences.
This is in a severe 23 accident area.
Are there certain conditions where stress 24 itself will cause a degradation of performance?
25 Interestingly enough, one of the accidents that happened iiV ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coserage 800-336-6M6 i
26461.0 39 REE I )
\\'/
1 recently, one senior operator literally passed out under 2
the stress of the situation.
3 MR. REED:
That is why you get back to aptitude 4
testing in the first place.
5 MR. JONES:
I think you are right.
6 We looked at guidelines for the evaluation of 7
visual displays units and published a very thorough report 8
on this which, incidentally, I have recommended to the e
9 human factors society as a human factors standard for use 10 in the United States.
It may well be.
11 The Italians used this particular publication on 12 their plant that is being built in Italy and used this
,-()
13 guideline for their particular guides in how to layout a 14 fairly complex advanced CRT display.
15' Then we looked at human performance in 16 nondestructive testing.
It turns out that in that area, 17 the performance of the individual operator, when he does 18 his nondestructive testing, may be more important than --
19 MR. MICHELSON:
Before you leave that slide, 20 perhaps it is in here or perhaps I don't understand, 21 obviously one of the important and simple elements of 22 man-machine interface is the proper labeling, color coding.
23 It is not advanced technology.
24 MR. JONES:
No, sir.
25 MR. MICHELSON:
Are you looking into those areas
'w)
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 804336-6M6
26461.0 40 REE.
p Y\\ /
'l and - finding out why some plants seem to be so poor about : it '
-2 whereas others have done a very fine job?
3 MR. JONES:
This is the function of the control 4
room design review.
As far as the human factors --
5 MR. MICHELSON:
I have in mind the pumps and 6
valves and pipes and so forth.
7 MR. JONES:
Yes, sir, we sure are.
8 MR. MICHELSON:
Are you getting int'o that?
9 MR.-JONES:
Into'what should be done aboutit?
~-
10 MR. MICHELSON:
Is that a part of what you --
11 MR. JONES:
It is a part of what we are going to 12 do next year.
(
13 MR. MICHELSON:'
Oh, I am sorry.
14 MR. JONES:
Well, nothing'like leading into a 15 good -- and then of course we-also did the standard review 16' plan publication on. human factors, both the whole human 17 factors-program and SPDS.
We worked on safety indicators, 18
.the NUREG on the safety. indicators program, which was an 19-attempt to develop some sorts of measures of safety
-:20 performance which' could be used in the evaluation of plant 21 performance,'and carried that on further into an analysis 22 of maintenance factors.
The whole business of safety 1
23 indicators is something that the NRC is looking at now, and 24 one way of evaluating safety and to some extent management 25 performance.
i ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
m 126461.0:
41 REE
- (]
'\\ >
1 We also published a. workbook and guidelines for 2
the Staff to use in assessing operating. licensees.
3 MR. MICHELSON:
Could you give me an' example of 4
a' safety indicator?
. MR. REED:
Give him the best one.
Give him the 6
very best one.
7' MR. JONES:
Human err'or.
8 I am,sorry.
I can't answ'er the best one.
Maybe
-9
.you can help me.
I don' t remember.
10 MR. REED:
Well, I think forced outage rate is 11'
-probably one of the'two best.
12 MR. MICHELSON:
That is what I needed.
I wanted
()
13 to understand what kind of indicators this might be.
Thank 14 you.
15 MR. JONES:
As a-human factors engineer, I know 16 about human error rates.
17 Let's talk about revision 2 of the human factors 18 program plan which is currently on its way to the printer.
19 Several things have-happened.
In the first 20
-place we had the budget cuts.
Research has cut out the --
21 its human factors program.
And the NRR budget has been 22 reduced quite a bit in this area.
We have a commissioners' 23 policy statement, Commission policy statement which shifts 24 the emphasis from development to auditing and encouraging 25 industries' self-improvement.
ACE-FEDERAL REPORTERS, INC.
20;.347-3700 Nationwide Coverage 800 336-6646
i 26461.0 42
.REE 1
MR. MICHELSON:
Was there some express basis for 2
believing that it was okay to go ahead and reduce these 1
3 programs or eliminate them because, perhaps, they aren't 4
that.important to safety anyway?
5 MR. JONES:
-No, sir.
On the contrary -- I have 6
to defend ourselves rather quickly.
If you look at the l
7.
actual allocation of dollars, first research is cut out(~lt 8
is human factors program.
But the human reliability part
~f 1
9 of research, PRA and that sort of thing,"is cont.inuing, and l
10 severe accident activities is doing human factors 'under a 11 different guise.
,1:2 As far-as the division of human-factors safety
{
()
13
.is concerned, the old NRR division, a large number or a 1
[
14 large part'of the individuals that worked in that group, in j
~
15 that division are not in the so-called licensing division 16 doing control room design reviews, audits, procedure 17 reviews, SPDS reviews,-all those things that we used to do 18 in the division.
And actually, as I will show you a little 19
-later, the number of individuals still working on the Staff 20 human factors program is essentially the same as it was
{'
21~
before.
So it really hasn't been reduced 23 in NRR but it has been, the research --
24 MR. JONES:
The research activities have been 25 cut back.
{
i a
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
)
26461.0 43 REE I
1 MR. MICHELSON:
Was the other part moved to some 2
other part?
3 MR. JONES:
The human factors pRA was moved to 4
receipt liability program branch.
5 MR. MICHELSON:
The part that was retained then 6
in total is just a percentage of the original human --
7 MR..
JONES:
Roughly one fifth.
m.
8 MR. MICHELSON:
So about 80. percent 'f'it cut o
. t 9
out.
10 MR. REED:
You mentioned that some of your 11 research now might be -- (Inaudible.)
12 I would like to ask fou, because whether the
,,(,)
13 ACRS Staff members -- with respect to the possible impact 14 of aptitude testing people is successful.
15 Are you going to perhaps pull together data with 16 respect to the plants that use aptitude testing or don' t 17 use it and what success or lack of success they have had?
18 MR. JONES:
He haven't but it is an excellent 19 idea.
I would like to make note of that.
20 MR. REED:
He sort of ran out of information and 21 he didn't cover the waterfront quite.
But we didn't get 22 any further conclusion of it.
23 MR. MOELLER:
Rig ht, you are correct.
24 MR. WARD:
Hould you fellows try to talk into L
25 your microphones, please.
,8 N
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6 L.
26461.0 44 REE.
- ry 1
MR.' JONES:
A-third item down here that has 2
occurred as f ar as the human factors program plan, with the 3.
elimination of the research program in human factors, this 4
essentially has become an-NRR plan.
It.is -- the 5
responsibility has been shifted to the division of human f
6-factors technology, whereas the earlier plans have covered 7
all the executive offices with all the offices of the'NRR.
8 And the revision 2 is kind of aimed at a broader 9
audience.
Ths detailed program plan, the schedules and-10 things that we had in earlier plans have been eliminated.
11 If you look at the resources real.briefly, you 12.
can see that the dollar numbers have gone down.
This was
-f )
i 13 S5,213,000 last year.
This number here is $1,132,000, 14 including some work on the national academy of science.
15 panel which both NRR and research are funding.
The human 16 reliability that was here.is still at a 900,000 level in 17 research, but it is not part of the human factors program 18 plan.
19 If you look at the total, the dollars that we
-20 have within NRR, a small amount in training, a good bit in 21 operator licensing because of the work we are doing in this, 22 a little bit in procedures, finishing up, this is 23
. essentially the project in Norway which was supported by 24 research and is now being supported by NRR.
That is a 25 major effort in the whole area of control room and ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationside Coverage 800-334 6646
l F
26461.0 45
~ 'REE
.r T.
's' 1
artificial intelligence, computerized procedures.
2 MR. MOELLER:
You quoted a 900,000.
Where was
[
3' that?
4 MR.. JONES:
In human reliability.
5 M R.' MOELLER:
But it shows zero.
6.
MR. JONES: 'That is true. 'That is not fundedL out of 'the human,f actors program. plan.
But the research 7
,8 budget for this line item, now in.the reliability program 9.
plan, PRA, is about 900.
g 10 MR..MOELLER:
Thank you.
l 11 MR. REMICK:
Is there anything that has been 12
.done on continuing exams by the NRC Staff?
.ym i
e i
'13 MR. JONES:
As far as I know, yes.
We plan to y A/
14-eliminate requalification exams.
There was originally an 15-
' idea that we would not conduct requal examinations.
We L
'16 were told to do requal examinations anyway.
17 MR. REllICK:
Is that because --
l" 18 MR. JONES:
I cannot tell you why.
All I know i
l 19 is, it was done.
20 MR. REED:
I think it is because of a NUMARC l
21 double cross, quite frankly.
22 What did I say, Charlie?
l 23 MR. WYLIE:
I heard you.
24 MR. REED:
The Staff came around to spot-check 25
. operators in a sensible way.
The ACRS recommended it two LO I~
ACE-FEDERAL REPORTERS, INC.
202 347 3700
. Nationwide Coverage 800-336-6M6 -
L-
F.I-I
,26461.0 46 L
REE 1
or three times by letter, I think.
And then NUMARC comes 2
in and talks against it.
I can't figure that out.
3 MR. JONES:
As far as the.new program is-4 concerned, section 1 of this is training and we have 5
already talked about the policy statement and the 6
self-regulation concepts and-INPO accreditation program's 7
that the NRC is supporting.
As far as plan activities.are
~
8
' concerned ~then, what we will do is evaluate the 9
implementation of this policy statement, do 10
' post-accreditation reviews after INPO has conducted an 11
. accreditation visit, do some performance based inspections,-
12 observe team visits in the INPO accreditation program,.
.()
.13 track pretty much what goes on, look'at the SALP process, 14-at any kind of archival data that we can have, and interact 15.
with INPO and with industry-in general, and finally report 16 to the Commission what recommendations we have for further 17 action a year ~from now.
18 The first. interim report is to the. Commission, 19 which is actually in the process of being written right now.
20 It may even be out from our office by this time.
21 MR. REMICK:
Do you plan additional 22 post-accreditation evaluation or visits, whatever you are 23
. calling them, after that interim report?
24 MR. JONES:
To the best of my knowledge, yes.
25 If we look at operator licensing exams, here is f%
~%)
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
1 26461.0 47 REE
, c~\\
(
?s/
1 where we made a mistake.
It is the last two items in your 2
handout rather than where it should be.
This is a 3
_ continuation of the ongoing process that I have talked 4
about earlier.
We are trying to make sure that we really.
5' do have. valid reliable examinations.
So we are working on
~
6 the RO and SRO catalog of abilities, and the INPO 7
-job-related.
We are looking at the validation of t3,e 8
. content of this by subjdct. matter experts.
We want to 9
consider do we need plant-specific catalogs to supplement.
10 the generic catalogs.
11 Incidentally, when we talked about 30,000 12 questions, test item questions, recognize a lot of these
(
- 13
_are plant specific.
In other words, they are particular
[
14' plants, not just generic questions cut across all plants.
i 15 There are a large number of generic questions.
j 16 So as we get more plant specific data, we will t
17 probably go up to 100,000.
We want to test the handbook 18 for examiners and particularly the construction of I
19 simulator scenarios.
We are getting started in this area.
L 20 This is new.
We want to make sure that it is a good 21 approach.
And the examiners are being trained on the 22 function-based procedures to make sure that when they go 23 out and do their examinations, that they really look at the j
24 function-based procedures rather than the equipment based 25 procedures that -- (Inaudible.)
I.
l I
ACE FEDERAL REPORTERS, INC.
202.347-3700 Nationwide Coverage 800-336-6646
-, - ~
~
.,. -.~.
l f26461.0 48 REE I-1 MR. REMICK:
Your data-bank questions, the
-2 30,000, now you are trying to move toward making sure that 3
the. questions are performance based, I think,.using the 4
learning objectives, how are you going to; cross-check those'
-5 30,000,~ whether they are performance based or not?
Is r
6 there any plan to do'that?
7 MR. JONES:
Obviously some of them will not be.
l 8
To th'e. extent practical, that is part of the examiner's i
l 9
handbook, that is the job of the examiner who writes the 1:
l.
10 question.
If he finds a question that ~is not perf ormance.
he should modify it, we tell him, to make sure that l
11
- based, l-L 112 it becomes performance based, and then that item would, we
. /( )
13 hope, be eliminated.from the bank.
E 14 MR. REMICK:
Are you now providing the examiners 15 who are_ preparing the written exams with the learning 16
- objectives from the training programs?
17 MR. JONES:
Yes.
18 Then, of course, continuing on the inevitable 19 work that we are doing in getting the rule on-the street.
20 That is part of L this year's program and we hope _ it will be 21 finished very shortly.
l 22 The next area we want to look at is the 23 procedures.
And we are continuing to look at emergency.
24 operating-procedures.
We are looking at guidelines for --
'25 guidance for normal procedures and abnormal operating
'O i
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6646
~26461.0 49 REE y.;.
-. s' l-procedures for some human factors deficiencies that exist 2
and we want to do a regulatory analysis to' determine if 3
additional guidance is required, i'f we have to go out,with 4
some sort of regulatory action.
5 In some of.the work we have done, we have 6-discovered there are significant problems in operators 7
transitioning between procedures.
Ther have difficult 8
deciding where they should go and in ma' king that-transition.
~
9 And so there is some draft guidance we are looking at on 10 the interprocedure interface that-is being developed.
We 11 will have to decide using a value impact analysis and so 12 forth as to whether this can be appropriate, whether it
[)
13 would be appropriate for us to take action in this area.
14 He are doing a valve impact analysis of the EOP 15 upgrade.
Research is completing a. study of -- a value 16 impact study _of the change from event base to function base.
17 Kind of a historical background, tell us what the problems 18 were and how it went.
And then we will do a value 19 assessment and try to determine where we need to go in the 20 future.
21 I should note incidentally that all the 22 maintenance procedure business is now over in the 23 maintenance program.
24 He are looking at accident stress on operators.
25 The research study that I mentioned to you earlier and the ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800-336-6646
26461.0 50
'REE 1
human factors issues branch will evaluate what follow-up 2
actions are appropriate.
We are considering asking for 3
-experimental work.
4 EPRI, of course, has been doing quite a bit of 5
work in the operator aids and expert systems, and we also.
6 asked the Halden project to look at computerized systems.
7 So Halden is developing a way of compu'terizing p,rocedures, 8
and not having -- putting the procedure,up on t'he board and-
~
9 reading it, but rather giving guidance to the operatpr-as 10 to where he should go based upon some sort of'a sens'ible 11 scheme that is expert, if you will.
12 This is also sponsored strongly by all the other
(_m) 13 nine nation signatory countries.
Everybody is interested
=
~
14 in it.
We recommended it and everybody latched hold of it.
15 MR. REED:
On this issue of stress, I think you 16 have to be very careful, I am not even sure that you could
-17 transfer that work to Englishmen from Americans.
You have-18 got to be very careful that you know the stress response of 19 the individual or his complacency or what before you start 20 out.
And you have to evaluate that first.
21 I had a control room operator who developed a i
22 mental breakdown, and after that was over and we went out 23 for analysis -- that is many years ago, 1960 -- when it was 24 all over, it was determined that he should have never been 25 a control room operator.
He just was not a personality and l
f ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800 336 6M6 u--
1 r
-w i
I l
.26461.0 51 j
i
'REE j
h tri f'\\- aptitude that was adjusted to that area.
So you have to be l'
2 very careful when you get involved with stress --
t 3'
MR. JONES:
I think I need to say that.we are 4
not talking about individual performance.
It is rather, 5
are there certain situations or conditions which are, by 6
their nature or design, unusually stressful and what can we 7
do to reduce those.
8 MR. REED:
They are stressful to some and not 9
stressful'to others.
10-MR. JONES:
I couldn't agree with you more.
-11 MR.-KERR:
Will natural mechanical ability l-12 testing bring that out?
1 h -)s
-13 MR. REED:
There are a battery of tests,_about
)
14 four, that will bring that out, yes.
I 15 MR. JONES:
There are some pretty good tests for
-16 stress -- the-ability of individuals to handle stress also.
t 17 MR. MOELLER:
When you mentioned the operator 18
-who passed out 10 or 15 minutes ago, was this during an
'19
. exam?
20 MR. JONES:
No, sir.
This was during an.
21 incident, an event.
And he performed very well, by the way.
22 I am not trying to say that that was bad performance on his 23 part.
On the contrary.
Apparently he did a good job.
It 24 is just that this was a very stressful event and he handled 25 it well, but after it was over with, you know how it O
u/
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmerage 800-336-6646
26461.0 52 REE ep
'/
1 happens.
2 MR. WARD:
How many more minutes do you need?
3 MR. JONES:
Not very long.
4 MR. WARD:
Why don't we let him zip along.
5 MR. JONES; Man-machine interface.
What we are 6
trying to do here is evaluate how well this whole business 7
of control room design reviews and so-forth is being done 8
by the regulatory, by the licensing divisions.
We 9-mentioned local control stations and we are looking at that
~10 to decide what kind of. action we think may or may not be 11 necessary.
12 He obviously have problems with annunciators and
(, )
13 what kind of pattern recognition and so forth.
We have 14 asked -- research on advanced annunciators is going on at 15 Halden and at EPRI and we are looking at what we need to go 16 about there.
And then there is a whole body of development 17 in the whole area of operator aids, advanced displays, CRTs, 18 that we think will be coming down the line even in current 19 plants, how are we going to regulate in this area.
20 Staffing and qualifications is pretty much a 21 continuation of what we talked about.
We want to evaluate 22 how well the industry is looking at its policy statement 23 and engineering expertise on shifts that was put out last 24 fall.
And provide information to the Commission on how 25 well it is.
The Commission was also requested and received
,d,m ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-336-MA6
26461.0 53 REE 1
a draft advanced notice of rulemaking for degrees in SROs.
2 What will happen to that, I don't know yet.
That is still 3
very much up in the air.
4 We are developing a policy statement on working 5
hours which I have already mentioned.
And we are updating 6
a series of regulatory guides and so forth.
The whole 7
staffing qualification area now is one of tracking what 8
industry does pretty much.
9 Management and organization, all work on 10 management and organization has ceased as far as the human 11 factors developmental plan is concerned.
Not that we are 12 not interested in management, it is just that --
$llI 13 MR. WARD:
We recognize that is not a problem.
14 MR. JONES:
That is right.
15
( La ug h te r. )
16 MR. JONES:
And the Commission's policy guidance 17 calls for moves towards performance based programs, the use 18 of indicators by I&E, to improve, solve and what have you.
19 Finally, the last new section is one we have 20 chosen to call human performance.
This is an attempt to 21 try to get a handle on this business of what can we do 22 about human error, human performance and related activities.
23 We are working with an EOD to define a protocol, to try to 24 get a better handle on what kind of LER in f o rma t ion 25 reporting is necessary.
We want to look at INPO's human
_j ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 80433MM6
26461.0 54 REE 1
performance evaluation system and the CN program and 2
finally, try to collect enough data that we can tell, we 3
can track the whole business of human performance and see 4
if there are areas where we need improvement in our 5
regulatory process.
6 MR. EBERSOLE:
Isthereanapproachtowardjaset 7
of guidelines or rules which' would include consideration of 8'
SPDS that would lead you to make a decision when you should 9
automate the function or when you should do it manpally?
10' MR. JONES:
To say that we are going that way in 11 general is absolutely true.
To say that we have a roadmap 12 to get there is not true.
We are working on the roadmap.
13 For example, I mentioned this visual display evaluation j _
14 criteria that we developed here last year.
This is an area 15
-- we are working in the area of artificial intelligence.
16 One of the items of the program plan is to try 17 to track the whole artificial intelligence area and come up 18 with where we need to go from a regulatory position as 19 industry moves to it, because industry is moving to it.
It 20 is a sensible way to go.
Expert systems at least, 21 computerized procedures and related activity.
22 That is all I have.
23 MR. WARD:
Thank you very much, Dan.
24 (Whereupon, at 3:10 p.m.,
the meeting was 25 adjourned.)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationside Coverage 800-336-6646
CERTIFICATE OF OFFICIAL REPORTER This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of:
NAME OF PROCEEDING:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 312TH GENERAL MEETING DOCKET NO.:
PLACE:
WASHINGTON, D.
C.
DATE:
FRIDAY, APRIL 11, 1986 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
(sigt)
W4 (TYPED)
REBECCA E.
EYSTER Official Reporter ACE-FEDERAL REPORTERS, INC.
Reporter's Affiliation O
APR 10 '86 14:04 SRL P02 J
April 8,1986 TO:
ACRS MEMBERS -
FROM:
D. A. WARD, CHAIRMAN SUBCOMMITTEE ON HUMAN FACTORS REPORT OF MEETING OF MARCII 19-20,1986 (he SW Attending the meeting were ACRS Members Moeller, Mich=' son, Wylic and
{m) myself and ACRS Consultant Gimmy.
This two-day meeting had three purposes. The subject the tirst full day was spent was an information session on one called the state-of-the-art in applying automatic monitoring and control functions in the operation of nuclear power plants. We heard from representatives from Staff and several representatives from the industry. The second topic was a review of the NRC's Human Factors program plan by the Staff.
I'll summarize what we learned on the first and third topics and then a t
representative of the Staff will review the present Human Factors Progress Plan for you. At the present time, I do not see a necessity for an ACRS report on any of these topics at this month's meeting. The third topic concerned the implementation of emergency operating procedures in actual plants.
On the first day of the meeting We heard 11 presentations on the state-of-the-art in automating monitoring and control functions. In an introductory presentation from the Staffit was stated that the early attitude of both the industry and the NRC toward computers was largely influenced by experience with big main frame computers and with the unreliability of this type of computer in a process setting.
_i However, in recent years, both the NRC and the industry have become aware that
/
'd computers specifically designed for monitoring and contro!!ing processes can be considerably more reliable than the earlier experiences indicated, and in fact, may
APR 10 '86 14:05 SRL P03 ACRS MEMBERS 2
April 8,1986 I
)
v' l
be extremely useful tools for aiding operators in managing both routine and l
emergency operations in nuclear power plants. It is interesting that the SPDS (Safety Pammeter and Di:; play System) which has been mandated as an addition to each plant as a post-TMI backfit seems to have served as sort of an introduction for the nuclear power industry to the use of modern process computers.
There was not really a specific requirement by the NRC that SPDS should be computer based but the characteristics desired for the SPDS and the availability of process computers that could have the appropriate characteristics has led really ~
every licensee to use some form of digital computer as the heart of his SPDS system.
However, despite the introduction to the computer age offered by the SPDS, the state of applications of computers in nuclear power plant control is not very advanced. We heard a report from Staff people who had meetings with NASA at the Kennedy Space Flight Center. Originally NASA made very little use of process p
computers in their ground control systems (as opposed to on-board systems). It has V
gradually, however, through the years, made more and more use until now the ground control systems at Kennedy Space Center are largely, what you might call, automated and make heavy use of process computers with, they believe, considerable success. They also reported that in NASA's design of a future space station, they have set aside 10% of the total development budget for the development of software and related automatic control systems.
We also heard a report from Leo Beltracchi of the Staff who sometime ago visited the Savannah River Plant and was briefed there by our Consultant, Mr. Gimmy, on i
the application of process computers in the monitoring and control of the Savannah River Reactors. This approach was initiated about 18 years ago and has gone through several generations ofimprovements. First there was only monitoring, then some close-loop control of the process was added. The next step was adding scram system safety circuits based on computers, a backup scram system, and most recently, addition of an alarm diagnosis system which is a form of an " expert system". Application of computers in today's nuclear power plant control is about at the stage where the Savannah River Plant reactors were in about 1970.
We also heard a report on methodology that was developed for so-called function allocation. This is a systematic approach for deciding which monitoring and control I,)
functions will be assigned to the human operator and which will be assigned to computers or other automatic control systems.
APR 10 '86 14:06 SRL PO4 ACRS MEMBERS 3
April 8,1986
)
We also heard a report from Professor Barclay Jones from the University of Illinois on research he is doing in what he calls the process of"embeded training".
By embeded training, he means combining the training functions for a nuclear power plant operator with his normal operational functions by providing them with a control system for the reactor that is part simulator. To most of us, this seemed to be like a radical, perhaps even dangerous concept, but Professor Jones is looking at it conceptually and he sees this as a possibility for fully automated control rooms of a future generation.
We also heard a report from Westinghouse on the approach they are taking on advanced control room design which is going to extensive use of plant computers and automation of the many process functions. heir approach is to move in very deliberate steps and their intent is not use for the fulli potential of the technology for near-term designs. However, the French, with similar plants, seem to be taking rather radical steps toward quite complete automation of the control rooms.
]
We also heard a report from Rom Duffey from EPRI who described the work that is going on at EPRI. This program is all directed toward application to the present population of plants, and not for future plant designs. He says that the utilities are very supportive of the research they are doing; in fact, each element of the research program has what he calls a host utility which is particularly involved in nuclear development work.
Finally, we heard a couple of good tutorials really from Mr. Bill Bertch and Mr.
George Niederauer of Energy, incorporated. They described in practical fashion the application that artificial intelligence and particularly expert systems will have in nuclear power plant operation. Then do not expect artificial intelligence to ever give close-loop automatic control of a process; for example, of the nuclear power plant process. Instead, it would automate certain observation, interpretation and decision analysis in nuclear power plant operation, leaving the final decision and actions to the human operator.
Next, we heard reports describing Emergency Operating Procedures developed for use at two specific plants.
The first was on behalf of the Hatch Plant of the Georgia Power Company. They
(^'s described and demonstrated a " flow chart" format. This is a very comprehensive V
and detailed approach to emergency operating procedures which uses large logic diagrams, or flow charts, to provide the initial response to any major plant upsets.
The purpose of the procedures under the flow chart was described as being to bound
APR 10 '86 14:08 SRL P05 ACRS MEMBERS 4
April 8,1986
,b the problem and stop the spread of degradation of the plant equipment or systems or for the process. Each flow chart ultimately leads through a logic path to a preferred recovery procedure in what is called "End Path Manuals". The second type of procedure we heard about was the so-called single column instruction format. These procedures were described to us and demonstrated as. applied at the Browns Ferry Plant of the Tennessee Valley, Authority.
'. As you have heard before, the NRC has required that each plant develop for itself
' some sort of emergency operating procedures which are " symptom oriented" rather than " event oriented". The NRCjust provided general guidelines and specifies a process for development of the procedures. The result is that there is a very wide variety and spectrum of styles and types of procedures in use at the many operating plants in the country. The two that we heard described may not be at the extreme ends of the comprehensive versus simplicity spectrum, but they are probably representative of something close to the ends of the spectrum.
(]
So far as which type of procedure is superior, I don't know, and I don't think V
anyone in the NRC Staff has taken a particular position on this. Certainly, each of the utilities believes that their system of procedures is adequate. Whether they think their own system is, in either case, the best approach that can be taken, I don't know.
{
The Staff approach to emergency operating procedures has been to insist that each licensee develop a process for developing its own EDPs and that the NRC staff then would audit and monitor that process with the expectation that if the process was going well, good emergency operating procedures would somehow appear in the l
plants. The NRC seems to have reached the decision that this approach has failed.
l They have monitored the process and often times found indications that there was a satisfactory process, but when they audited actual procedures, they found major
. problems with the all-important end product. Apparently, the Staff recognizes the
,need for more extensive auditing program of the actual final product, the emergency operating procedures.
I l
C)
o l
Meeting No.
Agenda Item Handout No.
312th 12.1 Title HINAN FAODRS I Authors J. Schiffgens E. Igne List of Docunents Attached 1.
Index
-g*********************
2.
Proposed Schedule for Discussion of Automation
)
in Nuclear Power Plants, the Human Factors Program Plan, and Status of Emergency Operating Procedures Implementation
- c.,,,,,,,,
3.
Status Report 4.
Summary / Minutes of the ACRS Subcommittee' Meeting on Human Factors, March 19,20, 1986, Washington, D.C.
Instructions to Preparer From Staff Person J. Schiffgens
- 1. Punch holes
[
]
- 2. Paginate' attachments l
- 3. Place copy in file box N
J E
l k'r8l t
. y, > <
\\v) 1 l
SECTION 12 l
INDEX
/N c
' ( j)
(
1.
Proposed Schedule for Discussion of Automation in Nuclear Power Plants (NPPs), the Human Factors Program Plan (HFPP), and the Status of Emergency I
Operating Procedures (EOFs) Implementation.
2 2
Status 3
1 3.
Summary / Minutes of the ACRS Subcommittee Meeting on Human Factors, March 19-20, 1986, Washington, D. C.
4
(
4
l i
l April 7, 1986 i
PROPOSED SCHEDULE FOR DISCUSSION OF AUTOMATION IN NPPs, THE HFPP, AND THE I
STATUS OF EOP IMPLEMENTATION ON FRIDAY, APRIL 11, 1986 1:15pm Subcommittee Report D. Ward 30 min 1:45 Review 1985 Progress on HFPP and Discuss Proposed 86'-87' HFPP D. Jones 30 l
l 2:15 Committee Discussion 15 m
/
o UNITED STATES 8'
NUCLEAR REGULATORY COMMISSION o
nE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
/" N.o WASHINGTON, D. C. 20555
[
! N%*..,*/
v April 7, 1986 MEMORANDUM FOR:
ACRS Members FROM:
J. O. Schiffgens, Staff Engineer j
SUBJECT:
STATUS REPORT ON AUTOMATION IN NJ
, THE HFPP, AND THE STATUS OF EOPs IMPLEMEtWA ION FOR THE 312th ACRS MEETING, APRIL 10-1Q,'1986 The Human Factors Subcommittee met on March 19-20, 1986 to a) explore the potential for automating more of the monitoring and control functions in nuclear pewer plants (NPPs) so as to reduce the burden on plant operators and enhance safety, b) review 1985 progress on the Human Factors Program Plan (HFPP) and plans for 86' and 87', and c) be briefed on the status of Emergency
,x Operating Procedures (EOPs) implementation.
(,)T
/
The first day of the meeting began with presentations by the NRC Staff on current Staff views and activities in the area of Automation of Monitoring and Control Functions in Nuclear Power Plants.
This was followed with presentations by university and laboratory researchers, vendor representatives, and consultants on the potential for and use of computers for NFP operation and control.
The second day of the meeting began with a Staff briefing on 1985 progress on the HFPP, the status of the ongoing HFPP, and the status of EOPs implementation.
This was followed by presentations on symptom based EOPs that use a) the " Flow Chart Path" format, and b) the " Single Column Book" format.
dill
E
,8
'o,,
UNITED STATES
/
NUCLEAR REGULATORY COMMISSION g
W
.m ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
[
WASHINGTON, D. C. 20555 o
(v} ' % * * * * * /
m April 1, 1986 l
MEMORANDUM FOR: D d A. Ward, Chairman, ACRS Subcomittee on Human j
tors FROM:
E. G. Igne, Senior Staff Engineer l
SUBJECT:
SUMMARY
/ MINUTES OF THE MARCH 19-20, 1986 MEETING OF THE ACRS SUBCOMMITTEE ON HUMAN FACTORS, WASHINGTON, DC r
A working copy of the combined Summary and Minutes for the subject meeting is attached for your review. Copies are being distributed to other ACRS members for their information or review.
A corrected version of the Minutes, after incorporation of any comments by you or others, will be transmitted to you for certification as to accuracy. After certification, the Minutes will become the official record of the subject meeting.
J
Attachment:
Summary and Minutes of the above subject meeting cc: ACRS Members ACRS Staff:
R. Fraley M. Libarkin T. McCreless J. McKinley G. Quittschreiber K. Gimy, ACRS Consultant I
E. IGNE:jk 4
4/1/86
WORKING COPY
,em.
DATE ISSUED: APR. 1, 1986
(
)
SUMMARY
/ MINUTES OF THE ACRS SUBCOMMITTEE ON HUMAN FACTORS v
MARCH 19-20, 1986 WASHINGTON, D.C.
The ACRS Subcomittee on Human Factors met in Washington, D.C. on March 19-20, 1986 to a) explore the potential for automating more of the monitoring and control functions in nuclear power plants so as to relieve the burden on plant operators and enhance safety, b) review the 1985 progress on the Human Factors Program Plan (HFPP), and c) be briefed on the status of Emergency Operating Procedures (EOP) implementation.
Notice of the meeting was published in the Federal Register on March 10, 1986 (Attachment A). The schedule of items covered in the meeting is in Attachment B.
A list of handouts kept with the office copy of the minutes is included in Attachment C.
There were no written or oral statements received or presented from members of the public at the
[\\
meeting.
E. Igne was the cognizant ACRS staff member for the meeting.
i
)
Principal Attendees ACRS NRC Staff D. Ward, Chairman D. Jones C. Wylie, Member F. Rowsome D. Moeller, Member J. Bongarro F. Remick, Member L. Beltracchi C. Michelson, Member B. Bolger K. Gimmy, ACRS Consultant W. Kennedy' E. Igne, ACRS Staff S. Weiss Others J. Carrera, Westinghouse R. Contratto, Westinghouse J. Easter, Westinghouse R. Duffey, EPRI P. Springer, Georgia Power Co.
1 L. Guewa, Georgia Power Co.
1 W. Williamson, TVA R. Knobel, Knobel & Assoc.
M. Paradies, Univ. of Illinois at Urbana-Champaign 4
- 8. Jones, Univ. of Illinois at Urbana-Champaign W. Bertch, Energy Inc.
G. Niederauer, Energy Inc.
HUMAN FACTORS 2
March 19-20, 1986 Meeting
~
'J In opening the meeting, D. Ward stated that the meeting is essentially a gathering of information on automation and its monitoring and control of.
plant parameters in order to enhance nuclear power safety. The subcommittee in particular should consider the following questions:
Are there NRC regulations or perhaps attitudes within the Staff that inhibit movement in the adoption of the technology that is available?
Should there be NRC regulations or attitude in the Staff that would actually promote moving in the direction of adoption of the technology that is available?
Should or can the ACRS do anything to promote or inhibit movement in this direction?
,/ %
i
\\
\\j D. Moeller stated that human factors aspects of the control room should include control room environmental conditions, i.e., temperature and humidity.
NRC Presentation S. Weiss, NRC, discussed NRC's past and present attitude toward automation or computers in control room in nuclear power plants and commented on a topical meeting on computer operators for nuclear power plant operation and control that was held in Pasco, Washington.
With respect to D. Moeller's question, S. Weiss state
- that the guidelines in NUREG-0700 give environmental conditions for equipment survitability in the control room. He would like to change the technical specification to include temperature considerations for the operator.
He further stated that he is aware of the prol'lem and some work to alleviate this oversight is ongoing.
HUMAN FACTORS 3
March 19-20, 1986 Meeting 77 I
\\j The NRC has no formal criteria for the use of digital computers in operational nuclear plants.
However, informal criteria have been established from experience gained in the safety evaluation of computer-based protection systems, e.g., Core Protection Calculator System and RESAR-414 Integrated Protection System. Another source of NRC experience has been the review of the Safety Parameter Display Systems (SPDS).
Subsequent to THI-2 accident the NRC established requirements for a SPDS in control rooms of nuclear plants. The nuclear industry has responded to this requirement by designing computer driven CRT display systems.
The Staff has surveyed six plants to determine the state of SPDS implementation, and to ascertain the scope and depth of review necessary for post-implementation audits.
In general, S. Weiss stated that the SPDS's surveyed have major deficiencies, and some are identified as follows:
Lack of SPDS availability because of gross system malfunctions, A) l.v Display of unreliable or invalid data and alarms, Poor acceptance of SPDS by operators because of reliability
- problems, Failure of management to integrate SPDS into the operational environment, Inadequate documentation of SPDS and failure to control system testing and modifications, and Slow SPDS response to some operator commands.
S. Weiss did state that two plants audited did well. These plants had extensive use of computers or worked closely with vendors or people with 4
1
HUPAN FACTORS 4
March 19-20, 1986 Meeting ry.
i computer experience. As a result of this study NRC is planning to perform post-implementation audits of the SPDS on all plants.
Problems similar to those described above have also been identified by the NRC during the evaluation of the emergency data acquisition systems as a part of the Emergency Response Facility appraisals.
In response to a question regarding the advisability of the use of SPDS in plants, S. Weiss stated that the concept of an SPDS is fine if its working, as shown in our audit of two good working SPDS.
It was further stated that as a regulatory agency, the NRC is trying to keep abreast of the technology that is developing so that the right questions during review can be asked. The NRC is not mandating a particular equipment to be used for the SPDS.
73 Next, S. Weiss presented a very brief overview of some papers that were
(
)
presented at the September 8-12, 1985 Pasco, Washington Topical Meeting on computer application for nuclear power plant operation and control.
L. Beltracchi, NRC, discussed a survey of Savannah River Plant (SRP) advanced control room design. The goal of the survey was to collect information on design of advance control rooms in order to establish design knowledge data base computer technology and its impact on safe plant operation. The SRP is a DOE production facility.
Each large reactor on-site is now equipped with multiple on-line digital computers that perform process data monitoring and display functions, and process control functions.
The computers also address functions to shut down the reactor when the plant approaches a violation of the safety function and are also used to provide expert advice to control room operators when multiple alarms occur, thereby reducing operator work loads during critical events. All computers have been backfitted into the operation of the plant. The initial design and operation of the control room were totally dependent on operator hard-wire technology.
HUMAN FACTORS 5
March 19-20, 1986 Meeting n,
(
)
v It was noted that major differences exist between the design and operation of the reactor plants at SRP and the design and operation of a commercial nuclear plant. Some of the differences are as listed below:
Plant design and operation are much simpler at SRP At SRP, all functions and tasks assigned operators are perfonned according to documented procedures Development of operator's procedures began during plant design As a rule, based expert system aids operators in responding to multiple alarms which identifies critical procedures.
In sumary, L. Beltracchi stated that SRP uses digital computers for process control, for safe shutdown of the reactor, and as an expert
[m
\\
q' )
system to provide realtime operator support during periods of information overload due to multiple alarms. These uses of digital computers have reduced operator workloads, improved productivity and process control, and improved operational safety.
W. Kennedy, NRC, presented a sumary of NASA presentation to the NRC Staff on NASA's (Kennedy Space Center, KSC) experience with automation.
The NRC Staff wanted to: (1) assess the role and function of the control room operations crew, the operator aids required to control and monitor operations, and the degree of automation used in operations; (2) identify and evaluate the design guidelines used to develop an advanced control room, such as the guidelines used to assign functions to operations, automatic systems, or digital computers; and (3) evaluate how the launch support facility operates using highly automated control rooms. This information will be used to establish a NRC data base of design guidelines and operational experience for use by the Staff in the review of (1) designs for advance control rooms for nuclear plants, and e
HUMAN FACTORS 6
March 19-20, 1986 Meeting D'
T (2) major revisions and modifications proposed for current control rooms. Establishment of this data base is important because designs for,
advanced control rooms based on computer technology have been submitted for NRC Staff review.
The KSC uses digital computers to automate many functions within the ground support systems and launch support systems. The complex operations required to prepare and support a vehicle's launch contain many hazardous steps, and some of these are manually controlled by human operators. The error-free human performance during critical steps of operation is being met through the use of computer-stored interrupt programs and control logic. The interrupt programs are called when an off-normal condition exists and detected from computer-monitored sensor signals. The interrupt programs perform diagnostics on the problem area and take predetermined action, based on logic stored in the computer.
Control logic is a specific form of sequence control, defined as system g
logic that monitors specified functions and takes direct action when an l
off-normal condition exists. Good performance of software is attributed to the design process, quality assurance and control, and the verification and validation methods used at KSC.
It was also mentioned f
that should the computer for a work station fail a hardwire backup control panel located at the work station is used.
Under this condition, the operator manually controls critical functions to maintain safe operation.
Based on the information provided by KSC personnel, it was stated that I
these tools would be useful to control room operators during emergency operations at nuclear plants. The technology is available and has been demonstrated.
W. Kennedy next discussed NRC's attitude toward future automation. He stated that control room automation applications are being considered.
{C
\\
He stated that automation is useful to both the industry and NRC's
1 HUMAN FACTORS 7
March 19-20, 1986 Meeting
/
V mission, and that artificial intelligence has significant potential in nuclear plant applications. He did caution that there are limits to what the NRC can require because of the backfit rule. Some NRC concerns toward future automation are as follows:
What applications of this new technology should or need to be regulated?
Who has responsibility for its use?
How to evaluate and regulate advanced technology?
M. Paradies, University of Illinois at Urbana-Champaign, discussed cognitive allocation and the control room to improve nuclear power plant performance.
He stated that a weak link in current connercial nuclear f3 plant design is an insufficient attention to the needs and capabilities
()
of the human operator. This flaw causes decreased plant reliability and reduced plant safety. To improve the design, these human errors must be considered early in the design process.
The allocation of functions needs to be considered not only in mechanical tasks to be performed but also the control requirements and the overall control philosophy.
" Cognitive allocation" is the allocation of the decision making process between operators and machines.
It defines the operator's role in the system. When designing a nuclear plant, a cognitive allocation starts the process of considering the operator's abilities. This is the first step to correcting the weak link in the current plant design.
He stated that industry figures show that 25% to 40% of all nuclear power plant outages are caused by human error.
If the average outage causes the plant to be down for one day, it will cost the utility about 4
$750,000 in replacement power cost alone.
HUMAN FACTORS 8
March 19-20, 1986 Meeting 7-I xg/
Using Rasmussen's decision making model, Mr. Paradies developed guidelines for normal and off-normal operations.
For normal operations the suggested guidelines are as follows:
Involve the operator in the control loop and help him refine his mental model of the plant.
Do not unintentionally increase the operators workload during events by adding tasks during normal operation that will carry over into abnormal situations.
Reduce the operator's monitoring load or change the form of the monitoring required.
Reduce paperwork and redundant tasks that take the operator out of
,q the control loop.
l
}
\\ j For off-normal events the suggested guidelines are as follows:
Maximize the use of the operator's decision making ability Maximize the operator's time sharing Simplify the operator's mental model Leave the operator in overall control, keeping the computer as a j
tool to be used to assist the operator.
j B. Jones, University of Illinois at Urbana-Champaign, discussed embedded training considerations. He stated that in the context of nuclear power plant operators, it includes operator instruction as an integral part of system control room and training occurs simultaneously with plant operations. The significance of the use of real nuclear plant control
HUPAN FACTORS 9
March 19-20, 1986 Meeting (y
(
))
rooms for training operators is to prevent operator boredom and maintain operator proficiency. Some problems with embedded training are (1) conflict between operational and training missions, installation in current operating plants would involve significant hardware and software modifications and (3) overcoming potential initial operator reliability decrement when changing from simulation to operation following an event.
J. Easter, Westinghouse, discussed Westinghouse views on advances in nuclear plant automation. He stated that the expected performance of nuclear plant aetomation set for the Westinghouse owners group is very high, i.e., target is two trips / year, but the control of the turbine-generator, which is already highly automated will contribute 1/2-1 trips per plant / year depending on the age of the system (newer units having the higher frequency). The Japanese trip rate is already in this range.
Some of the design goals are listed below:
/^N l
\\
)
Build the system so that no single failure will actuate more than one method of inserting reactivity.
The proper role of the I&C systems is to minimize transients to major plant components because the transients could jeopardize their integrity.
Do not violate the process with more sensor penetrations than r.acessary, which lead to the sharing of sensors between control and protection.
The machine role is still subservient to the human operator.
J. Eas.ter indicated that system reliability benefits favor the digital technology.
This technology offers the capability to build in large degrees of self testing, and the ability to take fault tolerant design approaches and make them reasonably cost effective.
HUMAN FACTORS 10 March 19-20, 1986 Meeting p.
U The new Westinghouse control room design will factor the applications of,
human factors engineering and artificial intelligence Expert Systems.
Westinghouse believes that by properly allocating tasks to man and machine and by continuing to press for the human factors community for principles applicable by I&C engineering personnel and for expanding the human factors understanding as applied to nuclear plants, that the growth of automation in nuclear plants can both aid in improving the safety of these plants as well as improve their cost effectiveness.
D. Duffey, EPRI, discussed the research on automatic monitoring and control functions of nuclear plants. Based on industry (including NASA, DOE and SRP) inputs, the EPRI research program in this area is to make the machine more tolerant to human intervention and error and to assist i
the operating and emergency staffs in decision making. He stated that the application of this new technology to existing power plants is the V
subject of current research and that implementation depends on actual benefits to plant operation, i
R. Duffey, in parting, stated that a seminar on SPDS implementation is scheduled on May 6-8, 1986, at the Lafayette Hotel in Boston and that the ACRS is invited to attend.
W. Bertch, Energy Inc., discussed the potential for and use of artificial intelligence (AI) in the nuclear power industry.
He stated i
that an expert system is not a close loop control process but is an operator aid because you are automating the observation, interpretation and decision for the operator and giving him the choice as to either pick his own decision or concur with the computer. The current capabilities are to focus on providing the correct recomendation to the I
0
HUMAN FACTORS 11 March 19-20, 1986 Meeting v
user. Justification for supporting the conclusions is provided which I
allows for verification of the conclusions.
i He stated that certain types of knowledge cannot be encoded with the current technology and trat validation forces all applications to fit the domain of the knowledge base. The operator could also overrely on the expert systems and following its recommendation would be irrestible.
Current industry developments are in the areas of technical specification monitoring, operator aids, diagnostic systems, and performance aids. Other domains within artificial intelligence include 4
the understanding of natural language learning, vision and robotics.
In these areas work has been started but results are not expected in the i
near future.
l George Niederauer, Energy, Inc., briefly discussed computers in nuclear iO power plants.
Currently computers are mainly used in monitoring roles and that little analysis and control are performed by computers.
Computers in the future should be used to make plants and personnel more expert in order to relieve people of tedium and make their jobs smarter.
The computer should be incorporated in a total unified plant information system that will perform analysis and design of support and control functions and be implemented by component, system and plant.
,i D. Jones, NRC, reviewed the 1985 progress of the Human Factors Program Plan (HFPP). The HFPP essentially started after the TMI era. This three-year program is now ended.
In the HFPP there were 33 TMI action items that pertained to human factors. All but six of those items are now complete.
Of the remaining six, several, and all, long-term items will require more than the three years to complete. Others are essentially complete and will be pursued by INP0 and NUMARC under a j
~
memorandum of understanding with NRC.
INP0 has been assigned the training and qualifications activities and NUMARC is working on l
self-initiatives on self-regulation. Also, 51 additional human factors w
HUMAN FACTORS 12 March 19-20, 1986 Meeting A,-
i 1
w/
items were recomended by the Human Factors Society.
NRR accepted a majority of these recomendations as being appropriate and a few recomendations were so-called industry activities, one of which is natural selection testing.
Implementation of these recommendations is in progress, although not as prompt and effective as NRR would want it to be--in part because of industry self-regulation.
A sumary of the elements of the HFPP was then presented. Of the seven major elements, D. Jones eliminated the discussion on the human reliability program because this element is essentially funded by the probabilistic risk assessment group, while the others are funded by NRR.
The six major elements discussed and a brief sumary of each is listed below.
O Staffing and qualifications. Only reactor operators at this time
,U need to be licensed. Regarding educational qualifications, it was stated that specialized course work, A.S. degree and B.S. degree, educational programs are similar in their job relevance in tems of coverage of knowledge list content remains to be covered outside existing educational programs. Approximately half or more of the academic knowledge list content remains to be covered outside existing educational programs.
It appears there is substantial reliance on plant training programs for coverage of academic knowledge necessary for licensed operator positions.
D. Jones stated that about 15% of current operators are degreed.
Training. A policy statement has been issued by the Comission accepting INP0 accreditation of training. NRR will evaluate INP0's training program for the next two years and report to the Comission its recomendation. NUREG/CR-4344 has been published on team skill training.
a 9
e
HUMAN FACTORS 13 March 19-20, 1986 Meeting
/%.
(
)
v' Licensing Examinations. Rulemaking (10 CFR Parts 55 and 50) is in the works on operator licensing qualifications, medical evaluation and simulators for exams. An examiner's handbook that tells how to prepare exams was prepared (NUREG-1121).
Procedures. An E0P inspection module which is used by inspection enforcement, resident inspector and others on their trips out has been developed. He stated that research on various alternative techniques and formats in presenting procedur information has been completed. The results of this study indicate that it doesn't make much difference what the technique or format is, as to presenting the procedure information, as long as it is presented logically, clearly and in a sensible manner.
Man-Machine Interface (MMI). At the present time E l deficiencies at local control station designs exist. There are no human h
engineering standards in control room stations. Work in this area for advanced technologies is ongoing.
Management and Organization (M&O). NRR has established a regulatory position on M&O for operating reactors. This position is in agreement with NUMARC.
I F. Rowsome, NRR, reviewed the HFPP, Rev. 2.
The principal differences between Rev. I and 2 of the HFPP are as follows:
Budget cuts have resulted in the elimination of human factors work by RES.
PRA-related human reliability work is ongoing at a reduced level.
Conformance with the Commission's PPG and policy statements of shifting the emphasis or requirements development to auditing and encouraging industry self-improvement.
J
HUMAN FACTORS 14 March 19-20, 1986 Meeting Revision 2 is aimed at a broader audience, and detailed program schedules and milestones are omitted.
The HFPP resources for FY 1986 will include 9 DHF and 3 training professional staff years. The total technical assistance budget for FY 1986 is about $1.34 million of which $175,000 is for the NAS panel, as compared with about $5.3 million for FY 1985. Aside from the NAS panel, only training, operator licensing, procedures and MI are being funded in FY 1986.
J. Bongorra, Jr., NRR discussed the status of the E0P upgrade program.
The objectives of the upgrade program are as follows:
Review origin of E0P upgrade program Reviewtheconceptoftheproceduresgenerationpackage(PGP)
Discuss Staff PGP-EOP audits Discuss results of the E0P upgrade program Review some options for improving the E0P upgrade program.
The requirements for upgrading E0Ps are based, in part, on TMI action plans (NUREG-0660 and -0737), guidelines for the preparation of E0Ps (NUREG-0899) and Generic Letter 82-33.
It was stated that the PGPs consist of plant-specific technical guidelines, writers guide, description of validation / verification programs and a brief description of the training program. The PGP-EOP audits were performed to check if PGP has been properly implemented and to determine if PGP is an effective method of review. A few plants have been audited with PNL assisting.
Results from the plants audited to date are disappointing and several Staff assumptions appear to be not valid. There seems to be little or no coordination of PGP with detailed l
control room design review, and having an approved PGP does not ensure it is used. Evidently in some cases E0Ps were written before the PGPs
f HUMAN FACTORS 15 March 19-20, 1986 Meeting were prepared.
Further, the most frequent problems with E0Ps are with the wtiter'sguide.
J. Bongarra, NRC stated that the options to the E0P upgrade program are as follows:
PGP submittals should be accompanied by sample of E0Ps PGP-EOP audits should be mandatory, not voluntary Accelerate PGP audit and review schedule Increase dialogue with industry.
R. Knoble, Knobel & Associates Inc., under contract to Georgia Power (Hatch), discussed a symptom-based decision tree approach to BWR emergencyoperatingprocedures(EOP). A movie was first shown of the useofasymptom-oriented, fully-integrated (SOFI) flowchart in the Hatch simulator for a complete loss of plant service water.
%J R. Knobel stated that the need was to create symptom-based E0P from BWROG symptom-based emergency procedure guidelines. With help from Essex Corp., Knobel and Asso, identified deficiencies in TMI E0Ps, as listed:
Single event oriented Confusion as to which procedures apply Excessive burden on operator memory System response seldom indicated Deficient in diagnostic information Contained many cross references Generally contained no charts or graphs Usability with multiple failure poor Operator awareness of total plant conditions questionable.
As a consequence, the operators do what they think is appropriate in i
c
/]O
\\
multiple event failures.
i i
\\
O
HLNAN FACTORS 16 March 19-20, 1986 Meeting (D
)
As a result, R. Knobel stated that a completely integrated all current E0Ps and sympton-based emergency guidelines into flow charts are used for immediate actions and prose for subsequent actions. At least two plants, Hatch and Brunswick, use this method.
W. Williamson, TVA, discussed Browns Ferry implementation of upgraded E0Ps in a single column instruction format. He stated that emergency procedure guidelines in the single column instruction fonnat exist for RPV control, primary containment control, secondary containment control and radiation release control. ATWS emergency procedure guidelines are still event based. The advantages of the single column format are as follows:
Easiest to write Most familiar to the operator Allows the writer choice of methods for identification of equipment
/]
controls and displays V
Implicit is less training required.
Its disadvantages are that it could 1) require more pages to present the necessary information and 2) be difficult to present contingency actions.
It was stated that at TVA the operators write the emergency guideline procedures.
The subcommittee stated that under pressure the operator may not be able to cope with the difficult cross-referencing and its high order of subheading.
Future Subcommittee Activities A subcommittee report to the full ACRS is planned for the April 1986 meeting.
/T NOTE: A transcript of the meeting is available in the NRC Public Document Room, 1717 H St., N.W., Washington, D.C., or can be
(
purchased from ACE-Federal Reporters, 444 N. Capitol Street, WashThgtonP D,C. 20001(202)347-3700.
7
8265 Federal Register / Vol. 51. No. 46 / Monday. March 10, 1986 / Notices e
functions in nuclear power plants to annual resiew, all petitions to modify
[ v 'Ithe proposed determination
}
ames final. an opportunity for a relieve the burden on plant operators the list of articles eligible for duty free.
(A anng will be published in the Federal and enhance safety.(2) review 19e5 tnatment under the Generalized System Register at a later date and any hearing prostress on the ilumen Factors Program of Preferences (CSP) and requests to request wdi not delay the effectise date Pfan. and (3) be briefed on the status of review the CSP status of any beneficiary of the amendment.
EOP tmptementation.-
developing country must be received no If the Commission decides in its final Oral statements may be presented by. later than the close of business, determination that the amendment does members of the public with the Monday. June 2.1986. The CSP provides concurrence of the Subcommittee for the duts-free importation of invulve a significant hazards Chairrnaru written statements will be qualifying eligible articles when consideration. a notice of opportunity fc,r a prior hearing will be published in accepted and made available to the imported frosi deslanated beneficiary the Federal Register and,if a hearing is Committee. Recordings will be permitted developing countries.The CSP la only durin one pornons of the authorized by Title V of the Trade Act granted. it will be held before any amendment is issued.
muung w a transcript is being kept.
of 1974. as amended, and has been and questions may be asked only bY implemented by Excecutive Order 11888 The Commission is seeging pubyic members of the Subcommittee.its of Novernber 24.1971 and mofified by
[e ermi t on of no i ificant bazards fo"ma e oral stat sh$uYd ti y siden ao consideration. Comments on the the ACRS staff member named below as Intmsted parHes or le.gs proposed determination may be far in advance as 's practicable so that telephoned to Elinor C. Adensam.
appropriate arrangements can be made.
gogrnments may suW peunons (1) to designale adddional articles as eligible Director of BWP Project Directorate No.
During the initial portion of the for the CSP. or (2) to withdraw, suspend
- 3. by collect call to (301) 492-8180 or meeting. the Subcommitte. along with I submitted in wrinng to the Secretary of any ofits consultants wbo may be or limit GSP duty. free treatment the Commission. U S. Nuclear present, rosy enchange preliminary acrorded either to eligible at:icles under the CSP or to individual beneficiary
.megulatcry Commission. Washingtes-views regarding mattera to be
. DC 3e55% Attic Docketing and Service cons:dered during the balance of the developing countnes with respect to specific CSP eligible articles or (3) to kanch. All comments received by meeting.
March 14.1966. will be considered in The Subcommittee will ther. hear otherwise modify CSP coverage, Also, naching a final determination. A copy presentations by and hold discussions any person mey file a request to have I
of the application may be examined at with representatives of the NRC Staff.
the CSP status of any eligible t
the Commission's Public Document its consultants. and othee interested beneficiary developing country 9m.1717 H Street. NW., Washington, persons regarding this review.
reviewed with nspect to any of the and at the Local Public Document Further rnformation regarding topics designation critieria hsted in
..som. Richland Public Library. Swift to be discussed. whether the meeting subsections 502(b)or502(c)of the Act and Northgate Streets. Richland.
has been cancelled or rescheduled. the (19 U.S. 3661 [b] and (c)).
Weshington 99352.
Chairman's ruling on requests for the Petitions and requests to modify CSP opportunity to present oral staternent's treatment should be subtrdtted In DeJed at Bethsda Maryland tbs 4 b day of and the time allctied therefor can be English. In 20 co les. In conformity with Sad ten obtained by a prepaid telephone call to regulations cod ed in 15 CFR. Chapter For the hudear Regdatory Commisu
- the cognizant ACRS staff member.Mr.
XX, especia!!y Part 2007 and addressed EEcor G. Adsasaro.
lohn Sch1[fgens (telephone 202/634-to the Chairman CSP Subcommittee.
Direc' tor. swA Pecct Directorate No.1 1414) between Et15 A.M. and 500 P.M.
Trade Policy StaU Committre. Of!!ce of Derision of 5MI Licensiny Persons planning to attend this meetin8 the United States Trade Representative.
are urged to contact the above named Room 517. 600 Seventeenth Street. NW..
(FR Doc. 86-316a Fded 3-74 8 45 eml Individual one or two days before the Washington DC20506.Further enasmo cooe meea scheduled meeting to be adsised of any information may be obtained by calling
,,/
changes in schedufe. etc., which may the GSP Information Center at (202) 395-have occurred.
eg71.
V Adyttory Committee on Reactor g
Safeguards. Subcommittee on Human Datest Wrch 4.1986.
Prospective petitioners and requestors Factors; Meeting Mortes W. IAarkin.
are strongly advised to review the CSP A ssistant Esecutive Duveforfar project regulations published in the Federal i
The AChts Subcommittee on lluman Rens,.
Register on Tuesday. February 11.19e6.
Factors wiD hold a rueeting on March 19 (rg Dec.aHtaa Filed s-r-as ess sul (51 FR 5035). Petitions and requests that and 20.1986. Roorn 1048,1717 H Sucef.
so,,a cooe me.ew do not conforrn to the requirements of 7
NW., Washington. DC.
these regulations wdl not be accepted i'
The neeeting wel be spes to public I0f 8"I*"
OFFICE OF THE UNITED STATES Notice of petitions and requests attendance.
TRADE REPRESENTATIVE accepted for review will be published in r
He agenda for subjecf meeting shall the Federal Register on or about be as follows:
Trade Policy Staff Committee; b
Wednesday. March 19.1988-8 30 A M.
Generaltzed Systam of Preferences
[",'
T h h unt!! the conclusion of business.
(GSP);Deadtlne for Acceptance of
. opportunity for IrWerested parties to Thrueday. March 20,19as-a.30 A.M.
Petitions Requesting Modification of comrnent en petitions and reqaests
. antil the concluston of business.
Uel of Art!cles Etigtble for Duty Free eccepted for review through public The Subocamittee wili a at to:(1)
' beadngs and written submissions Any we ; h h ;'vea est. la order to naoddications to the CSP reeulting from morr of the mcPor? ; r
- c o.
.n 1-P c.o & : t t44 CSP
,,,the itM CSP raenualreview will be,.,4:
b
. amine thotentti' fo... as at:r :
d T7MC//r/3, ;V/
c I
March 12, 1986 REVISION 4
PROPOSED SCHEDULE FOR THE MARCH 19-20, 1986 MEETING OF THE ACRS SUBCOMMITTEE ON HUMAN FACTORS Wednesday, March 19, 1986 Y:8 30am Opening Statement D. Ward 10 min 35 V 8:40 NRC Attitudes Toward Automation
- Past and Present Comments on the 9/8-12/85 Pasco Topical on Computer Appl-gV3 ications for NPP Operation and Control S. Weiss '
35
('
SD r sa M u ah h fl & l'
~
v 9:M Summary of SRP Presentation to the NRC Staff on Advanced l
Control Room Design L. Beltracchi 30
-S+46 Summary of NASA Presentation to N
W. Kennedy /
the NRC Staff on NASA's 30 4.0 Experience with Automation 10:M NRC Attitudes Toward Automation
- Future W. Kennedy 30 2L
- 10:%.
BREAK 15 4m 4'.10 Cognitive Allocation and the Control Room (U. Ill./SRL)
M. Paradies / 45 q
/mbedded b ones!
u Research Activities Trafning with Automated B.
~~
Syrcems (U. Ill.)
C. Hopkins 30 h \\ S'
--12.15 LUNCH 60
~
.e
. n.
1-March 12, 1986
.,. SCHEDULE / HUMAN FACTORS-..1986 REVISION 4
MARCH 19-20, MEETING
~
?_: 4 r
4 +-16pm Advanced Control Room Design J. Carrera/
60 min
(
and Evolution for Current J. Easter,
Control Rooms (Westinghouse) %a
^~
N Supervisory Information Management Alarm Management Automated Procedures - Normal and Emergency 2:15 Ho to Incorpo at Automat n
i to a N w P1 t Des gn (G ner Electr Pug 6
,.15 -
BREAK 15 5' >5
&% 3:30 Research on Automating Monitoring e
and Control Functions of NPPs
- p4-(EPRI)
R. Duffey 60 Procedure Tracking J?.cq.g Feedwater Control 0
Core Mo.titoring N
Computers in NPPs (Energy Inc.)
G. Niederauer/ 75 An Overview W. Bertch O
Automating Control Functions Use of " Artificial Intelligence" 4
and " Expert Systems" C4 4 EsAL.) 6M N
de N) 5.iSym ADJOURN
.g p't.gUNy(,.
Thursday, March 20, 1986
/
Q ws.J. W V 8:30am Review of 1985 Progress on HFPP L.;_:::
60 Discussion of " Ongoing" HFPP F. Rowsome 80 101 OG f o '. l f 10. ",0 BREAK 15 t
-10. &
Status of EOP Implementation J. Bongarro 80 IDtT Reviews of PGPs 4
Results of Staff EOP Audits
[
Discussion of Approaches
-b i
es 0
e
SCHEDULE / HUMAN FACTORS March 12, 1986 MARCH 19-20, 1986 REVISION 4
MEETING Taken t.o EOPs y
l ~
E0P Contents / Formats o
12' SO p
LUNCH 60 min 1:40
.S5 EOPs
" Flowchart" Format (GPC)
J. Springer /
i R. Knobel 120 t
%'.l 5
-4::iSpm BREAK 15 390 Sgro EOPs
" Single Column" Format (TVA)
W. Williamson 60 4.b 9 0 __ S (J q d g a u.ss um 6
ADJOURN g
- a s
,o
's o
- ~
1 8' b l
_U
m.
~-~~ ~
~ *ew a
..,m.
(,
LIST OF HANDOUTS ATTACHMENT C ACRS SUBCOMMITTEE ON HUMAN FACTORS MARCH 19-20, 1986 MEETING 2
WASHINGTON, D.C.
1.a S. Weiss, NRC, Regulatory Criteria for Use of Computers in i
Nuclear Power Plants t
1.b
- 5. Weiss, IE Information Notice 86-10 Safety Parameter Display System Malfunctioning 2.
L. Beltracchi, NRC, Survey of Advanced Control Room Design 3.
W. G. Kennedy, NRC, Sumary of NASA Presentation to the NRC Staff' on NASA's Experience with Automation 4.
W. G. Kennedy, NRC, NRC Staff Attitudes Toward Automation - Futurp L-s' 5.
M. W. Paradies, U. of Illinois, Urbana-Champaign, A Cognitive A
e Allocation to Improve Nuclear Power Plant Performance; also Cognitive Allocation and the Control Room 2
i 6.
B. G. Jones, U..of Illinois, Urbana-Champaign Studies of Improvei Operational Safety and Performance of Nuclear Power Plants O
7.
J. R. Easter, Westinghouse, Advanced in Nuclear Power Plant Automation: A View from Westinghouse 8.
R. B. Duffey, EPRI, Research on Monitoring and Control Functions l
9.
W. J. Bertch, Energy Incorporated, The Potential for and Use of
~.
Artificial Intelligence 10.a D. B. Jones, NRC, Review of 1985 Progress on Human Factors Program Plan (HFPP) 10.b D. B. Jones, Specialized Educational Programs' for Nuclear Reactor j.
Operators e
11.
F. H. Rowsome, NRC, Human Factors Program Plan, Revision 2 12.
J. P. Bongarra, Status of the E0P Upgrade Program i
[
13.
R. Knobel, Knobel Assoc., and J. S rin er, Georgia Power. Use of Symptom Oriented Fully Integrated SOF ) Flow Charts in Hatch Simulator for Complete Loss of P1arit Service Water; Emergency
-Procedure Guideline Implementation Program, E. I. Hatch Nuclear Plant 14.a W. Williamson, TVA, Implementation of Upgraded Emergency Operating Procedures (EOPs) in Single Column Instruction Format 14.b RPV Control E01-1 ATTACHMENT C 9
_,,,, _ --c
..ne.._,,-
_~_,,-n,,,..-
,_, n,--,.,_.
e.,
o NRR STAFF PRESENTATION TO THE o
ACRS SUBJECT HUMAN FACTORS PROGF, TAM PLAN:
PROGRESS AND PROSPECTS DATE:
ApRit 11, 1986 PRESENTER:
DANIEL B'.
JONES
- O PRESENTER'S TITLE / BRANCH /DIV
- PROJECT MANAGER: DHFT/HFIB PRESENTER'S NRC TEL. NO.:
x24879 f,
SUBCOMMITTEE:
-~
O STAFFING AND QUALIFICATIONS MINIMUM QUALIFICATIONS REQUIREMENTS
- LICENSINGOFADDITIONALOPERATIONhPERSONNEL--NOACTI EDUCATIONAL QUALIFICATIONS FOR OPERATIONS PERSONNEL NUREG/CR-4501 NUREG/CR-4411 O
RE6ULATORY GUIDE 1-8 -- PERSONNEL QUALIFICATIONS AND TRAINING SUPERVISOR (SRO AND STA) EFFECTS ON CREW PERFORMANCE IN SIMULATOR NUREG/CR-4280
- NO DIFFERENCE IN CREW PERFORMANCE 4
O
~
..-......-..-.n.-.
. -.... - ~.
r i
1 O
STAFFING AND QUALIFICATIONS LIMITS AND CONDITIONS OF SHIFT WORK SHIFT SCHEDULING AND OVERTIME (NUREG/CR-4284)
- BASIS TO REVISE NRC POLICY ON OVERTIME
- GUIDANCE ON ROUTINE 12-HOUR SHIFT Q
OPERATOR FEEDBACK -- MAILED SURVEY (NURE'G/CR-4139)
- EFFECTIVE WAY FOR NRC TO ACQUIRE INFORMATION:
RE:
OPERATORS
- PERSONNEL ATTITUDES: RE: STA SHIFTWORK, DVERTIME, AND STAFFING
~
O f
a
- O TRAINING j
l I
TRAINING REGULATION AND GUIDANCE i
l POLICY STATEMENT ON TRAINING AND QUALIFICATIONS
- INPO M0A ON ACCREDITATION TEAM SKILLS TRAINING -- NUREG/CR-4258 O
INSTRUCTIONAL SKILLS EVALUATION -- NUREG/CR-4344 NRC TRAINING EVALUATION PROGRAM
[
1 l
TRAINING REVIEW CRITERIA AND PROCEDURES (FOR NRC STAFF)
O
/
e-
- - ~.,. -. -.. - -
+ -
\\$.
o LICENSING EXAMINATIONS
~o PROCESS s.
RULEMAKING:
10 CFR PART 55 AND PART 50 i
k
- R. G. 1.8 -- PERSONNEL QUALIFICATIONS l
l i; f-
- R. G. 1.134 -- MEDICAL EVALUATION l!!!
l I-
- R. G. 1.149 -- SIMULATORS FOR EXAMS t
40 IMPROVING SIMULATOR EXAM PROCEDURES I
SIMULATION FACILITY EVALUATION CONTENT PWR KNOWLEDGE AND ABILITY CATALOG (NUREG-1122)
EXAMINERS HANDBOOK (NUREG-1121)
COMPUTERIZED EXAM QUESTIONS BANK
= > =.
/
O PROCEDURES E0P UPGRADE AND POST-lMPLEMENTATION AUDITS REVISED SRP 13.5.2, "0PERATING AND MAINTNANCE PROCEDURES" E0P INSPECTION MODULE (FOR ISE)
METHODS FOR EVALUATING ALTERNATIVE TECHNIQUES AND FORMATS O
I
{
MAN-MACHINE INTERFACE (MMI)
[ ()
MMI FOR EXISTING DESIGNS DEFICIENCIES IN LOCAL CONTROL STATIONS (NUREG/CR-3696) l ANNUNCIATORS COMPUTERIZED ANNUNICATOR SYSTEMS (NUREG/CR-3987)
ALARM REDUCTION TECHNIQUES
($)
MMI FOR ADVANCED TECHNOLOGIES SIMULATOR EVALUATION OF CRT DISPLAYS (NUREG/CR-3767)
EXPERIMENTAL ASSESSMENT OF AN EXPERT SYSTEM (NUREG/CR-4272) f DECISION MAKING UNDER STRESS (NUREG/CR-4040)
GUIDELINES FOR EVALUATION OF VDUs (NUREG/CR-4277)
HUMAN PERFORMANCE IN NON-DESTRUCTIVE TESTING (NUREG/CR-4277)
O l
i l
2 MAN-MACHINE INTERFACE (PMI) (CONT'D)
REGULATORY DOCUMENTS SRP 18,0 -- HFE SRP 18.1 -- CONTROL ROOM SRP 18.2 -- SPDS e
o 1
e up O
., _,. _., _,. _ _ _.... _, _ _, _ _ ~ _.
O MANAGEMENT AND ORGANIZATION (M40)
ESTABLISH REGULATORY POSITION - ors 4
SAFETY INDICATORS (NUREG/CR-3737) s EMPERICAL ANALYSIS OF MAINTENANCE FACTORS (N REVIEW OF INP0 SAFETY EVALUATION MATERIALS ASSESS PROCEDURES FOR Ols GUIDELINES AND WORKBOOK FOR ASSESSING ORGANIZATIO ADMINISTRATION FOR Ols (NUREG/CR-4125)
~
e e
g O
~
A*
_ _ _ es em ove ooek *j
?"
. w3, __.,. _:
h
O PRINCIPAL DIFFERENCES BETWEEN REVISION 1 AND 2 0F THE HUMAN FACTORS PROGRAM PLAN BUDGET CUTS RES HAS TERMINATED HUMAN FACTORS RESEARCH OTHER THAN THE PRA-RELATED HUMAN RELIABILITY WORK AND THE NRR BUDGET IS REDUCED
($)
CONFORMANCE WITH THE COMMISSION'S PPG AND POLICY STATEMENTS:
SHIFT FROM EMPHASIS ON REQUIREMENTS DEVELOPMENT TO AUDITING AND ENC 0URAGING INDUSTRY SELF-IMPROVEMENT REVISION 2 IS AN NRR PLAN ASSIGNED TO DHFT; REVISION 1 COVERED ALL ED0 OFFICES REVISION 2 IS AIMED AT A BROADER AUDIENCE:
DETAILED PROGRAM SCHEDULES AND MILESTONES ARE OMITTED 4
0
([)
HFPP RESOURCES O
B.
TECHNICAL ASSISTANCE BUDGET FY 1985 FY 1986 NRR' RES TOT.
0 1.
STAFFING 8 QUALIFICATIONS 125 115 240 0
50 2.
TRAINING 445 0
445 50 413 3.
OPERATOR LICENSING 695 476 1171 413 0
34 4.
PROCEDURES 300 560 860 34 660 5.
MAN-MACHINE INTERFACE 0 1032 1032 660 6.
MANAGEMENT a ORGANIZATION 250 115 365 0
0 0
7.
HUMAN RELIABILITY 0 1100 1100 0
0 0
7.
(NEW) HUMAN PERFORMANCE 8.
NAS PANEL 43 132 175
(])
TOTAL 1815 3398 5213
1.
TRAINING (MAINTENANCE AND TRAINING BRANCH)
{)
BACKGROUND POST-TMIRECOMMENDATIONSkNDTHENUCLEARWASTEPOL ACT OF 1982, SECTION 306, LED STAFF TO PROPOSE A TRAINING RULE (SECY-8k-76, 76A, 76B)
COMMISSION ELECTED TO PROMULGATE POLICY STATEMENT (FR 11147, MARCH 20, 1985) IN LIEU OF RULE TO ALLOW INDUSTRY TWO YEARS TO IMPLEMENT THE INP0 ACCREDITATION PROGRAM O
CURRENT AND PLANNED ACTIVITIES EVALUATE IMPLEMENTATION OF POLICY STATEMENT POST-ACCREDITATION REVIEWS PERFORMANCE-BASE INSPECTIONS OBSERVE INP0 TEAM VISITS AND ACCREDITATION BOARD O.
'"^'"'"S (' "")
.O SALP ARCHIVAL PERFORMANCE DATA INTERIMPROGRESSRkPORT(MARCH 1986)
ACCREDITATION STATUS (MARCH 1, 1986) 310 0F 610 PROGRAMS AT 54 SITES " READY FOR ACCREDITATION" O
^
PROBLEMS IDENTIFIED WITH " ANALYSIS" AND LEARNING OBJECTIVES, ELEMENTS NRC AND INPO DISCUSSING IMPROVEMENTS FINAL EVALUATION REPORT AND RECOMMENDATION FOR FUTURE ACTION (MAY 1987)
REVISE SRP SECTION 13,2 BASED ON 1986 DECISION O
($)
III, PROCEDURES (HUMAN FACTORS ISSUES BRANCH)
SCOPE:
ALL OPERATING PROCEDURES.
MAINTENANCE PROCEDURES COVERED BY M8SPP CURRENT AND PLANNED ACTIVITIES:
EMERGENCY OPERATING PROCEDURES (E0Ps)
GUIDANCE FOR NORMAL PROCEDURES (0P) AND ABNORMAL OPERATING PROCEDURES (A0P) g-)
HUMAN FACTORS DEFICIENCIES REGULATORY ANALYSIS TO DETERMINE IF REGULATORY GUIDANCE JUSTIFIED TRANSITIONS BETWEEN PROCEDURES:
OPERATORS HAVE DIFFICULTY IN TRANSFERRING BETWEEN PROCEDURES.
DRAFT GUIDANCE ON PROCEDURE INTERFACES BEING DEVELOPED.
VALUE-IMPACT (V/I) ANALYSIS OF E0P UPGRADE:
RES COMPLETING RETROSPECTIVE V/I STUDY OF THE CHANGE FROM EVENT-BASED TO FUNCTION-BASED E0Ps.
RISK BASED VALUE ASSESSMENT AND COST DATA TO SHARPEN FOCUS
(])
OF FUTURE DEVELOPMENT WORK IN PROCEDURES.
O i i i ea0CeDuRES (CONT D)
-i STRESS OF ACCIDENTS ON OPERATORS:
RES STUDY WITH RECOMMENDATIONS, HFIB WILL, WHEN RESOURCES PERMIT, EVALUATE WHAT FOLLOW-UP ACTIONS ARE APPROPRIATE, 1
RELATED RESEARCH EPRI ADVANCED OPERATOR AIDS / EXPERT SYSTEM HALDEN STUDY OF COMPUTERIZED PROCEDURES O
i i
e 1
l O
w-,,--
-nw
-e---,-v
--,,,--,,,-we-,
,,,--,e-~,
-,e--,w, e
-o,--
,,----,m--.,,-.,-a,-r
---m r--,
m,,
IV, MAN-MACHINE INTERFACE (HUMAN FACTORS ISSUES BRANCH) c3 U
BACKGROUND HFIBISTRACKINGTHEEVALUATION5BYTHEDIVISIONSOF LICENSING 0F CONTROL ROOM DESIGN REVIEWS AND SAFETY PARAMETER DISPLAY SYSTEMS
'AN IE INFORMATION NOTICE HAS BEEN ISSUED ON SPDS PROBLEMS, P00R AVAILABILITY UNRELIABLE DATA AND ALARMS O
POOR OPERATOR ACCEPTANCE AND/0R INTEGRATION INTO E0Ps WE WILL EVALUATE THE NEED FOR ADDITIONAL REGULATORY ACTION 1
CURRENT PLANS LOCAL CONTROL STATIONS C0KMONLY HAVE HF PROBLEMS CONDUCIVE TO HUMAN ERROR.
WE PLAN TO SCOPE THE PROBLEMS, EVALUATE THE V/I INCENTIVES FOR REGULATORY OPTIONS, O
n s
,-.,e-r--
IV, MAN-MACHINE INTERFACE (CONT'D)
(g b
OPERATOR OVERLOAD DUE TO ANNUNCIATORS:
THE NRC HAS YET TO RESOLVE THE PROBLEM OF T00 MANY ALARMS / ANNUNCIATIONS DURING SIGNIFICANT UPSET EVENTS, RESEARCHONADVANCEDANN0NCIATORSYSTEMSHASBEENDONEBY RES, IS UNDERWAY AT HALDEN AND EPRI, THE STAFF WILL FORMULATE OPTIONS FOR RESOLUTION AND PREPARE REGULATORY
- ANALYSES, ADVANCED CONTROLS, OPERATOR AIDS, AND DISPLAYS:
O THE STAFF IS TRACKING DEVELOPMENTS IN BOTH THE NUCLEAR AND OTHER INDUSTRIES.
WE ARE GEARING UP FOR SAFETY EVALUATIONS OF LICENSEE PROPOSALS TO MINIMIZE THE REGULATORY DISINCENTIVES AND REWARD THOSE LICENSEES WHO PROPOSE TO IMPLEMENT ADVANCED CONTROLS AND DISPLAYS, 2
O a
O V.
STAFFING AND QUALIFICATIONS (HUMAN FACTORS ISSUES BRANCH)
CURRENT AND PLANNED ACTIVITIES,
POLICY STATEMENT ON ENGINEERING EXPERTISE ON SHIFT ISSUED, OCTOBER 1985 COMMISSION PAPER ON INDUSTRY PLANS FOR IMPLEMENTATION TO BE DEVELOPED BY JULY O
COMMISSION REQUESTED AND RECEIVED A DRAFT ADVANCE NOTICE OF RULEMAKING REQUIRING DEGREES OF SR0s IN FEBRUARY POLICY STATEMENT ON SHIFT WORKING HOURS AT THE COMMISSION'S REQUEST, THE STAFF IS EVALUATING THE NEED TO UPDATE THE CURRENT POLICY STATEMENT AND THE TWO GENERIC LETTERS 82-12 AND 82-16.
O
4 O
V.
STAFFING AND QUALIFICATIONS (CONT'D)
UPDATES ARE-IN THE WORKS FOR REG, GUIDE 1,114, " GUIDANCE TO OPERATORS AT THE CONTROLS AND TO SENIOR OPERATORS,,."
AND THE CORRESPONDING SRP SECTION 13,1,2, "0PERATING ORGANIZATION" TO CONFORM WITH THE STAFFING RULE 1
10 CFR 50,54(M)(2)(III).
I O
l
)
I 4
s O
r--*
~v
---,,,,,-,-,,,e+
,,m,,,,,,-e.,,-,,,.-,,---.,,--_.w,,-n,c.m---,.,
-w--
,,w_,,,,,,,,,
---en,_,_e,-,,,.,--,,r
-m,,--,
e,----
O VI.
MANAGEMENT AND ORGANIZATION (HUMAN FACTORS ISSUES BRANCH) e 4
ALL WORK ON MANAGEMENT AND ORGANIZATION UNDER THE HFPP IS I
SUSPENDED:
THE COMMISSION'S " POLICY AND PLANNING GUIDANCE, 1986" CALLS FOR A MOVE TOWARD PERFORMANCE-BASED PROGRAMS.
Q THE NRC IS PLANNING TO BE MORE RESPONSIVE TO ACTUAL INDUSTRY PERFORMANCE IMPROVEMENTS TO SALP ARE UNDERWAY f-DEVELOPMENT OF PERFORMANCE INDICATORS.
l 1
!O l
t
+w- -
,,wewm-,
,--_._,,----.m.
---,-----,------c,
,...,. - -,.w
..,. - +
O VII. HUMAN PERFORMANCE (HUMAN FACTORS ISSUES BRANCH)
A.
DEVELOPMENT OF IMPROVED HUM N PERFORMANCE EXPERIENCE REPORTAGE CONCERN:
CAUSITIVE FACTORS OF HUMAN ERRORS (AND NOTABLE SUCCESSES) ARE ESCAPING REPORTAGE.
1.
HFIB IS WORKING WITH AE0D TO DEVELOP A PROTOCOL FOR INCIDENT INVESTIGATION TEAMS TO GUIDE THE INVESTIGATION OF THE ROOT CAUSES OF UNUSUALLY P0OR (OR GOOD) OPERATOR PERFORMANCE DURING THE INCIDENT.
2.
HFIB PLANS TO WORK WITH AE0D ON LER REPORTAGE GUIDANCE, AND WITH IE AND THE REGIONS ON AUGMENTED INSPECTIONS.
3.
WE PLAN TO AUDIT AND EVALUATE THE TREATMENT OF HUMAN PERFORMANCE IN INP0'S SEE-IN PROGRAM AND THE NEW HUMAN PERFORMANCE EVALUATION SYSTEM.
O
l O
I VII HUMAN PERFORMANCE (CONT'D)
B.
RISKANDPRECURSORSTdDIES CONCERN:
WE NEED BETTER VALUE/ IMPACT TOOLS TO ASSESS PRIORITIES, 4
BUDGETS, PERFORM REGULATORY ANALYSES, BACKFIT RULE, O
PLANS:
WORKING WITH DRA0 ASSISTANCE, WE PLAN TO REVIEW SELECTED PPECURSORS AND PRAs TO ASSESS THE EXISTING HF REGULATIONS FOR RISK RELEVANCE, IDENTIFY GAPS IN THE REGULATIONS, AND PREFABRICATE MODELS OF THE DIFFERENCES IN RISK TO BE EXPECTED OF CHANGES IN HUMAN RELIABILITY IN EACH OF THE CONTEXTS WE-D0 (0R SHOULD) REGULATE, FOR USE IN PRIORITIZATIONS, TO FOCUS DEVELOPMENTAL PROGRAMS, AND IN V/I STUDIES FOR STANDARDS DEVELOPMENT, O
,-~,m--<---aw
. w e
- ------, -- -~--
,ew~---,-n,-----
,--y-,
.r-
,----,m,
-g--.
?
- O 0PERATOR LICENSE EXAMINATIONS (OPERATOR LICENSING BRANCH)
G0AL 1
ENSURE THAT NRC-ADMINISTERED EXAMINATIONS GIVE A CONSISTENT, VALID MEASURE OF R0 AND SR0 COMPETENCE FOR SAFE OPERATION q
CURRENT AND PLANNED ACTIVITIES L
CONTINUE DEVELOPMENT OF CATALOGS OF R0 AND SR0 KNOWLEDGE i
1 AND ABILITIES BASED INITIALLY ON THE INP0 JOB-RELATED ANALYSES (I)
VALIDATION BY STAFF AND SUBJECT MATTER EXPERT REVIEW (II)
ASSESS NEED FOR PLANT-SPECIFIC CATALOGS TO SUPPLEMENT GENERIC CATALOGS HANDBOOK FOR EXAMINERS WILL BE PILOT TESTED AND IMPROVED NECESSARY, E.G., CONSTRUCTION OF SIMULATOR SCENARIOS l
EXAMINERS ARE BEING TRAINED ON FUNCTION-BASED PROCEDURES 1
THROUGH SPECIAL AND REFRESHER TRAINING COURSES lO i
O OPERATOR LICENSE EXAMINATIONS (CONT'D),
UPDATES ARE PLANNED FOR 10 CFR 55, "0PERATORS' LICENSES" R.G. 1.149, " NUCLEAR POWER PLANT SIMULATOR FOR USE IN OPERATOR LICENSE EXAMINATIONS" t
)
R.G.1.8, "QUALIFICATI0N' AND TRAINING 0F PERSONNEL FOR NUCLEAR POWER PLANTS" R.G. 1.139, " MEDICAL EVALUATION OF NUCLEAR FACILITY PERSONNEL REQUIRING OPERATOR LICENSES" 1
O
.