ML20248B139
| ML20248B139 | |
| Person / Time | |
|---|---|
| Issue date: | 09/27/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1759, NUDOCS 8910030111 | |
| Download: ML20248B139 (358) | |
Text
_ _ _ _ _ _ _ _ - _ _ -
c.
O
. J R G \'A _
UNITED STATES NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS In the Matter of:
)
MEETING: )
) 2 SUBCOMMITTEE ON HUMAN FACTORS )
)
)
O I
I j
Pages: 1 through 281 Place: Bethesda, Maryland l
- "'"'*"{mj*j g.:fCe COD [ letai?
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ifo W -m. ,nit f .iv.o..P
_. .n.r.%w,
- . C..;
O HERITAGE REPORTING CORPORATION 0,fiedelReporters 1229 L Street, N.W., Suite 648 f;j00;jii, - - ;- Wasidagton, D.C. 20005 Y f }' 7,'y - .
(202) 628 4888
1 PUBLIC NOTICE BY THE.
2 UNITED STATES NUCLEAR REGULATORY COMMISSION'S 3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4 September 27, 1989 5
l+
6 7 The contents of this stenographic transcript of 1-8 the proceedings of the United States Nuclear Regulatory 9 Commission's Advisory Committee on Reactor Safeguards 10 (ACRS), as reported herein, is an uncorrected record of the 11 discussions recorded at the meeting held on the above date.
12 No member of the ACRS staff and no participant at.
13 this meeting accepts any responsibility for errors or 14 inaccuracies of statement or data contained in this
() 15 transcript.
16 17 18 19 20 21 22 23 24 25-O Heritage Reporting Corporation (202) 628-4888
UNITED STATES NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS In the Matter of:
)
MEETING: )
)'
SUBCOMMITTEE ON HUMAN FACTORS )
)
)
Wednesday, September 27, 1989 Room P-110, Phillips Building I 7920 Norfolk Avenue-Bethesda, Maryland The meeting convened, pursuant to notice, at 8:30 a.m.
BEFORE: DR.'FORREST J. REMICK, Subcommittee Chairman Associate Vice-President for Research Professor of Nuclear Engineering The Pennsylvania State University University Park, Pennsylvania SUBCOMMITTEE MEMBEht., PRESENT:
MR. DAVID A. WARD Research Manager on Special Assignment E.I. du Pont de Nemours & Company Savannah River Laboratory Aiken, South Carolina MR. CARLYLE MICHELSON Retired Principal Nuclear Engineer Tennessee Valley Authority Knoxville, Tennessee, and Retired Director, Office for Analysis and Evaluation of Operational Data U.S. Nuclear Regulatory Commission Washington, D.C.
(Continued on the next page)
Heritage Reporting Corporation (202) 628-4888
2 SUBCOMMITTEE MEMBERS PRESENT , (Continued) :
MR. JAMES CARROLL Retired Manager, Nuclear.. Operations Support
-Pacific ~ Gas & Electric' Company San Francisco, California ACRS COGNIZANT STAFF MEMBER:
E.- IGNE'-
NRC STAFF PRESENTERS:
. SHER BAHADUR, RES
' SANDRA FRATTALI, RES THOMAS-G. RYAN, HFB/RES CARL E. . JOHNSON, HFB/RES ERASMIA LOIS,.HFB/RES INDUSTRY REPRESENTATIVES:
JOE COLVIN (NUMARC)
RICH ENKEBOLL (NUMARC)
BRUCE MARTINIS (BALTIMORE. GEE)
JERRY ELLIS (SYSTEM ENERGY RESOURCES)
CHUCK SPRUNK (PA. POWER & LIGHT)
O. JOHN ROSS~- (BALTIMORE G&E)
ALSO PRESENT:
LOREN BUSH, NRR MR. MORRIS' CID FELD,,RES contractor MR. COFFMAN DR. AZARM WILLIAM.VESLEY JOHN WRANTHRALL PETER'APPEGUIANI DR. THURBER DR. MODARRES (U. of Md.)
i
' /9 Heritage Reporting Corporation V (202) 628-4888
0 1 E B Q C E E~Q 1 N G E 2 DR. REMICK: The meeting will now come to order.
3 This is a meeting of the ACRS Subcommittee on 4- Human Factors.
5 I am Forrest Remick, Chairman of the Subcommittee.
6 Other ACRS members in attendance today are Mr. Ward, Mr.
7 Michelson and Mr. Carroll.
8 Today's meeting will be to discuss: 1) .the access 9 authorization rule; and 2) performance indicators.
10 E. Igne is the Cognizant ACRS Staff Member for 11 today's meeting.
12 The rules for participation in today's meeting 13 have been announced as part of the notice of the meeting
() 14 that was published in the Federal Recister on September 13, 15 1989.
16 The meeting is being conducted in accordance with 17 provisions of the Federal Advisory Committee Act and the 18 Government in the Sunshine Act. We have received no written 19 or oral statements from members of the public.
20 It is requested that each speaker first identify 21 himself or herself and speak with sufficient clarity and 22 volume so that he or she can be readily heard.
23 A couple of comments here on the agenda. It is 24 possible.that the discussion on access authorization will 25 not take the full morning. It may or may not.
. Heritage Reporting Corporation (202) 628-4888
1 But we are going to see if the staff who are going 2 to discuss performance indicators might be prepared to come 3 in earlier in case we do finish the access authorization 4 - rule earlier.
5 I mention that for those in the audience who might 1
1 6 be interested.
7 The other thing is, some may know that the 8 President of INPO. Zack Pate, sent a letter to James Taylor 9 on September 8 on performance indicators.
10 We had asked the staff under the discussion for 11 performance indicators to come prepared to discuss that 12 letter. But the staff will not be prepared to discuss it 13 today. So I mention that also for anybody who might have
() 14 been particularly interested in that.
15 The staff will not be discussing the INPO letter 16 of September 8.
1 17 The first matter before us, the access 18 authorization rule, if my memory serves me well, goes back 19 many years.
20 I think at one time it was even referred to as the 21 insider rule. Am I correct on that, staff?
22 And I know it goes back to the 1981-82 time frame 23 at least, and probably before that. There has been a lot of 24 discussion. It was initially I think proposed as a rule and 25 then with some industry discussion with the Commission I Heritage Reporting Corporation Os (202) 628-4888 t
l l
0
\
(V~)
1 believe it was decided that it would be developed as a 2 policy statement.
l 3 I think it went out for public comment as a policy l
l 4 statement. More recently the Commission decided that it 5 would issue the access authorization as a rule. So it has 6 been around for some time.
7 Basically throughout that time I think it has had 8 three primary elements: a background check, the 9 psychological screening test, and then the behavioral 10 observation by supervisor.
11 I think those have been the three elements.
12 ACRS indirectly commented back some months ago, 13 because it so happened when we were reviewing the fitness
() 14 for duty rule which came to us by the Office of Nuclear 15 Reactor Regulation, about the same time we are aware that 16 the Office of Research had put out for public comment I 17 believe or were proposing to put out for public comment, the 18 access authorization rule, and both talked about drug 19 testing and so forth. And so ACRS made some comments to the 20 Commission that shouldn't we get our act together. And 21 these things, if they are talking about the same thing, we 22 should know what is going on in the other part of the house.
23 I think it has been under revision since that 24 time. This is the first ACRS has seen it since that time.
25 And without further ado on my part, and before T Heritage Reporting Corporation
(~J
( (202) 628-4888
6 V%
1- asking the staff to summarize what is'in the access 2 authorization rule, I will ask if any other members of the 3 subcommittee.want to make any introductory comments.
4 (No response) 5 DR. REMICK: All right. Mr. Sher Bahadur from the 6 Office of Research is going.to'make the presentations, both 7 presentations on the access authorization rule this morning.
8 So at this time I will turn the meeting over to 9 the staff for their presentation.
10 An d don't hesitate to correct my history, my ad 11 hoc history, if it is incorrect.
12 MR. BAHADUR: Thank you, Mr. Remick.
13 What I am going to do is to give a brief
() 14 presentation using the handouts that are being distributed.
15 and use the viewgraphs just to explain some of the depicted 16 notes that I made on the handouts.
17 (Slides being shown) 18 MR. BAHADUR: My name is Sher Bahadur. I am from 19 the Regulations Development Branch in the Office of 20 Research.
21 And today, we will talk about the access 22 authorization program for the nuclear power plants.
23 As Mr. Remick has pointed out, this rule has been 24 with the agency for about ten years or so. And at this 25 time, going through basic steps, we were just about ready
/ Heritage Reporting Corporation (202) 628-4888
f 7 b with the final rule that, according to the original 1
2 schedule, was to go to the Commission on the 10th of October 3 for their final approval.
4 About two weeks back, we presented this rule to 5 CRGR. The normal progression is, the staff develops a 6 package, takes it to CRGR, gets their concurrenc'e, goes to
- 7. EDO and side by side, presents that to ACRS, as we are doing 8 today, and then incorporating all the comments and 9 submitting it to the Commission for their approval.
10 When we took it to CRGR about two weeks back, they 11 raised certain questions, questions which were very basic 12 and which kind of made us look into some of the issues a 13 little more deeply.
I() 14 The rule is a complex rule. It does require, as l 15 Mr. Remick has pointed out, three basic elements: the 16 psychological assessment, background investigation and then 17 the behavioral observation by the supervisors.
18 None of this can be digitalized into numerical 19 values. All these are very subjective procedures, and there 1 20 is an awful lot of range of emotional opinions about these.
l l- 21 We have been grappling with these complexities 22 ever since we have been trying to develop the rule.
23 When we went to CRGR, the questions were so basic 1
! 24 that we cam = back and started thinking afresh about the 25 rule.
Ileritage Reporting Corporation O- (202) 628-4888
o .1 As a result, the package that you have in your 2 hand today could be somewhat different than how the rule 3 would eventually look.
4 It may go back to the same pattern as you do have.
5 But what I am trying to say is that we are still in the 6 process-of' evolution, we are still in the process of 7 developing the rule so that it is acceptable to all. And it 8 does meet the performance objective that we eventually are 9 trying to achieve.
10 DR. REMICK: Mr. Bahadur, that raises, I think, a 11 couple of questions in my mind.
12 One, I was not aware of this, and it was my 13 impression that the staff was looking for an ACRS letter.
() 14 I think depending on how much further development 15 might be necessary, I think this raises a question, are we 16 at a point where we might want to take it to the full 17 committee'on October or we might want to defer.
18 And the question is, the committee will have to 19 decide whether it wants to write a letter if it is still 20 under development.
21 MR. BAHADUR: That is a very valid comment, sir.
22 And before I mislead you in thinking that the 23 changes will change the philosophy of the rule, let me 24 correct myself.
25 The basic attributes in the rule remain the same, Heritage Reporting Corporation O (202) 628-4888
1 with just the fine tuning, here and there.
2 Frankly, we had-the option of either requesting to 3 cancel this meeting or to come here and give you a progress 4 report on the issues that we are right now grappling with.
5 And we get the benefit of your opinion on some of these 6 things, and as the meeting progresses, maybe we can 7 collectively make that judgment whether we are at this point 8 to go through the full ACRS meeting and you to write a 9 letter, or perhaps we have to come back again.
10 DR. REMICK: Okay. Fine.
11 MR. BAHADUR: So as I was mentioning, this 12 presentation is going to be more of a progress report, more 13 like making you aware of the questions, the issues that we
() 14 are grappling with, and see as to what the current thinking 15 is and where should we be going, in what direction.
16 As you point out, sir, the history of the rule, we 17 published the rule in 1984. It was a proposed rule that was 18 published for public comment. And it did generate an awful 19 lot of comment. There was a lot of public interest in the 20 proposed rule.
21 The major comment came from the industry, NUMARC 22 specifically, who suggested that we should drop the idea of 23 proposing a rule.
24 They suggested that they had developed a 25 guideline, that the industry has decided to accept Heritage Reporting Corporation s (202) 628-4888
em 10 b 1 voluntarily, that we should just issue a policy statement, 2 which will endorse those guidelines, and then drop the issue 3 right there.
4 Consequently, the staff developed a proposed 5 policy statement, attached the industry-developed guidelines 6 in that policy statement as an appendix, went for public 7 comments in 1988.
8 The comments again came from various sources:
9 psychologists, psychoanalysts made comments; the public in 10 general made comments; the industry was also very much 11 interested.
12 While we were developing these comments into some 13 resolutions, we were analyzing it, evaluating it,
() 14 Congressional committees held their hearings. The SINR 15 hearing came up in 1989. And at that time Hugh Thompson 16 went in front of the committee.
17 When he came back from the Congressional 18 committee, he came with an express understanding that 19 Congress wanted permission to make a decision sa soon as 20 possible and then get on with this idea of access 21 authorization.
22 An option paper was developed for the Commission 23 and the staff recommended that the Commission should go the 24 route of proposed rule.
25 The reason why we decided to recommend rule rather
(~)
's >
Heritage Reporting Corporation (202) 628-4888
ri 11 t
1 than policy statement. wuer first of all, the public comments 2 were overwhelmingly in favor of the rule.
3 Secondly, even the Congressional intentions 4 indicated that they were somewhat opposed to this perceived
.5 voluntary commitment on the part of industry for a program 6 like this.
7 The Congress wanted NRC to have a little more 8 authority in the matter.
9 From our point of view, the rule would be more 10 effective, especially if we are talking about a state where 11 the laws could be in contradiction to what we are trying to 12 do.
13 For example, in Michigan, you cannot use the data
() 14 developed through a criminal history for employment. And 15 there are certain small nuances here and there which will 16 make it a lot easier to enforce a rule than to depend on a 17 policy statement and keep hoping that the industry will 18 continue the guidelines.
19 The NRC of course will have the oversight on these 20 physical programs the same way as we do have for the fitness 21 for duty rule, for example, or the guard training. And then 22 of course a proposed rule would give us a well-defined 23 mechanism by which we can evaluate the programs of industry.
24 With this background, I will just briefly go 25 through the performance objective of the rule and go through Heritage Reporting Corporation O- (202) 628-4888 L __ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . . . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _
12
' (q
~
/
L 'l the basic attributes.
lL l 2 DR. REMICK: Mr. Bahadur, a nit that I observed in 3 reading the Federal.Recister notice, I think it is a typo, 4 on Page 2 of the Federal Reaister notice it indicates that 5 public comments were due in May of 1989. That presumably l 6 was May, 1988.
1 L 7 MR. BAHADUR: That is correct, sir.
8 DR. REMICK: Okay.
9 MR. BAHADUR: So in April of 1989, soon after the 10 SINR hearing and the option paper that was developed by the 11 staff, the Commission directed the staff to develop a final 12 rule, a rule with a clear performance objective.
13 And the objective was that the program should
() 14 provide a high assurance that any individual who is being 15 granted access to vital and protected areas should be both, 16 first of all, trustworthy and reliable and also there should 17 not be any unreasonable risk to public health and safety, 18 for any act, including radiological sabotage.
19 In order to do that, the rule that has been 20 developed, which is in front of the ACRS subcommittee today, 21 contains these basic elements.
22 MR. MICHELSON: Before we get into that, let me 23 just kind of ask you a question.
24 on your previous slide you talk about no 25 unreasonable risk.
Heritage Reporting Corporation O (202) 628-4888
l 1
1 That indicated to me that you had made some kind 2 of risk determinations on sabotage or are you just kind of 3 using that term differently than we normally think?
4 MR. BAHADUR: That is true. The term has been 5 used in a non-numerical reference.
6 MR. MICHELSON: You haven't done any kind of 7 probabilistic study to determine what the risk levels are, 8 might be, or anything else?
9 MR. BAHADUR: No. There is no risk analysis, 10 numerical.
11 MR. MICHELSON: I just wanted to be sure. Thank l
l 12 you.
13 MR. BAHADUR: The elements of the rule that is in
() 14 front of you is major attributes. It also has the 15 provisions for unescorted access in a manner other than just l
16 the normal, usual access.
l l 17 For example, it gives you the provisions for
.18 temporary access. It also gives the provisions for those 19 who already have access as of the day the rule will be 20 promulgated.
21 If reinstatement is done, then it goes the i
22 provisions for that, and also the transfer of sccess 23 authorization from one entity to the other, from a vendor to 24 a contractor, and various combinations.
25 MR. CARROLL: What does reinstatement mean in that l
Heritage Reporting Corporation (202) 628-4888 l
14 O 1 context?
2 MR. BAHADUR: For example, I had an authorization 3 in 1988, and I stopped working for a while. And when I came
~4 back and rejoined'a similar kind of work in some other 5 place, would I be reinstated automatically, or would I have 6 -to go'through the entire background investigation, --
7 MR. CARROLL: Okay.
8 MR. BAHADUR: -- psychological assessment, et 9 cetera.
10 DR. REMICK: But isn't it more specific? I don't 11 remember exactly. But am I correct that within 365 days or 12 something like that it can be reinstated under certain 13 conditions?
l O. 14 MR. - UR: That's rieht. ror the j
l 15 reinstatement, if you have the interruption of not more than l-16 364 days In other words, within a year, then you could be 17 reinstated without going through certain moves.
18 The rule also provides a screening for.the 19 situations where the plants may be closed during the cold 20 shutdown period.
21 There is a review process; the protection of 22 information; and of course records and audits.
23 DR. REMICK: Are you going to address things like 24 review process where you differ with the guidelines or are l 25 you going to take that up specifically, those areas where Heritage Reporting Corporation (202) 628-4888 L________._ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . . _ . _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
( the staff differs with the guidelines?
1 2 MR. BAHADUR: Yes.
3 DR. REMICK: All right. Fine.
4 MR. BAHADUR: Let's ta3k about the major 5 attributes, first.
6 The rule has, in its present format, the rule is a 7 general rule. It is accompanied by a regulatory guide, a 8 guide which endorses the guidelines developed by NUMARC.
9 There are three basic attributes: namely, 10 background investigation; psychological assessment; and the 11 behavioral observation program. They are defined in the 12 industry guidelines. Our regulatory guide does endorse that 13 and !n turn the rule lists these three attributes as being
() 14 the most important compulsory attributes to be followed for 15 the rule.
16 On the background investigations, the data such as 17 employment history, credit check, military history if any, 18 and this kind of information, the proper ID, this type of 19 information would be collected. And the industry would 20 follow the guidelines.
21 MR. MICHELSON: Excuse me. On the employment 22 history, in the guidelines, I don't see any mention of the 23 possibility that a person may take several months off, 24 terminate his employment acceptably, and take several months 25 off and go off to some foreign country and engage in an Heritage Reporting Corporation O (202) 628-4888
16 I'
l 1 extensive program to learn how to sabotage, and then come s
2 back and get re-employed, acceptably. And that seven and 3 eight month gap doesn't seem to be accounted for nor even 4 interested in.
5 MR. BAHADUR: Let's say that a person was employed 6- by a utility. And the scenario says that now he goes to a 7 country where he can learn how as an insider he can commit a 8 sabotage, in seven to eight months, and comes back.
9 MR. MICHELSON: Sure. Don't you have to account 10 for'all of yoor time somehow, whether you are employed or 11 not? Or do you? And do you have to account for any foreign 12 travel?
13 MR. BAHADUR: The reinstatement would require that
() 14 the interruption should not be more than a year.
15 And therefore, he may not have to --
16 MR. MICHELSON: Yes. Okay. This is less than a 17 year's interruption.
18 MR. BAHADUR: So he has gone, he has either gone 19 on a leave from where he was or he has resigned and then he 20 is coming back.
21 MR. MICHELSON: Yes.
22 MR. BAHADUR: The only check, and this would be a 23 behavior monitoring program. He comes back and he is being 24 monitored by the supervisor. And the supervisor has known 25- him before.
Heritage Reporting Corporation O- (202) 628-4888
17 1 MR. MICHELSON: No. It's a different company.
2 MR. BAHADUR: If it is a different company, then 3 he is going to review his employment history and thereby he 4 will have to be telling what he is doing the last seven or 5 eight months.
6 MR. MICHELSON: Well, yes. He can easily make up 7 a story of course about the last seven or eight months. And.
8 you would check those stories.
9 I didn't find in here that you even asked for 10 absences or checked on absences from employment.
11 MR. BAHADUR: Sandy, do you have something on 12 this?
13 DR. FRA? TALI: Yes, that is correct.
14 MR. BAHADUR: Because this question came up 15 earlier, also, and we have looked into it.
16 MR. MICHELSON: What is correct?
17 MR. BAHADUR: This is Dr. Sandy Frattali, also in 18 the Office of Research, in the Regulations Development 19 Branch.
20 DR. FRATTALI: That question was raised in public 21 comment.
22 I want to point out that the person we are talking 23 about has had a clearance program, is a person who at some 24 point has had a permanent, valid, access authorization. He 25 has gone through a security program.
Heritage Reporting Corporation O- (202) 628-4888
l
-s 18-U. l' MR. MICHELSON: That was before he went off to his 1
2 training camp.
3 DR. FRATTALI: Yes. But part of this is to.
L 4 understand that this is a person who we have determined 1
5 meets our standards.
6 We.had to come up, or the industry had to come up, 7 with a time, a break in service that you could reinstate.
8 This is an arbitrary time.
9 MR. MICHELSON: I'm not even going to reinstate.
10 I'm just gcing to go and start all over again.
11 DR. FRATTALI: Or start all over again. But this 12 is a person who has been determined as being trustworthy and 13 reliable previously.
() 14 We're saying that, within a year, and this is an 15 arbitrary amount of time, for a workable program, that 16 within a year we assume that this person will remain 17 trustworthy and reliable for a period of time.
18 This is simply, was a judgment by the industry.
19 We could see no real reason for making the time shorter. It 20- was something that has been in the guidance for a while and 21 we did not take exception to it.
22 MR. WARD: I think you are answering the wrong 23 question.
24 MR. MICHELSON: Yes.
25 MR. WARD: He's not talking about reinstatement.
Heritage Reporting Corporation O (202) 628-4888
1 He's saying, look at the employment history. The employee 2 worked for seven years for WalMart. Then he went off to --
3 MR. MICHELSON: He went on vacation.
4 MR. WARD: He went someplace for seven months, to 5 sabotage camp. And then he came back and worked for WalMart 6 for five years.
7 Do you have any way of determining, looking at 8 that gap in employment?
9 DR. FRATTALI: No. We are only assuming that a 10 person who has been in the situation of having had a proper 11 access authorization, that the reinstatement and the l
l 12 transfer conditions are good for a year.
13 MR. MICHELSON: Maybe, as David pointed out, maybe
( 14 he didn't even work under any kind of access authorization 15 before. He just came to work for you.
16 Do you ask him only about previous employment, or 17 do you ask him about previous vacations as well?
18 DR. FRATTALI: Oh, I'm sorry.
19 MR. MICHELSON: Particularly if they are out of 20 the country.
21 MR. BAHADUR: I think'I understand the question.
~
22 The question is whether we ara asking a person 2 !
. what he has done other than --
24 MR. MICHELSON: Does he fill in all the gaps?
25 MR. BAHADUR: -- educational history and l
I O
v Heritage Reporting Corporation (202) 628-4888
O 1 employment.
2 MR. MICHELSON: Yes.
3 MR. BAHADUR: And the answer is no.
4 MR. MICHELSON: You don't ask him what he's done?
5 MR. BAHADUR: What we can do is to add something 6 in the employment application: "Have you been to a sabotage 7 camp?"
8 MR. MICHELSON: No, no. I think you have to 9 account for all time periods for five or ten or however long 10 you think you need to look back.
11 But you need to account for all time periods.
12 MR. BAHADUR: Right. And I think it should also 13 include the traveling history, is that the suggestion?
O 14 Ma. M Castson: wae= 1 v cce==t th e -e -
15 you also check.
16 MR. WARD: I think you are asking if there is 17 something in the, not just employment history, but sort of 18 accounting for a person's time.
19 DR. FRATTALI: Yes.
20 MR. WARD: Equivalent to a "Q" clearance, say.
21 DR. FRATTALI: No, it is not equivalent to a "Q" 22 clearance.
23 MR. WARD: I mean, parallel with it.
24 DR. FRATTALI: Yes.
25 MR. WARD: In accounting for time.
Heritage Reporting Corporation O. (202) 628-4888
- ---_--_------------_-------_____j
(:) 1- DR. FRATTALI: There are two things that you are 2 looking at. Th; character and references, the afore 3 developed character and references must cover the five years 4 and there would be questions there as to the activities and 5 the character of that person.
6 The second thing that you look at is the criminal--
1-l 7 history, which does cover the entire period, for the check.
8 And the third thing that gives you some 9 information, some continuity information that you can get, 10 is the credit. check. There is information there that would 11 give someone doing this clearance program something to look 12 at.
13 So the period is covered, although not
() 14 specifically.
15 MR. MICHELSON: Well, let me ask it one more time, 16 and very specifically.
17 If a person is absent from employment for seven 18 months, do you inquire into where he was and do you check 19 into where he was?
20 DR. FRATTALI: That I would defer to someone who 21 does the security clearance. And I think the answer is yes.
22 MR. MICHELSON: Well, I hope it is in this paper 23 here, very clear.
24 DR. REMICK: Mr. Colvin would like to make a 25 comment.
(~' Heritage Reporting Corporation (202) 628-4888
l p 22
() _
1 MR. COLVIN: Yes, sir. Let me take a crack at 2 trying to clear this up.
1 3 If you look at the guidelines, and simply from an 4 employment history point of view, Paragraph 621 on Page 4 5 indicates that not only do you look at verification of 6 employment for periods of 30 days or less, but Item E 7 indicates that activities during interruptions of employment 8 in excess of 30 days must be verified.
9 So while I agree with Dr. Frattali, and that is 10 the other part of the answer, from just simply an employment 11 history point of view, we do verify periods in excess of 30 12 days, of non-employment.
13 So when you take that and consider that in company r
i 14 with the criminal history check, with the background, with 15 the other elements of the developed character references and 16 other items that Dr. Frattali indicated, you really do cover 17 those periods and try to come up with a high assurance that 18 that individual can be granted unescorted access.
19 MR. MICHELSON: The last time we had these 20 discussions a year or more ago, I asked the question about 21 to what extent you even account for or verify absence to 22 foreign countries, and the activities engaged in while in 23 absence in foreign countries.
24 And as I vaguely recollect, that was a very fuzzy 25 area as to what you were doing to do, if anything.
Heritage Reporting Corporation
(~}
%- (202) 628-4888 l
i-23 1 MR. WARD: I guess I still don't understand. The 2 industry guideline has covered the point Mr. Michelson has 3 raised, but the rule hasn't, apparently. Is that it?-
4 MR. BAHADUR: The rule --
5 DR. REMICK: The reg guide adopts the guidelines, l
l 6 with four exceptions.
7 MR. BAHADUR: Yes.
8 MR. MICHELSON: I'm just making sure that I l
9 understood "E." I was aware of "E" but I wasn't quite sure 10 what it meant, and I'm still not quite sure what 11 verification is going to involve where absences from the 12 country are involved.
13 MR. BAHADUR: I think that is an are;a in which we
() 14 need further investigation.
15 MR. MICHELSON: Well, it needs to be clear in the 16 guideline, or in the regulatory guide, as to what your 17 intentions are. It is not clear in the guideline, either, 18 as to what the intentions are, by the word verify, 19 particularly if they are absent from the country.
20 Maybe an industry representative would like to 21 expand on that.
22 MR. COLVIN: Perhaps during the discussion, we are 23 going to make a presentation and discuss the guideline, that 24 we might cover the points related specifically to the 25 guideline. I think it might be a more continuous process.
Heritage Reporting Corporation (202) 628-4888
24 l O. MR. MICHELSON: Good.
1 2 MR. COLVIN: And it perhaps will respond to your a l
3 question.
4 DR. REMICK: That would be precedent setting, if 5 one of our subcommittee meetings was very orderly.
6 Please continue, Mr. Bahadur.
7 MR. BAHADUR: Thank you.
8 The second major attribute is a psychological 9 assessment. And the way the industry guidelines have 10 described this, and also our regulatory guide embraces that',
11 is that first of all, the psychological assessments will be 12 done by a personality test, or any other acceptable method.
13 And if there is any problem highlighted by the test results,
() 14 then a clinical interview will follow.
15 The third major attribute would be the behavior 16 observation program. Something which will be done by the 17 supervisors. And if the supervisors observe any changes in 18 the behavior, any small infractions here and there, then the 19 matter will be referred to management and the management 20' will make the judgment.
21 MR. MICHELSON: Along the same line in the 22 behavior observation program, maybe the staff could tell me 23 what they would do in the case where they have an otherwise 24 fully acceptable employee but he happens to like to 25 subscribe to Soldier of Fortune magazine and go off to
, Heritage Reporting Corporation (202) 628-4888 I
25
,,c
\_)f 1 training camps once a year, and there are several of those 2 even in the U.S.
3 Is that unacceptable behavior?
d 4 MR. BAHADUR: That is more of a detail criterion.
5 MR. MICHELSON: I just want to get some feel for 6 what becomes unacceptable, where the threshold is.
7 MR. BAHADUR: If you are asking a personal 8- opinion, sir, then I think 3t is fairly acceptable. Simply 9 because somebody reads Soviet Land doesn't mean that he is 10 communist.
11 MR. MICHELSON: Well, let me expand on it 12 slightly. Maybe I'll take my two weeks off and go to Libya 13 for one of their training schools.
() 14 Would that be acceptable behavior? So there it a 15 line somewhere but it is outside the U.S.?
16 MR. MORRIS: I think we want to point out that 17 this rule and regulatory guide which would endorse the 18 NUMARC guidelines relies on a variety of elements that would 19 go into a decision by the management with regard to whether 20 access authorization should be, in this case, withdrawn, I 21 guess, for someone for whom it had been previously granted.
22 And I don't think it would be appropriate, 23 personally, to have one single element, such as what you are 24 raising, be the deciding factor.
25 I think the management of the utility would have Heritage Reporting Corporation O- (202) 628-4888 1
i l
i
- - _ _ _ _ - _ _ - _ _ __ _._w
26 q\ J l to make an overall assessment of that person's reliability 2 and trustworthiness and based on all the information they 3 have, if that information came to their attention, then they 4 would I believe act upon it appropriately, and it would be I 5 guess under the continued observation program.
6 MR. MICHELSON: That alone would still be 7 acceptable behavior, going off to soldier of fortune camps 8 once a year?
9 MR. MORRIS: I'm going to stand by my answer. I 10 think that overall assessment of all the factors would be 11 the way you would decide it. I would not say one element 12 would be unacceptable in my mind.
13 MR. MICHELSON: Okay.
in
() 14 DR. REMICK: When you say " acceptable," do you 15 mean from the staff's standpoint in auditing?
16 MR. MICHELSON: No.
17 DR. REMICK: Or from the utility's standpoint?
18 MR. MICHELSON: From the utility's standpoint.
19 DR. REMICK: There might be many answers to the 20 question.
21 MR. MICHELSON: Yes, I think there possibly would 22 be more than one answer to that one.
23 MR. BAHADUR: One thing I would like to point out 24 here is that this program is only a screening program. It 25 is like a mesh that we are slightly trying to tighten so Heritage Reporting Corporation f'}
k- (202) 628-4888 l
l _ _ _ _ _ _ _ _ _ _ _ -
J s. 27 1 .that more undesirable elements will be quite easily 2 differentiated from reliable and trustworthy people.
3 It is a very subjective way of course, a widely 4 accepted way. But nevertheless, if you just take only one 5 single component, it will be-difficult to make a judgment 6 just based on that single component.
7 And that is why this array of three major 8- attributes are essential for having a very effective 9 screening program.
10 MR. MICHELSON: It's a difficult problem, though.
11 Keep in mind, some people like to keep a lot of armaments in 12 their house, perfectly legal, semi-automatic weapons. And 13 they are a collector of a large number of these devices.
() 14 Does that kind of behavior raise any concern?
15 MR. BAHADUR: That is more of a hobby.
16 MR. MICHELSON: Well, they are all hobbies. Going 17 to the training camps is a hobby for some people.
18 MR. BAHALUR: I understand. But there are an 19 awful lot of things which a supervisor may not even be privy 20 to.
21 What I do in my personal life, the supervisor may 22 never know.
23 But once I take the personality test, once I 24 answer these series of 120 questions which goes into every
~
25 detail, or every facet, maybe that might show up somewhere.
(] Heritage Reporting Corporation v (202) 628-4888
l I
i 20
-)
x_/
1 Or maybe my day to day behavior. These are just some of the 1 2 precautions that could be placed in the prog::am and hope 3 that they will give the results that we are seeking.
4 Do those screening tests ever go into arms in the 5 home? That's not the screening you do at all?
6 MR. BAHADUR: There are questions in that, one of 7 the tests that I read, out of interest, had questions like 8 what do you like to do in your leisure time.
9 And then they give you three or four answers, and 10 then choose and pick the one --
11 (Extraneous noise) 12 MR. MICHELSON: That is our alarm system.
13 MR. BAHADUR: Have I said something?
() 14 MR. MICHELSON: That is our alarm system. There 15 has been a security violation.
16 MR. BAHADUR: All right.
17 So there are a series of questions which may look 18 into that aspect of the personality.
19 MR. MICHELSON: But do the questions get that 20 specific so that we know if the person has a particular 21 interest in --
22 MR. BAHADUR: I do not know, sir. It would be 23 difficult for me to make a judgment on that basis.
24 DR. REMICK: Let me ask a hypothetical question.
25 It might sound like it is off the wall. But it is not.
1 i
/~ Heritage Reporting Corporation k} (202) 628-4888
- - - - - _ - - - - - - - - - _ _ _a
i E
r 29-
{
1 There is a reason for.asking it. Dr. Frattali might be able 2 to respond.-
3 If there was a security clearance, let's say'a "Q" 4 clearance, would psychological assessment be needed?
5 DR. FRATTALI: There is no requirement for 6 psychological assessment in our current "Q" clearance.
7 program.
8 DR. REMICK: Okay.
9 DR. FRATTALI: But there is a more stringent 10 background investigation for the "Q" clearance program, 11 which would hopefully discovery such aberrations, any 12 aberrations, .if they exist.
13' ~DR. REMICK: Okay. That's what I expected to be.
( ).. 14 the answer.
15 MR. MARD: Your question was you wouldn't need the 16 psychological screening --
17 DR. REMICK: That-is right.
18 MR. WARD: -- if you had a "Q."
19 DR. FRATTALI: At this point in time.
20 MR. MORRIS: Do we understand your question to.be 21 if an employee of a utility had a "Q" clearance?
22 DR. REMICK: No. It'is actually DOE-related. But 23 just out of curiosity since we had some people here and we 24 talked about security clearance.
25 The question comes up if you have a "Q" clearance f Heritage Reporting Corporation (202) 628-4888
)
("i 30
\._) I 2
1 system, does it make sense also to have psychological 2 assessment.
3 MR. BAHADUR: This question may have different 4 implications.
5 For example, I have a "Q" clearance. I leave NRC 6 and I join a utility company.
7 Yes, I will have to go through the background 8 investigation, the psychological assessment and then the 9 behavior observation program.
10 On the other hand, if I continue to have my 11 employment with NRC, with a "Q" clearance, and I need an 12 access in some of the plants, I will be given the access 13 because I am certified by the NRC. That is the provision
() 14 there.
15 DR. REMICK: Yes.
16 MR. BAHADUR: So it depends, as to where you are 17 at that time, when you are applying for the access.
18 DR. REMICK: Well, I don't want to confuse the 19 question. It is not related to our discussion. But I 20 thought it was a good opportunity to ask some people who 21 might know.
22 MR. WARD: Did you get an answer?
23 DR. REMICK: I thought the answer was that if 24 there, if you have a "Q" clearance process you probably do 25 not need a psychological assessment.
f']
N Heritage Reporting Corporation (202) 628-4888
1 1 MR. WARD:. Oh.
2 DR. REMICK: That was my interpretation.
3 DR. FRATTALI: You do not have a psychological 4 assessment. Whether you need one or not --
5 DR. REMICK: Would it make sense --
6 MR. WARD: But I think he is asking your 7 professional opinion here. Let me ask the questions. He is 8 asking your professional opinion, if you had a "Q" 9 clearance-like process, would it then be necessary to have 10 the psychological screening?
11 MR. MORRIS: I understand this to be a 12 hypothetical question.
13 DR. REMICK: It is hypothetical, yes.
_ () 14 MR. MORRIS: We face the question in our thinking 15 in this way.
16 NRC employees are granted access, unauthorized 17 access, on occasion, to facilities. And we do not require a 18 psychological screening test for them. They have a "Q" 19 clearance.
20 The question comes to my mind or came to my mind, 21 well, is there something anything like that kind'of a 22 judgment made that would come out of a psychological 23 screening test and evaluation that would go over and beyond 24 what you would learn from a "Q" clearance, which would be a 25 background check, continuity check, talking to references, Heritage Reporting Corporation O (202) 628-4888
I fs 32
. .id 1 the same kind of thing that might be done in this program, j i
2 And in terms of those people who work for the~NRC 3' who are granted access, unauthorized access, unescorted 4 access, I'm sorry, we believe that most of those people have 5 been subjected to some kind of an evaluation by their 6 management, maybe not by professional psychologists, but 7 there is a continued association with individuals before 8 they are given that privilege by NRC to go and be able to go 9 into those areas unescorted.
10 We have a judgment that we make about their 11 reliability and trustworthiness that goes beyond what you 12 can get out of a simple "Q" clearance, I believe. And i 13 have talked about that.to several of the managers within the
() 14 agency. And I think they are agreeable that that is 15 something that happens.
16 Whether this extends to DOE people or not I don't 17 know. But the way we faced it with the NRC question was 18 that we believe there is a judgment made regarding the 19 trustworthiness and reliability of a person based on 20 observation of that person's characteristics before he gets 21 unescorted access.
22 I don't know whether that gets close to the 23 answer.
24 DR. REMICK: No, I understand. But it's not quite 25 the professional -- I was looking for just an opinion on if Heritage Reporting Corporation :
(202) 628-4888 i
O l' one thought professionally that a "Q" access clearance 2 process replaces the need for psychological assessment.
3 MR. MORRIS: I think maybe the answer would be, 4 maybe a better answer than~an opinion, although I will let 5 Sandy answer, is that we do not require it.
6 DR. REMICK: That I understand. That I 7 understand, yes.
8 DR. FRATTALI: The consensus of opinion in the
- 9. ' security community is that no, it is not necessary for a"Q" 10 clearance. But that is always subject to looking at it 11 again and re-evaluating it again. But at this point in time
- 12. that is the consensus of opinion in.the security community.
13 DR. REMICK: Thank you.
l I 14 MR. MICHELSON: The thing that bothers me a little l
15 bit about this business is that when you see in the 16 newspaper examples of people that have come in after years 17 of employment and bring an automatic weapon in end kill 18 several people, the fellow employees say gee, we didn't know 19 he was that kind, you know. Their psychological observation 20 of him through employment connections for years never gave 21 any suspicion.
22 Yet, when the reporters go and talk to the 23 neighbors, they knew all about this fellow and his 24 characteristics and so forth. He was a different person at 25 home than he was on the job.
Heritage Reporting Corporation O (202) 628-4888 l
L 34 l' And so a little bit of inquiring around the 2 neighborhood might be far more of a psychological 3 examination of the individual than observation even by the 4 fellow ~ employees, or even by a supervisor.
5 MR. MORRIS: But that kind of evaluation, dat a 6 gathering, is done before the access is granted.
7 MR. MICHELSON: For a "Q," but not under these 8 guidelines, I didn't any kind of inquiry at all.
9 MR. MORRIS: Not after the grant of the unescorted 10 access, but before it, yes, that kind of evaluation is done.
11- MR. MICHELSON: You check employment history and 12 education and criminal and military.
13 I didn't see any. Now, do you think that you will
() 14 actually go out and do an inquiry, then?
15 MR. MORRIS: I guess I would refer to Page 6 of 16 the guidelines. It seems to be saying, and I will make sure 17 that NUMARC agrees with this, but it seems to be saying that 18 you are going to examine references supplied by the 19 applicant, two additional references.
20 MR. MICHELSON: Okay. That covers my concern.
21 Yes. Thank you. I just didn't notice it on the'next page.
22 DR. REMICK: Do you want to continue, please?
23 MR. BAHADUR: Thank you, sir.
24 So the third major attribute should be the ;
i 25 monitoring of the changes in behavior of an employee, either ]
(] Heritage Reporting Corporation (202) 628-4888
)
)-
1 by supervisor, by supervisor mainly, coworkers, and then if
'2 there is any problem, then the problem could be referred to 3 management for an action or decision or judgment.
4 I would not go into the-detail of what else is 5 contained in the rule right now, but would like to place in l
L 6 front'of this subcommittee the issues that were raised by 1
7 CRGR, when we went in front of the committee about two weeks 8 back.
9 The first issue that they raised was that if the 10 industry is doing quite well in this area, if we haven't had 11 any sabotage in the last 20 years, they now come to us and 12 they say that they have developed a guideline, that they are 13 voluntarily embracing, and they promise they are going to
() 14 follow it, why do we need a rule? What is the benefit that 15 we are going to get out of this, and how much are we going 16 to spend? Will the money spent by the industry be worth the 17 assurance that we are going to get? Is it going to do 18 anything more than what the industry is doing right now?
19 DR. REMICK: Sounds like a reasonable question.
20 MR. BAHADUR: Yes. Somehow, when you work on a 21 rule which has been with us for the last ten years, such 22 questions don't come into mind. they are so reasonable that 23 they don't just come.
24 And two weeks later we went to CRGR and they asked 25 this question. And that made us look afresh at the need for Heritage Reporting Corporation O (202) 628-4888 1
3 36 O
1 this rule under the present circumstances.
2 And I would like to place in front of the 3 subcommittee the thinking that we have behind as to why do 4 we need this rule.
5 DR. REMICK: Fine. But it went out for public 6 comment as a policy statement, right?
l 7 MR. BAHADUR: It went to public comment in '84 as 8 a proposed rule.
9 DR. REMICK: Right. But then later.
10 MR. BAHADUR: Went for public comment in 1988 as a 11 policy statement.
12 DR. REMICK: Right.
13 MR. BAHADUR: And then the staff analyzed both k 14 pros and cons for both of these and went to the Commission 15 with the option paper, gave the Commission the option to 16 choose one or the other, or none. And the Commission 17 decided to go with the proposed rule.
18 DR. REMICK: The reason I asked that question, you 19 made it sound as if this was a new thought to the staff.
20 But it wasn't, because you went out for public comment on 21 the policy statement.
22 MR. BAHADUR: The only thing is, ten years is a 23 long period.
24 The thing we thought ten years back, intuitively, 25 later might have some questions. I mean, it is just a good
/~ Heritage Reporting Corporation kTl (202) 628-4888
t.
37 1~ time to turn around, examine that question.
2 MR. MORRIS: Let me add to this general thought 3 that also at the CRGR meeting the idea came up that a number 4 of utilities already have commitments approximating some 5 fraction or all in some cases maybe of the NUMARC 6 guidelines..
7 And if that is the case, and they are already.in 8 their. license conditions, .that perhaps a rule would not be 9 necessary if those are already essentially da facto 10- -requirements.
11 And so what we were not able to do at that 12 meeting, and I think not based on previous surveys that had' 13 been taken, would be to have a good snapshot of just where I) 14 utilities were with regard to incorporating all these kinds 15 of provisions into license commitments that they had made 16 formally.
17 Since then we have done some more survey, which 18 Mr. Bahadur is going to tell you about, to get a better 10 picture of that.
20 But that was behind this question, is why do you 21 need it if these commitments are already made.
22f And now he can give you the answer to sort of how 23 we have analyzed that issue.
24 MR. BARADUR: As Bill mentioned, the question was 25 what the industry is doing right now, what are the Heritage Reporting Corporation O,' (202) 628-4888 l
1 commitments and how they are going to continue to do so.
2 And at the time we went to CRGR, I did not have 3 that detailed picture in my mind.
]j 4 The second issue that they raised was that if the
- 5. regulatory guide is endorsing the industry-developed
'6 guidelines, why aren't we doing a clean job of endorsing the 7 entire guideline? Why are we coming up with these four i 8 major exceptions, the exceptions which are entered'in your L
9 rule?
10 And the last one was, the provisions of cold 11 shutdown, the provision by which you screen the people 12 having access in the power plant during the cold shutdown 13 period, why it seems to be more liberal than the existing
() 14 practices in the industry.
15 Of course, I couldn't give the answer to-this 16 question at that time either, because I did not have much 17 idea about what the industry was doing.
18 And as I presented to you today, maybe I don't 19 have it right now, either. And maybe I will ask NUMARC to 20 give us a little more enlightenment in this area towards the 21 later part of the presentation.
22 Let me turn to a slide which is somewhat busy, but 23 it gives us a basis of what is happening, where we are and 24 why we think we need the rule.
25 We took a survey of 20 utilities. And when I say l Heritage Reporting Corporation Og (202) 628-4888
<- 39
%j 1 we did a survey, that means we re-examined their physical 2 security plan.
3 We just selected these 20 utilities randomly and 4 then saw what they had committed to.
5 And this is what we found.
6 We found that right now their physical security 7 plans require that they meet the standards, ANSI 18.17, for 8 the access authorization.
9 And we found that while five utilities were at 10 least meeting those three requirements given in these 11 standards, the two did not meet all these standards.
12 They had the background investigations. They also 13 had the monitored behavior program. But they did not go
() 14 through the psychological assessment.
15 The survey also found that at least ten utilities 16 went beyond the 18.17 but were meeting the revised ANS 3.3 1
17 where these three components: the background investigation, 18 psychological assessment and the behavior monitoring, were 19 prescribed a little more in detail.
20 For example, they have listed out that you check l
21 your employment, education, criminal records and references, 22 and then it also gave something about the assessment.
23 At least ten utilities were meeting those 24 standards.
25 Three utilities out of 20 went beyond these Heritage Reporting Corporation O' (202) 628-4888
C:)
I standards, only in their commitment.- In practice, maybe 2 many more are doing it. We have no knowledge of that. But 3 in their physical security plan, at least three indicated 4 that they were doing a little beyond the 18.17 to the tune 5 that they had better education information, their military 6 history was there, they were also going to the credit check.
7 That is the picture of the industry today, on 8 paper, commitment. There are at least 1, industries, 15 9 utilities which are meeting the standards 3.3, two are below 10 and three are above.
11 DR. REMICK: Is the inference there that ANSI, or 12 ANS 3.3 is not as datailed as the guidelines? Is that the 13 proper inference I should make from that statement?
() 14 MR. BAHADUR: The ANS 3.3 is not as much in detail 15 in the background investigations, does not have the 16 components listed out as much as the guideline does.
17 It also provides for a personality test and if 18 something goes wrong, then a clinical interview.
19 And the personality test could also be replaced by 20 technically acceptable other ways of finding the similar 21 characteristics.
22 MR. MORRIS: Maybe you were going to make this 23 point. But I want to make it clear that we are not 24 asserting that many utilities may not be doing somewhat more 25 than these commitments on paper.
' ( Heritage Reporting Corporation (202) 628-4888
)
1 We don't know just where they stand in that
, 2 regard. Maybe NUMARC can shed some more light on that.
3 But those are over and above anything that we 4 could document that.would be something that would be 5 equivalent to the requirements of the rule.
6 MR. MICHELSON: You screen all the reports that l
7 come in from the utilities on minor infractions.of security 8 and so forth.
9 Is that correct? Do you automate that to some 10 extent? Or you just kind of read it?
11 MR. BUSH: There is an automated system. I'm 12 Loren Bush, NRR. We do have a system where the question of 13 program findings are summarized.
() 14 MR. MICHELSON: I wasn't thinking of just the 15 inspection programs, although those are also important. I 16 was thinking of the one the utility has to submit when they 17 find they are in violation of one of the requirements.
18 I forgot which part of the regulations require a 19 report be submitted. It is submitted separately from LERs 20 because it is considered to be of a more confidential 21 nature.
22 MR. BUSH: I believe you are referring to 23 reporting requirements --
24 MR. MICHELSON: That is probably a good 25 description, yes.
f Heritage Reporting Corporation (202) 628-4888
g) 42 I k
Ll 1 Do you automate those reports as they come in?
2 MR. BUSH: Yes. NMSS has the responsibility --
3 MR. MICHELSON: Are you acquainted with the 4 results at all?
5 MR. BUSH: Say again? ,
l }
6 MR. MICHELSON: Are you acquainted with what j 7 results you have seen?
8 NR. BUSH: Not offhand. But I can speak generally 9 that we are getting a number of reports of agencies, and the 10 log entries.
11 MR. MICHELSON: Let me ask the question a little 12 differently.
13 If a utility finds that they have experienced some r's
(_) 14 difficulty, maybe the fellow was discovered with drugs 15 onsite, and then they check into it a little bit, if they 16 find that he, they look back to see well, why did we clear l
l 17 this guy to begin with, for accese, if they find some l
18 deficiencies in his character that should have showed up, do 19 they have to report that to the NRC7 20 MR. BUSH: No.
21 MR. MICHELSON: They don't. So you aren't really 22 aware of how good the screening process is working 23 presently, unless they are required to report such 24 deficiencies in their screening process?
25 MR. BUSH: If they did report anything it would be Heritage Reporting Corporation
(~)
(202) 628-4888
43 0 1 voluntarily. Or the other alternative would be for the 2 inspoctor looking into their program might come to us.
3 MR. MICHELSON: Because I has leard a few horror 4 stories in which when they do discover the fellow with drugs 5 onsite or alcohol abuse problems onsite, it turns out that 6 he wasn't very well screened to begin with, and it should 7 have been picked up long before.
8 But I don't know if I just hear the isolated 9 instances or whether this is a little more prevalent than we 10 :e . ght suspect.
11 MR. BUSH: This whole area has been fraught with 12 problems over the years, particularly on the part of L3 contrnetors, with background information --
lll 14 MR. MICHELSON: What concerns me is how well is 15 the industry really doing today when you say the; are going 16 by this or that guideline or ANS standard, how good is it 17 really working as opposed to the fact that they have adopted 18 the standard? Do we know?
19 MR. BUSH: That's difficult, because if people are 20 going to commit fraud in order to get people employed, then 21 they are going to do it no matter what kind of rule we have.
22 You' re talking about shat:ing the trees.
23 MR. MICHELSON: I just wondered to what extent we 24 are even !nformed of the problem and to what extent we have 25 a kind of a collective view of how good or how bad the Heritage Reporting Corporation G (202) 628-4888
44 1 situation is.
2 I guess the answer is really you don't really have 3 a systematic.way of collecting such information or if you do 4 at least you are not aware of a summary result.
5 MR. BUSH: No.
6 MR. MICHELSON: Thank you.
7 MR. MORRIS:. May I ask, are you asking about-8 detailed information about possible problems individuals may 9 have in'their backgrounds or are you talking about actual 10- security or safeguards, violations or problems that might 11 have occurred?
12 MR. MICHELSON: What I am talking about is where 13 there is a violation and they check into the individual,
() 14 they.found that they wondered why they should have cleared 15- him to begin with. And it is that part of the aspect. It 16 is kind of a root cause analysis of the violation.
'17 NR '. MORR.IS: Yes. Precisely.
18 MR. MICHELSON: And that tells me roughly how good 19 all these present programs are working and therefore how
- 20. much change if any is needed.
21 MR. BUSH: I'm sorry. I was being force-fed 22 information.
23 (Laughter)
, 24 MR. BUSH: I was reminded that there is a 1
25 provision in the reporting requirements that if the licensee O Heritage Reporting Corporation (202) 628-4888 1
45
.fs '
t
-1 realizes or recognizes that certain elements of for example 2 the background information were not conducted properly, then 3 we do expect the licensees to report that to the NRC.
4 MR. MICHELSON: Do you compile those reports?
5 Have you; looked at the results of that compilation in some 6 systematic way that would give you a feel for how good the 7 present programs are working, and therefore whether or not 8 they would even need to be fixed?
9 MR. BUSH: What we did was we retrieved data from 10 the inspection histories and identified the violations that 11 have been identified over the last ten years. And I think 12 it was six licensees advised of contractor violations.
13 MR. MICHELSON: But that is only a small sample of O 14 the tota 1 numder.
15 MR. BUSH: Yes.
16 MR. MICHELSON: That just happens to be those that 17 are picked up out of inspection reports.
18 MR. BUSH: That is correct.
19 MR. MICHELSON: And I am interested in a larger 20 spectrum of information, because that's too small a sample 21 to be useful.
22 MR. BUSH: Well, to be honest with you, we also 23 have a problem with licensees not reporting things that we 24 would expect to see.
25 MR. MICHELSON: So at least from that viewpoint, O Heritage Reporting Corporation (202) 628-4888
46 1 it.is nard to say whether the present programs are working 2 or not, at least on the basis of that information.
3 Thank you.
4 MR. MORRIS: I would just mention that the 5 industry guidelines and this package of rule and reg guide 6 would~ require annual audits by the utility and I would think 7 that those -- NUMARC may be able to go into this more --
8 would include something like an analysis of the type you are 9 talking about. If they note that there are problems, they 10 would periodically try to root those out and figure out what 11 their implications are.
12 So it may be going on voluntarily by a number of
-13 people in the industry to try to. understand this, but as
() 14 yet, without the rule and reg guide, it'would not be part~of 15 our regulations, other than what you have just heard about 16 here.
17 MR. MICHELSON: Maybe NUMARC, when they make that 18 presentation, could tell me what they know in this regard.
19 Thank you.
20 DR. REMICK: Do you want to continue, please?
21 MR. BAHADUR: What the staff did after noticing 22 that at least ten utilities are meeting the ANS 3.3 is did 23 some cost estimation, just to have an idea as to where they 24 are and how much are they spending, and how much they may 25 have to spend should they go to the guideline or the Heritage Reporting Corporation O (202) 628-4888
-____m_..._ _ _ _ _ _
47
-f-(
x) 1 proposed rule.
2 MR. MICHELSON: Just for clarification, how many 3 utilities are we-talking abriut? About 52, something like 4 that?
5 MR. BAHADUR: Total?
6 MR. MICHELSON: Yes.
7 MR. BAHADUR: What is the total? 54.
81 MR. MICHELSON: We are talking about a small 9 fraction of the total even in this chart.
10 MR. BUSH: This is sites, not utilities.
11 MR. MICHELSON: Oh, those are sites.aS 12 MR. BAHADUR: Sites.
13 MR. MICHELSON: Okay. Then how many sites are 14 .there in total?
15 DR. FRATTALI: This survey is 20 sites.
16 MR. BAHADUR: How many sites are there in total?
17 ER. FRATTALI: 111.
18 MR. MICHELSON: That's reactors. 70-something?
19 Okay. That's all I was trying to find out.
20 MR. BAHADUR: It's about 25 percent of the total, 21 which are about between 75 to 80 sites.
22 MR. MICHELSON: Thank you.
23 MR. BAHADUR: And before I ask Dr. Cid Feld to 24 give you the assumption that he made in making these cost 25 estimates in case there is a question, let me just summarize
() Heritage Reporting Corporation (202) 628-4888
)
I im 48
(_)
1 as to what we did. l 2 We assumed that there is a typical site out there, 3 which is at this level of adequate protection. And the cost 4 of being at this level was estimated to be $12 million.
5 They are spending $12 million as they are analyzed over a 6 30-year period, per site.
7 Should they decide to embrace the guidelines as 8 three of the sites have already done so, and to practice it, 9 then the additional cost would be $60,000, again, over a 10 period of 30 years.
11 If the rule goes the way it is worded, and if the 12 four exemptions that the regulatory guide takes to the 13 industry guidelines, then the additional cost to the
() 14 industry would be an additional $250,000.
15 MR. MICHELSON: The savings for transferability 16 tells me I guess that the present guidelines do not require, 17 or do not have a provision for transferability?
18 MR. BAHADUR: The present standard --
19 MR. MICHELSON: Yes. No. I was reading the 20 guidelines.
21 MR. BAHADUR: -- would not have the provisions for 22 transferability because the program is not uniform all over 23 the industry.
24 Should the guidelines be accepted, should the 25 program become uniform, and I see some headshaking in the Heritage Reporting Corporation C) (202) 628-4888
,-)
, 49 v
1 NIRCMRC corner. Before I ask them to say anything and bust 2 my balloon, let me just point out a couple of things here.
3 Right now the transfer provisions are not there.
4 There are vendors out there, contractors out there -- the 5 Bechtels and the GEs -- who may be running more than 15 6 access authorization programs within their shop, so that 7 they can meet the need of the individual utility programs.
8 Once a uniform program is laid out over the 9 industry, that cost will be reduced.
10 It is the indirect benefit to the industry. It is 11 nothing doing to the regulator, to NRC. But that is a cost 12 which is going to be in the negative column if you compare 13 them as to what they are spending today.
() 14 DR. REMICK: Do you want to respond to that?
15 MR. COLVIN: Yes. Perhaps it would be worth 16 taking a moment on that, Dr. Remick.
17 We have a provision in the guidelines for 18 transferability of access authorization.
19 I think that in order to utilize that provision, 20 we ought to recognize that no matter whether we put in place 21 the guidelines under a policy statement or whether we have 22 an NRC rule, I don't believe that we are going to achieve a 23 standardization or uniformity throughout the industry as far 24 as access authorization programs go.
25 We are going to certainly achieve some minimum
- Heritage Reporting Corporation (202) 628-4888
._ ._ _______-___m
r~g 50 V. 1 level of standardization, but for the most part, .many 2 licensees have programs-in effect that are now part of their 3 current license, and I don't believe the intent is to-have 4- them back off from some of those current requirements to put 5 in place, you know, to meet the requirements of either the 6- guideline or the reg guide or whatever the appropriate 7 guidance document is.
8 So I think it is incorrect to assume that we are 9 going to have all standard programs in each one of the 54 10 utilities.
11 To accomplish the provision of transferability and 12 to ensure that it is viable, though, in our guidelines, 13' we've put in a requirement to say that in order to transfer-(jf 14 . access authorization or unescorted access authorization from 15 one utility to another, you can do that even though you 16 don't meet all the requirements of an individual program, as 17 long as you meet all the requirements of both the access 18 authorization rule provided by the Commission and, and this 19 is an "and" function, the fitness for duty rule that is 20 established by the Commission.
21 So if you meet the minimum requirements for your 22 panel of drug screening that are in the fitness for duty 23 . rule even though you might test for additional drugs, you
- 24. could accept transferability of an individual without the 25 additional drug testing as long as you met the requirements Heritage Reporting Corporation (202) 628-4888
~
l: 1 of both the rule and access' authorization and fitness'for 2 duty.
3 So we did try to account.for transferability in
.4- that sense.
5 DR. REMICK: Thank you.
6 MR. BAHADUR: I think that is what I said before.
y 7 If the industry embraces the guidelines, whether through.the 8 proposed rule or a policy statement, there.is a benefit, a 9 perceived benefit, because of the uniformity of some sort.
10 Right now the Standard 18.17 says you go out and 11 do these three things, and do it the way you want.
12 What are the chances that all the 54 utilities 13 will do it only one way? I think very small.
I) 14 On the other hand, if you go out and say these are 15 the three performance objectives that you have, here are the 16 guidelines -- which is one acceptable way of doing it -- the 1 ~7 chances are the majority.of the utilities will do that 18 rather than reinventing the wheel.
19 I mean, it is a judgment call one way or the LO other. Whether we go the proposed rule route or the policy 21 statement route is a different issue.
22 Once the guide es are in place and industry is 23 committed to those guideline, either voluntarily or through 24 a rule, there is a perceived advantage.
25 And that is that the transferability will save a Heritage Reporting Corporation O (202) 628-4888
1
(~\ 52
(/.
l' certain amount of money. And that is what I was trying to 2 put out here.
3 MR. MICHELSON: When we talk about 4 transferability, do we mean that the paperwork that the 5 previous employer might have performed during the tenure of 6 the employee at that location would be transferred or just a 7 note that says he is a good guy?
8 MR. BAHADUR: The provisions for transfer would 9 say that you can accept an individual transfer from a 10 contractor, vendor or another licensed facility, provided 11 number one, his employment has not been interrupted for more 12 than 365 days and that you have verified the interrupted 13 time in writing.
$f 14 You also have a true identity and you have written 15 confirmation of his authorization by the previous employer.
16 MR. MICHELSON: It looks like if you really want 17 to save some money, there ought to be a provision that there 18 is a dossier that goes with the individual from location to 19 location so they don't do all their homework over again each 20 time.
21 Otherwien, I don't know that transferability saves 22 much unless you just -- I guess maybe there is jt; tat no, I 23 was thinking of the reinstating like a "Q" clearance. Here 24 I guess you don't even reinstate, you just transfer and you 25 accept the previous employer's statement that he is good,
(} Heritage Reporting Corporation (202) 628-4888
L L 53 l'
1 and not know the basis, because each utility you said has a 2 different kind of program and they are not necessarily 3 ' standard.
4 MR. BAHADUR: Right now.
l 5 MR. MICHELSON: Right now. And under the new
- 6. provision, you have a certification he met the minimum 7 standard, namely, the regulatory guideline. You would !
8 certainly, you can get a statement, and then you would know 9 roughly-what kind of checks had been done because he met the 10 requirements of the regulatory guide.
11 MR. BAHADUR: So the question is as to why do.we 12 require that in the rule?
13 MR. MICHELSON: No. I'm trying to determine what
() 14 transferability really means, in case of a utility that has 15 more stringent requirements coming, getting an employee from 16 a utility that only met minimum requirements.
17 MR. BAHADUR: If the rule was out there, the rule 18 with the regulatory guide and the guidelines and based by 19 the guide, and if a person has gone through that program, 20 that is the NRC-approved program --
21 MR. MICHELSON: And that's what you testified to.
22 MR. BAHADUR: -- now, Mr. A works in one utility, 23 has gone through that program, has the authorization . He 24 desires to go to a different program. Within 365 days he 25 would do so without going through the entire background Heritage Reporting Corporation O. (202) 628-4888
1 54
.,/
V information check --
1 i 2 MR. MICHELSON: Even though the new utility had l
L 3 more stringent requirements that the guide?
4 MR. BARADUR: That would be up to the utility to j 5 go beyond the call of duty. But at least he.has gone 6 through the NRC-approved program.
7 MR. MICHELSON: Okay.
8 MR. BAHADUR: And the utility would not be at l
9 fault to accept that.
10 MR. MICHELSON: There is no provision to ever 11 recheck.these things, I guess. Is that right?
12 Is there a requirement at the end of five years 13 you must do another background check, for instance?
l( ) 14 MR. BAHADUR: Sandy, what is the current opinion 15 on that?
16 DR. FRATTALI: The current rule does not contain a 17 reinvestigation procedure.
18 However, we are considering requiring 19 reinvestigation for it just as we have recently required 20 reinvestigation for our own "L" clearances.
21 There has been an initiation of rulemaking for 22 that.
23 MR. MICHELSON: That will be a separate rule from 24 this?
25 DR. FRATTALI: Yes, it will. It is a new issue Heritage Reporting Corporation
('d')
s (202) 628-4888
l 1 that has to go out'for public comment. -It has not been out 2- for public comment.
'3 MR. MICHELSON: So right now the intention is, 4' 'though, it goes on in perpetuity without rechecking. Thank 5 you.
6 .DR. REMICK: In fact, I think somewhere in the 7 documents you sent, that.was an internal decision, to 8 separate that.out from this so that this is not held up, the
.9 access authorization is not held up, while that goes out for 10 public comment.
11 Is that correct?
12 DR. FRATTALI: Yes, sir.
13 DR. REMICK: Okay.
14 .MR. BAHADUR: So the objective of showing this 15 . exercise was to indicate that a typical utility is spending 16 $12 million to reach the level of adequate protection, and 17 would be expending another 2.5 to 3 percent of that money 18 should we decide to go the enhanced assurance for increased 19 safety.
20 DR. REMICK: I can't read that note completely on 21 the right there. It's not printed. What does that note say 22 about the $12 million? Current per reactor?
23 MR. BAHADUR: Cid Feld, could you please give your 24 basis for these numbers?
25 DR. REMICK: What does it say, first?
Heritage Reporting Corporation O (202) 628-4888
56 1 DR. FELD: That would be the cost per reactor over 2 the' remaining 30 years.
3 DR. REMICK: Okay. Not an annual-cost?d
'4 DR. FELD: That is correct.
5 MR. MICHELSON: That is not a site, either, it is 6- per unit.
7 DR. FELD: It is per reactor.
8 DR. REMICK: Okay. That's 5. percent discount 9 rate. Okay.
10 Do you want to proceed.
11 MR. BAHADUR: So the question in front of us, 12 before we go to CRGR, would be is the 3 percent additional 13 cost worth the increased assurance that we are going to get?
() 14 In addition to these, I have listed out six 15' reasons why does the staff at this time consider that we 16 need a rule.
17 The 51rst being, when we did the survey we found 18 at least 10 percent of the sites were not meeting even the 19 minimum standards. -By promulgating this rule, we will 20 ensure that those licensees would also be committed to the 21 basic standard to improve their program.
22 It would provide a uniform standard for NRC to 23 evaluate all other programs. Right now, our inspectors do 24 not have a standardized program to compare with. The only 25 thing we have to go by is either the 18.17 standards or the Heritage Reporting Corporation f'} (202) 628-4888
57
.O 1 3.3.
2 It would provide a well-defined mechanism for 3 effective enforcement. And should we find deficiencies in 4 the program, we would be able to very clearly indicate where 5 the programs are deficient. It will be easier from that 6 point of view, rather than looking at 50 different programs 7 developed by 50 different people.
8 At this time we know that at least three sites are 9 committed to the guidelines and I have been told that in 10 practice there are many more sites which are practicing the 11 methods, the procedures, the ways which are suggested in the 12 guidelines. But to ensure that they stay at that level, 13 throughout the life of the plant, this rule would make that
. 14 clear.
15 DR. REMICK: That's not quite clear to me. If 16 they make a voluntary commitment, the staff usually comes 17 along with a confirmatory order or something equivalent, and 18 it becomes a formal commitment - you referred to that 19 earlier -- how could they possibly drop it, then?
20 MR. MORRIS: It seems that the mechanisms would be 21 the alternatives, if you were going along those routes, 22 would be to actually do that, go about the process of 23 rewriting all of the licenses to upgrade to some level, say 24 the guidelines.
25 And if that were done and had been done at this O Heritage Reporting Corporation (202) 628-4888
58 1 time, I think some of the reasoning for the rule, this is 2 the issue that came up before, - would be less positive.
3 If we put the rule and the' regulatory guide out, 4 and let that stand as the standard, then I don't think you 5 would have to go through that process of one by one 6 determining what status the licensees had attained on a 7 voluntary basis. This would be a way to just say put it on 8 the books and say here is what the standard of the agency is 9 and expect it to be met.
10 DR. REMICK: Okay.
11 MR. BAHADUR: The rule provides a proper guidance 12 for the plants which will be licensed in future.
13 And of course there will be an industry-wide
() 14 standard for access authorization program, which I mentioned 15 earlier would provide transferability that would be a 16 considerable saving to the industry.
17 DR. REMICK: I can't help but make the comment 18 there is nothing on there that talks about assuring safety.
19 It is all inspection and enforcement.
20 MR. BAHADUR: The assurance of safety is implicit 21 in the entire program, whether we go through the proposed 22 rule route or we go through the policy statement route.
23 DR. REMICK: So it goes unsaid in other words?
24 MR. BAHADUR: In both cases, the mere action has 25 been taken, with that goal in mind.
) Heritage Reporting Corporation (202) 628-4888
59 i k-)
L 1 Now, we are trying to find out the ways-by which 2- we can achieve that.
l H 3 As NUMARC suggested, we could achieve that by just 4 letting the industry voluntarily accept the guidelines the 5 way they are and just leave it there.
6 The other possibility is to go to policy statement 7 endorsing the guideline. And the third possibility is go to 8 the proposed rule, prepare regulatory guide which endorses 9 th6 guideline.
10 I would move on to the second major concern that 11 the CRGR had raised, and that was the four exemptions that 12 we had taken to the guidelines developed by NUMARC.
13 The.first exception that we took was the military
() 14 history. The way I understood, the guideline said the 15 history would be limited to five years only. If a person 16 has claimed that he was in the military six years back, for 17 example, the background would not go into the military 18 history.
19 The staff feels that the military history should 20 be total.
21 One of the reasons why we feel that way is because 22 for a civilian,- if during his lifetime he has committed a 23 crime, it goes in the FBI files. So you go through the 24 criminal check and you.will find everything crime-related 25 that he has done would be in his file.
Heritage Reporting Corporation O (202) 628-4888
es 60 s-) 1 On the other hand, if you were in the military,
]
1 2 and if you commit a crime, it would not go into the FBI )
3 files.
4 So to bring the military people at par with the j l
i l 5 civilian, the staff felt it necessary that the military 6 history should be checked in total.
7 MR. MICHELSON: Does that first bullet mean that 8 if you served more than five years ago, you don't check the 9 military history at all?
l 10' MR. BAHADUR: That's the way the guideline --
11 MR. MICHELSON: That's the way the guideline says 12 it.
13 MR. BAHADUR: The guideline says you are applying
() 14 for authorization in 1989. You were discharged out of the 15 military in 1982. We don't worry about that history.
16 MR. MICHELSON: But they ask whether it was 17 honorable or not.
18 MR. BAHADUR: But that, I'm told, is a very 19 general document.
20 MR. MICHELSON: Oh, yes.
21 So that is the first exception.
22 The thinking the staff developed also indicated 23 that whether you go for the five year total, the cost --
24 well, I replace it.
25 We feel if you are longer than five years out of Heritage Reporting Corporation O (202) 628-4888
61-1 the military, then of course, there is no cost on the 2 military history.
3 But if you ask for the military history, it just 4 doesn't stop at five years. You get the total history. So 5 there is- no additional cost in case you were tx) make the 6- check.
7 I'm told that there is an informal agreement now 8 between the military people and FBI, that they will be 9 sending the criminal records to FBI, but that is only an 10 informal basis. Perhaps an MOU is being developed right
, 11 now. And that will only be effective starting from last 12 year, that is, 1988.
13 So anyone who has been in the military before that
() 14 would still be an unknown quantity to us.
15 The second exception that we took was the-l 16 guideline asks for a reliable personality test. The 17 guideline said if the personality. test is administered, make 18 sure it is reliable.
19 The staff's thinking was the test should be not 20 only reliable but should also be valid.
L 21 MR. MICHELSON: Could you tell me how you can have 1
22 a reliable test that is not valid? What would be an 23 example?
24 MR. BAHADUR: I have a very good example. And 25 after Morris has listened to it about eight times, he said Heritage Reporting Corporation O (202) 628-4888 1
.________---___-_____-__a
i n 62 j U 1 if I tell it the ninth time he would walk out. So with that
[.
2 risk, I am going to tell it to you.
3 Let's say that you want to be comfortable in this 4 -room. One of the factors of the comfort could be 5 temperature. Temperature would be a valid parameter for 6 your comfort in this room. But you would need a reliable 7 thermometer to measure that.
8 Simply because the thermometer is reliable would 9 not mean that the temperature is a valid parameter for 10 comfort.
11 First, you have to determine what parameter would 12 be valid to your situation.
13 MR. MICHELSON: Said that way, I understand it,
() 14 because I understand temperature and thermometers. But I 15 don't understand psychology so well.
16 MR. BAHADUR: Well, let's try it both together.
17 Because I have the problem with psychology myself.
18 We are trying to find certain characteristics in
~
19 the atmosphere of the nuclear industry. We are trying to I
20 assure against the threat of a radiological sabotage.
21 Let's say one parameter is color. If the person 22 doesn't like a pink color, he is unreliable and 23 untrustworthy. But that certainly is not a valid parameter.
24 And we can easily discard that.
25 On the other hand, maybe you want to find out if Heritage Reporting Corporation O- (202) 628-4888
,- . 63 C' 1 this man has a fear of closed space, if this man likes to 2 use guns. Then there is a difference of opinion.
3 And once we say those are the characteristics we 4 are going to find out, then we will find a test which will 5 give us reliable results.
6 MR. MICHELSON: Okay.
7 MR. BAHADUR: Because the test is reliable does 8 not'mean it will give you the valid parameters. That is the 9 thinking of the staff. We took this exception. By 10 differentiating it, we may clarify this language in the 11 regulatory guide. We don't have to make an exception so 12 blatant.
13 DR. REMICK: I assume that MHPI has been found to
() 14 be both valid an reliable for this purpose.
15 MR. BAHADUR: Sandy, would you comment on that?
16 DR. FRATTALI: It is a valid and reliable test for 17 characteristics. Now, it is not a valid and reliable test 18 for the nuclear industry. That is why we can't ask for a 19 valid test for the nuclear industry. It has not been 20 validated for the nuclear industry.
21 But it will be valid for characteristics that we 22 want to address that relate to trustworthiness, reliability 23 and the threat of radiological sabotage.
24 And if the industry wants to regard that as a 25 reliable test, that is fine.
["N Heritage Reporting Corporation (202) 628-4888 i
.y; 64 1 But'it can't be valid for the color pink, because 2 that is not a characteristic that we care about.
3 DR. REMICK: Has there been any psychological 4 screening test that has been validated for the nuclear 5 industry?
6 MR. BAHADUR: The validation per se has not been 7 given to any test specifically for nuclear industry.
8 However, there are tests which are perceived. valid 9 based on the characteristics that could be applied to the 10 nuclear atmosphere.
11 DR. REMICK: Repeat that last sentence again.
12 MR. BAHADUR: I was afraid you would ask me that.
13 First of all, we may have to make a list of the
() 14 characteristics which are applicable in the nuclear 15 environment. And then we can go to the tests and we will 16 find that there are certain tests which are valid to measure 17 those characteristics or to reflect those characteristics in l 1
i 18 the personality.
19 DR. REMICK: I guess I am not seeing through this.
20 It is my understanding from what I have read here that i 21 essentially all the nuclear industry is using MMPI.
22 What does this say about MNPI continued use? Are ;
23 you saying they have to validate it for the nuclear use or l 24 are you saying that it is acceptable as is?
25 I don't know what the implications are. )
i Heritage Reporting Corporation O (202) 628-4888 i
i 65' i
.O. 3 MR. BAHADUR: Sandy, I see you raise your hand.
2- . Will you take this question?
3 DR. FRATTALI: We're saying the test, if the
~4 licensee chooses to use the MMPI, we're saying that's a 5 valid test.
6 We're not saying that the licensees have to 7 validate it for the nuclear industry. That's not what we're 8 .asking.
9 There are some psychological studies being done 10 specifica11y'with populations in the nuclear industry that 11 are looking to find valid tests for the nuclear industry, 12 but they are not yet available, they are not yet developed.
13 MMPI is a reliable and valid test. IPAT is a
() 14 reliable and valid test. There are other reliable and valid 15 tests.
16 Utilities choose to use MMPI I think because it is 17 .the most well known test. And it is an acceptable test at 38 this point. No one has said it is not an acceptable test.
19 I would rather say it that way.
20 DR. REMICK: So once again MMPI would meet the 21 reliable and valid psychological test requirement that you 22 are proposing for the reg guide?
23 DR. FRATTALI: Yes.
24 DR. REMICK: Okay.
25 MR. BUSH: If I might interject here, don't be Heritage Reporting Corporation O (202) 628-4888 l
66 l (~N_
L \_)'
! 1 misled in that MMPI is the only test --
2 DR. REMICK: No. But somewhere I read that a 3 large number of them were using it.
4 MR. BUSH: Yes, a large number. But there are l 5 many, many sections --
6 DR. REMICK: I see. Okay. .Thank you.
7 MR. MORRIS: I want to stress that if you look at 8 the way the industry guidelines are written and the way'our 9 earlier version of the rule and the reg guide have been 10 written, reliance is not placed entirely upon the screening 11 test.
12 I mean, you are adding into that the judgment of a 13 qualified analyst who could understand what the test means l() 14 and make some kind of a judgment about whether it has 15 indicated something that you would be concerned about or 16 not. And so I think that what we are looking at in this 17 interim period in the fine tuning process is to try to 18 determine just how specific we have to be in our rule and 19 our reg guide about the combination of testing and 20 evaluation that is going to be required or is the standard.
21 And right now I think we are moving in a direction 22 that would say it is either the test accompanied by analysis 23 by a qualified individual or some other equivalent method.
24 So I think that the state of the scientific 25 knowledge in this area is not sufficient that we want to Heritage Reporting Corporation (202) 628-4888
p
- 67 k'
1 lock into anything that would be anything like a requirement 2 for a particular test.
3 And we want to leave a lot of latitude to the 4 industry to come up with new, improved methods. And I think 5 that is the way we are moving right now.
6 DR. REMICK: I understand that. 1 just was trying 7 to get what is the significance of adding the word " valid."
8 What does it really mean, what is the potential impact. And 9 I think you have answered that.
10 MR. CARROLL: I'm not sure you have. I'm still 11 having problems with the semantics here.
12 You're telling me that MMPI is a reliable and 13 valid psychological test. But you are also telling me that
() 14 it has not been validated.
15 Now, there's a disconnect there. What is the 16 meaning of the word " valid" in this context?
17 DR. FRATTALI: I have to be careful, because I am 18 not a clinical psychologist. So excuse me if I get this 19 wrong.
20 MR. CARROLL: l'm not, either.
21 DR. FRATTALI: Okay. When you validate the test, 22 you are also looking at a population.
23 Now, it is my understanding that there is going to 24 be a new version of the MMPI that has been validated for 25 different populations. And as it happens, one of the
(, ' ' Heritage Reporting Corporation N
(202) 628-4888
l-68 O 1- populations, I have been told, are nuclear operators. These 2 are operators in the control room. This is not valid for 3 the sense that we are looking at here for access 4 authorization.
5 But this test has taken a population, the 6 operators, has determined some characteristics that they
.7 felt made for the characteristics that they want, and 8 actually provided a data base that validate the test.
9 Now, for the test in general, the test that has 10 been moved, the population that it was validated for, the 11 original test, the test that the licensees are using now, 12 was not nuclear power plant operators.
13 So that, although it is a reliable and valid test
() 14 for a general population, psychologists don't like to say 15 for a specific part of the population it is a valid test.
16 They don't want to say it has been validated for a 17 nuclear power plant, for this. And that is why we can't 18 require a valid test for the nuclear industry if there isn't 19 a valid test, a test that has been validated, for the 20 nuclear industry. I guess that is what we are coming around 21 to.
22 MR. CARROLL: Okay. So why are you then taking 23 exception to the industry guideline that just wants to say 24 reliable?
25 MR. BAHADUR: Sir, that is the question exactly O Heritage Reporting Corporation (202) 628-4888
69 1 that was asked by.CRGR about two weeks back.
2 The way I understand, and as I mentioned earlier, 3 we are no longer very strong on this exception.
4 There are thoughts whereby we are saying we can 5 just qualify the language in our regulatory guide rather l
6 than take an exception.
7 The way I understand is-this. MHPI is there. The 8 n ajority of the utilities are using it, it is a widely 9 acceptable test in the nuclear surroundings. It has not 10 been validated, because the need wasn't there.
l 11 Should the rule require that a test be validated, 12 it is a different issue. The rule only requires that the 13 test should be valid and reliable. It is asking for a
() 14 judgment call by the management of the utility which is 15 going to perform the test.
16 How strongly the staff feels about this: we are 17 at the crossroads, not sure whether we are going to take the 18 exception, we are going to make the exemption, or we are 19 just going to clarify the language in the regulatory guide.
20 MR. MICHELSON: Could you tell me just briefly --
21 MR. MORRIS: Let me try. I think we are involved 22 in a semantic mire here.
23 The term of validation, the term valid, is being 24 used I think in two different ways, sometimes in the same 25 sentence, I think.
O Heritage Reporting Corporation (202) 628-4888
s 70
? . ]
1 I believe that in some sense the psychologists, l 2 the scientists would like to think validation, as Dr.
l l 3 Frattali has been saying, as testing the test with different 4 types of populations for different types of characteristics.
5 And that has not been done to include the l
6 population of employees at nuclear plants with regard to the 7 possibility of committing radiological sabotage. That has 8 not yet been done in the scientific sense that a 9 psychologist would like for it to be done in order for him 10 to say this is validated.
11 What we are saying is we believe that in the use 12 of the testing as a screening tool accompanied by an 13 assessment by a trained psychologist, qualified psychologist
() 14 or other medical person, that the test would be valid in 15 order for that person to use it, accompanied by the 16 management of the utility to make a judgment about whether 17 or not they would make a determination that the person is 18 reliable and trustworthy.
19 So I would say that our use of the term valid is 20 in that more general sense and is not in the sense of the 21 psychologist's scientific use of the term valid.
22 In that regard, having defined what we are 23 generally thinking about in terms of valid, Mr. Bahadur said 24 we are now trying to determine what the worth is of using 25 both attributes, reliable and valid, in an exception to the
, Heritage Reporting Corporation
' (202) 628-4888
O 'l industry guidelines.
2 And as we go through it more and more and discuss 3- it with you and the CRGR and others, we are finding it less 4 and less tenable to try to use two attributes. And it is-5 too complex an issue to make it more complex by adding a 6 bunch of modifiers to the testing.
7 And I think we are moving in that direction now.
8 L
9 10 11 12 13 O 14 15 16 17 18 19 20 21 22 23 24 25 Heritage Reporting Corporation O (202) 628-4888
gs 72 1 MR. MICHELSON: Would you tell me just briefly why 2 you validate psychological tests. Maybe that will help me 3 understand what valid means. I think that I know how you do 4 it, but now I am not sure.
5 How trould you validate a psychological test to 6 determine that the person might be prone to cauce 7 radiological sabotage?
8 MR. BUSH: If I could, let me try to answer your 9 question and back up a little bit and try to answer what the 10 difference is between reliability and validity. Reliability 11 means that you consistently get the same result. Validity 12 means that you get the result that you are looking for.
13 MR. MICHELSON: What is a result in a case of a A
(_) 14 psychological test?
15 MR. WARD: If I could ask a question which would 16 help me and I think that it might clarify. When you say 17 valid, that the test is valid, you want it to be a valid 18 indicator of certain characteristics in the person.
19 Now what are those characteristics?
20 MR. MICHELSON: I think that you must have 21 individuals who have been shown to have done certain things, 22 and they must also have taken a test.
23 MR. BUSH: I was going to get into that.
24 MR. MICHELSON: I do not know how you do that in 25 this case, f')
Heritage Reporting Corporation (202) 628-4888
73 1 MR. BUSH: If I can use a variation of the example 2 of colors, for example the reliability. If you are looking 3 for the color red, the test each and every time will 4 identify the color red when it is present. If it is not 5 valid, you think that you are measuring for the color red 6 but you are really measuring for the color blue. So you are 7 being misled as to the results.
8 Now the problem with validating a test is that you 9 usually need a sizable population of people who passed the 10 test and people who failed the test. In the population at 11 large, we have people who are in prisons. So the 12 psychologists are able to go get profiles from the people 13 who are in prisons or who exhibited antisocial behavior or
() 14 things of that nature. So they know that these are the 15 people who are the outliers. So they can validate the test 16 as identifying it.
17 The problem with the nuclear industry if we are 18 going to look for people who have committed sabotage, we do 19 not have a population of people who are in prison for having 20 committed sabotage who we can go administer a test to and 21 find out what kind of results there are. So that is where 22 the problem is in validating the test for the nuclear 23 industry.
24 MR. MICHELSOH: And hopefully you will never have 25 such a population.
Heritage Reporting Corporation (202) 628-4888
I 1
i 74 1 MR. BUSH: E:actly.
2 MR. MICHELSON: So valid then blows my mind as to 3 what it means.
4 MR. BUSH: So now what we are limited to really is 5 the gathering of the population at large and the prison 6 populations and looking for antisocial behaviors or 7 behavioral defects or personality defects that would reflect 8 the type of person that we would be a little bit 9 apprehensive about.
10 MR. MICHELSON: So you never can really validate 11 it for the nuclear industry unless you sort from these 12 prisoners the prototypes of those who are in the nuclear 13 industry somehow, and say that is typical of nuclear
() 14 industry people who might do this. That is the sort of 15 thing that you h>re got to do, and that would be very 16 difficult I think.
17 MR. BUSH: So I think that reliable is probably 18 the right word.
19 DR. REMICK: It sounds more and more like this is 20 not really an exception but it is a clarification. I think 21 that we best proceed.
22 MR. BAHADUR: The third exception is regarding the 23 review process. The industry guidelines indicate if a 24 permanent employee of the utility is not given the 25 authorization for access then he had the right to appeal.
O Heritage Reporting Corporation (202) 628-4888
n 75
\~)
1 The appeal proces. is only extended to the permanent 2 employee of the utility.
3 The exception that the staff is making is first of 4 all the appeal process should go to the employee of the 5 vendor and the contractor also. And secondly, it should 6 only be limited to permanent but to temporary employees 7 should also be included in that.
8 It is a debatable issue. NUMARC has come up with a 9 very strong position against it. It is a legal issue. Our 10 lawyers are battling with NUMARC lawyers. And it is not a 11 very clear cut issue either. Because both lawyers are using 12 the same cases to support their cause.
13 So at this time we have an open mind. If we are
() 14 told that legally it is not necessary for us to have this 15 exemption, we will drop it.
16 MR. MICHELSON: Are you going to tell us more 17 about this later or is this it?
18 MR. BAHADUR: I wish I could tell you moro about 19 it, sir.
20 MR. MICHELSON: What you have told me, I really do 21 not have any basis to agree or disagree with the exception.
22 I de not understand it yet.
23 MR. BAHADUR: And as I understood myself, I 24 consider this to be more of a legal issue rather than 25 technical.
[]' Heritage Reporting Corporation (202) 628-4888
7 m, 76 sj 1 MR. MICHELSON: There are technical issues 2 irrespective of legal issues. The question is who can 3 commit sabotage, only permanent employees or can temporary 4 employees do it as well. And the answer is clear, it is 5 either one can do it.
6 MR. BAHADUR: Yes, and that is a very good 7 question. Our lawyer, Bob Farmer, is not present here. But 8 if Mr. Morris would like to add something.
9 MR. MORRIS: I think maybe to clarify this. This i
10 review process does not have to do with whether or not you 11 make a review to grant or deny unauthorized access. What it 12 has to do with is upon denial would there be an appeal 13 available to a temporary contractor or employee.
() 14 MR. MICHELSON: I am sorry, I misunderstood it 15 completely. That is not what I thought you meant. The 16 whole bullet confused me I think.
17 MR. MORRIS: That is not at issue at all. This 18 does cover anybody who goes on the site, this background 19 check and so forth.
l 20 MR. MICHELSON: And NUMARC did not take issue with 21 that?
22 MR. MORRIS: That is not the issue. It is a legal 23 issue about the rights.
24 MR. BAHADUR: Maybe I misguided you.
25 MR. MICHELSON: Well, I just am not thinking too
() Heritage Reporting Corporation (202) 628-4888 !
- _ - _ - - - - - - - - - - - - - - - )
77 I.
1 well.
2 MR. BAHADUR: The way that the requirements are if 3 the person goes through the access authorization application 4 and his application is rejected and he is not granted 5 authorization, and as a result of this rejection he loses 6- his employment, then should he or should he not have the 7 right to appeal. That is the question.
8 MR. MICHELSON: I see. That is a different issue 9 entirely.
10 .DR. REMICK: It is a due process consideration.
11 MR. BAHADUR: NUMARC says a permanent employee of 12 the utility will have that right. The staff while 13 developing this exemption says it also should be extended to
() 14 temporary employees and also the employees of the 15 contractor.
16 The fourth and the last issue is the issue of 17 audit. The guidelines are that a utility will audit its 18 authorization program every two years. But the vendors' and 19 contractors' programs would be audited every year. The 20 staff agrees with that but adds a little clarification. If 21 a utility subcontracts a portion of its program, that 22 program should also be audited once a year. An exemption l
23 was to that effect, and it is quite likely that we might 24 clarify that in the regulatory guide. And again there could 25 be a different interpretation.
Heritage Reporting Corporation
(*~} (202) 628-4888
c._ - _ _. _ _ _ - _ ._ . - _ _ _ _ _ _ - _ - _ - _ _ _ _ _ _ _ _ _ _ _ -
1 1 :.
gJ 78 1 DR. REMICK: Excuse me, a question for
- 2. clarification.
3 If they subcontract part of their program, do you 4 .mean their access authorization program?
5- MR. BAHADUR: Right. For example, in the 6 background investigation they subcontracted the credit check 7 and the criminal history for example. The exemption says in 8 that case that the utility should audit its program once a 9 year and not twice.
10 MR. CARROLL: In its entirety or just that part?
11 MR. BAHADUR: That is where lies the 12 interpretation. Some of the staff are very strongly of the 13' opinion that it means the entire program. Others have
() 14 indicated that they meant only the contr actor. And that is 15 a clarification that we have to develop. That is the basis 16 that we are right now in the process of developing.
17 Sandy, do you have something to add to it?
18 DR. FRATTALI: I think that it was the other way 19 around. We feel strongly that only the part that is 20 contracted out needs to be audited every year, but there is 21 some feeling that maybe the whole program should be audited 22 every year.
23 MR. BAHADUR: So there are two schools of thought 24 right now and we are in the process of developing it.
25 MR. CARROLL: Well, let me ask this, assuming that Heritage Reporting Corporation O* (202) 628-4888
l'
. k' 11 'you do not contract any of it out, how often should the
-2 program be audited?
3' MR. BAHADUR: Once in two years.
4 MR. CARROLL: Once in two years?
5 MR. BAHADUR:- Correct.
6 MR.-CARROLL: Okay.
7 MR. MICHELSON: There is one item.that did not 8 appear -- go ahead, because I have got another subject.
9 MR. CARROLL: And what does this audit consist of,
- 10. how do I do this, do I get somebody who is not directly
-11 involved'in the' program who works for my utility, the QA 12 department or somebody?
13 MR. BAHADUR: Dr. Frattali.
- ( ) 14 DR. FRATTALI
- The minimum audit criteria are part 15 of the guidelines, so that is how you do the audit. The
' 16 rule has the requirement or the -- I guess there are two 17 things.- Statements of consideration address that issue and 18 says that the utility can do its own internal audit provided 19 that it is done by qualified people not involved in the 20 program, for example a quality assurance person if he is 21 qualified in security could do the audit.
22 And that is the intent of the rule and there are 23 some words under audit procedure that support that, in the 24 rule on how to do the audit. The actual criteria for the 25 audit are given in the guidelines.
O Heritage Reporting Corporation (202) 628-4888
r~s . 80 1 MR. CARROLL:- And that is every two years?
2' DR. FRATTALI: Every two years, yes.
3 DR. REMICK: And there is a transferability of
-4 that audit from one utility to another, is that right?
5 DR. FRATTALI: Yes.
6 DR. REMICK: So not every utility would be going 7 into the same contractor and redoing the same audit, is that 8 correct?
9 DR. FRATTALI: Yes.
10 MR. MICHELSON: One item that did not appear as an 11 exception which bothers me just a little bit, and maybe you 12 can explain to me what the grandfathering provision of this 13 guideline and regulatory guide will be.
-( ) 14 DR. FRATTALI: The rule provides for 15 grandfathering the authorization. That any person who has 16 the access authorization on the day or within 365 days of 17 the date of the promulgation of this rule would 18 automatically be granted access authorization.
19 MR. MICHELSON: What does mean though, granted 20 under what program. Maybe I have got a utility which had a 21 minimal program and did not meet the guidelines at all, but 22 it granted so-called access authorization.
23 That access authorization now grandfathered under 24 this more comprehensive program?
25 DR. FRATTALI: That is correct.
O Heritage Reporting Corporation (202) 628-4888
81 e3
,V 1 MR. MICHELSON: Can you explain why that is 2 acceptable?
3 MR. BAHADUR: This provision reflects our 4 confidence in the existing system.
5 MR. MICHELSON: I do not have any basis for 6 confidence.
7 MR. BAHADUR: For adequate protection that the 8 utility is committed to.
9 MR. MICHELSON: If you are really confident that 10 they are already doing it, you do not need any of this. But 11 there is some concern about that level of confidence as well 12 as making this thing a little more legal. You know, it 13 bothers me a little bit that you let minimal programs go
( 14 through in perpetuity and the guy is never checked. If he 15 got in under the rule and under the grandfather, he will 16 never be checked.
17 MR. BAHADUR: And you can approach this question 18 from the angle that currently the utilities are at the level 19 of providing adequate protection and we are talking about 20 enhanced after.
21 MR. MICHELSON: You mean already now that they are 22 already adequate?
23 MR. BAHADUR: The way that we see the commitment 24 of the utility on paper in their security plans that the 25 commitment to either 18.17 or 3.3 is there.
Heritage Reporting Corporation (202) 628-4888
p57) d i: : ,
p- .
' -~ 82
?r.
E 1 MR.'MICHELSON: For every utility?
E 2- MR. BAHADUR: What the guidelines have'done is to 3 raise that level of adequate protection to the enhanced 4 safety.- .The insider rule whereby we'went to the fitness for 5 duty, the random check for drugs, and then the access-6 authorization, these are all combined to provide you 7 enhanced safety and enhanced assurance against the insider 8 committing the sabotage.
[ 9 MR. MICHELSON: Really only enhanced assurance 10 .against.a new potential. insider coming in but not the ones 11 who are already there.
~12 MR. MORRIS: Let me see if I can help with this 13 one a bit. We are talking about people who are employed by
() 14 the utility who have already been granted unescorted access, 15 who have been at the utility for some period of timer.
16 MR. MICHELSON: Right.
17 MR. MORRIS: Here is where the factor in my mind 18 at least comes into play with regard to the fact that there 19 is now a familiarity with that person that is not achievable 20 for a new person coming onto the site for the first time.
21 So now for those people for whom you have some familiarity, 22 we believe that the rationale would be that you would not
- 23 need to go back through and go through those elements that 24 might have been missing from your existing access 25 authorization investigation program, to go back and redo all Heritage Reporting Corporation (202) 628-4888
83 0 1 of that for all of them.
2 MR. MICHELSON: It is not a question _of redo. In
- 3. my case, it was never done.
4 MR. MORRIS: Yes, fill in the gaps of the program.
5 Because what I am saying is that what I think fills in that 6 gap'is the continued familiarity with the person's 7 reliability and trustworthiness by having an association 8- with him. So the management has made that judgment and 9- continues to make that judgment, and we do not see a need to 10 go back.and do parts of a program simply because he did not 11 have them in effect earlier on. That is the rationale for 12 it. The basic rationale I think is that continued 13 association with the individual.
() 14 MR. MICHELSON: But I think that I can correctly 15 characterize it that if we do have insiders there that we
}-
16 are unaware of that that will remain, that we will only r
17 screen new people coming in for this potential. If they i 18 never took a psychological test, we are not now going to 19 give them one. If we never checked their background, we are 20 not going to check it now and so on. One of the elements 21 will be implemented of course.
22 MR. MORRIS: It could be, and the other two would 23 be absent. There is one factor that I think that I will 24 remind you of. These elements, these basic elements, of the 25 program are not perfect. And we do not claim that you are Heritage Reporting Corporation O (202) 628-4888
84 1 going to capture every potential saboteur with this kind of I
2 a combination. There is certainly a possibility that you i
3 could have someone there, or you could have someone even 4 after implementing this program who would get unescorted 5 access.
6 So there is a small residual possibility that 7 there would be a problem that would be lingering. But we 8 are going to be looking at the reinvestigation issue, and 9 that may bring this issue up again.
10 MR. MICHELSON: What would you investigate if you 11 never investigated to begin with? You would still have the 12 grandfathering I assume.
13 MR. MORRIS: That has not been decided yet, just
() 14 how we are going to deal with that.
15 MR. MICHELSON: So really the rule might go beyond 16 just the question of reinvestigation. It might be 17 questioning also the grandfathering provision.
18 DR. FRATTALI: It is not quite as clear cut.
19 First of all, as far as we can determine, the psychological 20 evaluation, that everyone has had the psychological 21 evaluation.
22 MR. MICHELSON: Every utility has done this you 23 are saying.
24 DR. FRATTALI: Even the two that are not committed 25 to it in the plan, who were not committed to do it.
O
< Heritage Reporting Corporation (202) 628-4888
<- 85
\_/
1 MR. MICHELSON: Were they done by competent 2 psychologists and so forth as required by the guidelines or 3 was it just a piece of paper that they handed out and filled 4 out?
5 DR. FRATTALI: I do not know, I am sorry. But 6 what I really want to point out that last year or two years 7 ago that we had FBI criminal history check. So that 8 everybody currently employed by the utility has had a 9 fingerprint check. Now this pretty much fills in a very big 10 gap for these people who are grandfathered.
11 MR. MICHELSON: Does that also include the 12 temporaries and so forth that would be under this 13 guidelines?
( )- 14 DR. FRATTALI: Yes, 7357 is a separate rule that 15 requires anybody who sets foot on the site of the utility 16 have submitted their fingerprints for a FBI criminal 17 history.
18 MR. MICHELSON: So part of the background 19 investigation is assured anyhow even with grandfathering?
20 DR. FRATTALI: Yes.
21 MR. MICHELSON: So the only part then -- well, you 22 also assured me that the psychological test at least had 23 been administered and whether or not by appropriate people 24 is another question. But you have already done all three 25 elements for the previous employees then. You clearly have O Heritage Reporting Corporation (202) 628-4888
(- 86 1 observed their behavior for at least a year. You did the 2 psychological assessment.
3 MR. CARROLL: No, not necessarily.
4 MR. MICHELSON: I thought she said that they all 5 had done it.
6 MR. CARROLL: At least in my experience there was 7 the three year rule. If I had been employed by the utility 8 for three years and my supervisor was willing to say that I 9 was reliable and trustworthy, I did not take the test.
10 DR. FRATTALI: I was referring to what the 11 procedure was today. There may be people far enough in the 12 past that -- if I could give just a little bit of history.
13 A prior version of the rule grandfathered only people who
() 14 had been cleared by one of the ANSI standards, 18.17 or 15 3.3. As you saw from the survey, all but two had been 16 cleared by those programs. Had they been cleared by those 17 programs, they had a psychological evaluation of some type, 18 that was the requirement.
19 What we are saying now is that we do not need to 20 have that caveat in there any more for grandfathering, 21 because basically we feel comfortable that if not everyone 22 then most people have had the psychological evaluation and 23 perhaps everyone has had it. That is why we removed that 24 cavedt from it.
25 MR. CARROLL: I do think that is right. I come
(' Heritage Reporting Corporation (202) 628-4888
87 1 from the utility industry. When I retired two years ago, I 2 had never had a psychological evaluation. Every year my 3 supervisor, I do not know if it was every year but at some 4 interval, had to sign a piece of paper and I used to sign 5 them for people who worked for me that I was grandfathered 6 because I had been a reliable employee for the last three 7 years or something.
8 MR. MORRIS: I believe when I mentioned this 9 earlier that my reasoning was that that factor would weigh 10 on my mind as the basis for grandfathering. And even though 11 as Sandy says that we think that there are many programs 12 which includes psychological testing that do not necessarily 13 include it as a written commitment but do so maybe
) 14 voluntarily that we cannot really attest and cannot document 15 as the basis for this decision any particular fraction of 16 the industry that does that.
17 I mean it is a general belief, but I think that it 18 is supplemented by that feeling that there are a number of 19 people getting psychological tests, and the fingerprint rule 20 is the continued familiarity with the trustworthiness and 21 reliability of the employees.
22 MR. MICHELSON: Well, the grandfathering provision 23 in the guidelines seems to be quite clear. You do not 24 really need to do anything if you have been around for 25 awhile like 365 days. Yet you tell me that all of this
'~)/ Heritage Reporting Corporation (202) 628-4888 l
t 88
,q'.
(_/. I stuff is done anyhow so that it is no problem.
2 Why do we not just add these as caveats in the 3 unlikely event that it has not been done. I really believe 4 that some kind of a check ought to have been done on people 5 who have been there 366 days besides just the behavioral 6 observation for that one year period.
7 MR. BUSH: I might make a comment that I am aware 8 of a few cases where the licensees got what would be termed 9- adverse information back from the FBI as a result of the 10 fingerprint check and there were criminal records. And I am 11 not aware of anybody, although there might have been some 12 people who were terminated because of the adverse 13 information.
I) 14 Basically what happened in many cases and in fact 15 one case the individual had been employed for over ten years 16 even though he had a criminal record, or in other cases 17 convictions involving drugs or something like that ten or 18 fifteen years before. And the licensee in the cases that I 19 am aware of chose to weigh heavily on the fact that the 20 individual had been a faithful employee for ten or fifteen 21 years and did not take action to terminate.
22 MR. MICHELSON: Well, I was not questioning 23 whether or not action was taken. I was just questioning 24 whether or not the information was even available to the 25 utility. And I think that you have assured me that these i
O Heritage Reporting Corporation (202) 628-4888 4
i 1
89 LO. .
1 tests were done and so forth. Why not put them in as 2 caveats then.
3 MR. BUSH: I guess that what.I was trying to point 4 out is based-upon what we have seen so far that even if they-
- 5. did come up with old information or adverse information that 6 the utility probably will'not take action anyway.
7 MR. MICHELSON: That.would be their' judgment to 8 make. It is our judgment as to what kind of.information 9 they should have in front of them when they do make their 10 -judgment. And you are saying none if it is over 365 days.
11 If I am reading the grandfathering hopefully correctly, it 12 says that we do not need to know anything about him-if he 13 has been working for us for 366 days.
I) 14 MR. BUSH: The grandfathering has no relationship 15 with the 365 days.
16 MR. MICHELSON: Well, it is in the grandfatharing 17 provision. Non-escorted access authorization with 365 days, 18 then you do not need to do this.
19 DR. REMICK: But that is assuming that the person 20 had been granted unescorted access under the' program at that 21 time.
22 MR. MICHELSON: Yes.
23 MR. BAHADUR: And that it did not expire.
24 MR. MICHELSON: Yes. If I have got a program in 25 effect for 366 days, anybody coming in at that time or Heritage Reporting Corporation (202) 628-4888
- s. 90 J 1 before is not automatically grandfathered irrespective of 2 what that program may have required if anything at that 3 time. I do not know what it might have required two years 4 ago.
5 MR. BAHADUR: That is true.
6 MR. MICHELSON: And I think that some caveat ought 7 to be on that.
8 MR. MORRIS: We thought about this at some length, 9 and we felt that it would just not be cost effective to 10 force the issue of going back and filling in these elements 11 that had not been done previously and we did feel that we 12 could justify it.
13 MR. MICHELSON: I thought that you almost assured
() 14 me a moment ago that you have done all of this, I mean that 15 the utilities have done this, that they have done the 16 psychological tests and so forth under the ANSI standard, 17 and therefore it would not be a big burden.
18 MR. MORRIS: Let me be clear. We looked at 19 commitments that a sample of twenty utilities had made in 20 their licenses, and we found a range of different degrees of 21 specificity in those commitments, some even including very 22 little information about what the commitment would be.
23 Agreed that those few percent may not have done a 24 significant fraction of the elements of the program.
25 Nonetheless those employees would have been employed there f' Heritage Reporting Corporation (202) 628-48B8
91
, A 1
\_)
1 for some period of time, at least-a year.
2 MR. MICHELSON: Not a year -- well, yes, one year.
3 MR. MORRIS: About a year. So you would have some 4 chance to determine whether they were likely to be 5 untrustworthy and unreliable. And a combination of that 6 factor plus the fact that we do believe that a large number 7 of employees have been subjected to most of the program led 8 us to the conclusion that it would not be cost effective to 9 go back and put a requirement in that would require that 10 these gaps be filled in so to speak.
11 MR. MICHELSON: I do not know what cost effective 12 means, because you never determine benefits but you just 13 determine cost. You do not know what the benefit is of
() 14 doing any of_this.
15 MR. MORRIS: Maybe we can just close this. We 16 understand your concern, and we are sensitive to it. Now we 17 have made that judgment, but we are going to be reflecting 18 on this.
19 MR. MICHELSON: If the number had been like three 20 years of employment or something like that, I would have 21 been much less concerned. But one year is an extremely 22 short time in which to do the behavioral observation. You 23 know, lacking no other information, I would think that you 24 would want to go back a little further or alternatively 25 implement all three major attributes of the program.
O Heritage Reporting Corporation (202) 628-4888
I 92
' sJ' \
1 MR. MORRIS: We will do that based on what your 2 concerns are.
3 DR. REMICK: It is unlikely that a person would 4 have been granted an unescorted access on the first day of 5 his job, so it would only be 366 days.
6 MR. MICHELSON: I am not going to argue a day or 7 two or a month or two.
8 DR. REMICK: There certainly must be some program 9 for screening those people.
10 MR. BAHADUR: We understand the concern you 11 raised, and I do not think that we have all of the answers 12 at this time. But when we go back, we will rethink about 13 this provision.
() 14 DR. REMICK: Yes. We are running quite a bit 15 behind time. I suggest that we move on.
16 MR. BAHADUR: If I am allowed to continue with two 17 more slides, then I will summarize the presentation.
18 DR. REMICK: Fine.
19 MR. BAHADUR: I want to come to the third question
-20 that was raised by CRGR and that was on the provisions of 21 screening people during the cold shutdown period. As you 22 can see, the provisions are that in a utility which is 23 closed by cold shutdown where they do require a massive 24 number of people to do some job for which the cold shutdown 25 period has imposed on them, that they will need to give O Heritage Reporting Corporation (202) 628-4888 i
l
1 authorization in a hurry.
2 And at that time, they could provide access 3 authorization in those areas which have been devitalized.
4 Now devitalized may not be the right word, since the 5 regulations only define what the word vital is. So let's 6 say those areas which are not considered vital during that 7 time provided that they follow the following three 8 conditions. And these are based on the NUMARC guidelines to 9 some extent.
10 First of all, the background investigation would 11 only be limited to fingerprinting and not the whole detailed 12 investigation as you do for the rest of the authorization.
13 It would not require a psychological assessment, but they
() 14 would undergo the behavior observation.
15 And because we are making these concessions, to 16 make sure that we compensate these concessions without 17 jeopardizing the safety, there are three measures that the 18 utility will be required to take. First of all, they will 19 inspect all of the vital and protected areas after the work 20 has been completed. They will go through the appropriate 21 testing procedures before the start-up. And of course last 22 of all, the access will be granted only to the 12 3 non-operating unit in a multi-unit site.
24 There are some questions in our mind, because 25 first of all we do not know whether the industry right now Heritage Reporting Corporation O' (202) 628-4888 l
l - - _ _ - - - - - - - i
~
l 1 is taking this option at the cold shutdown or are they 2 sending only those people who have been authorized for the 3 access in the normal plant. We do not know that, but that l
4 is one thing that we need to find out from NUMARC.
5 We also need to have more understanding on the 6 test procedures before the start-up that the guidelines 7 indicate. Is it going to the entire year or the test year.
8 And that is something that we need to get a little more 9 understanding on.
10 MR. MICHELSON: Unfortunately there are many dirty 11 tricks that can be played by lifting wires and so forth that 12 are very difficult to find. They are almost impossible to 13 find by visual inspection without going terminal by terminal
() 14 which would be hopelessly impossible or alternatively even 15 by testing. It is no trouble coming up with a number of 16 dirty tricks that will do the job without ever being 17 detectable on the start-up again without very extensive 18 testing.
19 MR. BAHADUR: And your comment reflects exactly 20 the sentiment which was given to us by CRGR. They feel that 21 this provision perhaps is too liberal considering all of the 22 dangers to the safety which might be created by the 23 onslaught of these massive people going through this access 24 in a quick manner. But we need to get more information and 25 we need to get more understanding of where the NUMARC O Heritage Reporting Corporation (202) 628-4888
_ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ~
.-_______-___-___7_
95 O 1 guidelines are coming from.
2 MR. CARROLL: Put into your thinking the fact that 3 I can go down to the drugstore and buy a calendar clock that 4 I can use to trigger a bomb at any time in the next fifty 5 years if I want to program into it.
6 MR. BAHADUR: These are the three major concerns 7 that the CRGR raised about two weeks back. I have tried to 8 give you some idea about the current thinking that the staff 9 had on this. And based on that, that seems to be our future 10 course of action. We need to resolve issues with NUMARC, 11 especially the issue on the cold shutdown. We need to 12 understand more about the test procedures that they are 13 talking about before the start-up. We need to get a more
() 14 realistic picture of how the industry right now is taking 15 this particular option.
16 And then of course within the NRC offices also we 17 need to resolve certain issues. For example when I say 18 certain issues, some offices are of the opinion that the 19 rule should be very_ general, brief, and refer everything to 20 the regulatory guide and the guidelines. Others feel 21 strongly about putting some flesh and bones into the rule 22 itself. And those are some of the issues that we have to 23 resolve within the offices. We will do that and we will go 24 to CRGR for their concurrence.
25 And I would like to add here that although we may Heritage Reporting Corporation (202) 628-4888
1
,7 96 1 do these changes that the basic philosophy and the basic 2 performance criteria of the rule is going to be preserved 3 the way that I presented that in front of you. We will come 4 back to you if you wish. And then once the package is 5 completed, we will submit that to the Commission for their 6 approval.
7 DR. REMICK: Well, we can decide the question of 8 whether we want you to come back or not at the end of the 9 NUMARC presentation. But my inclination at the moment is to 10 suggest that we not go to the full committee next week but 11 have you come back, but we will discuss that with'the 12 subcommittee later on.
13 Are there other further questions of Mr. Bahadur
() 14 at this point?
15 (No response.)
16 DR. REMICK: Seeing none, thank you very much for 17 the presentation from the staff. I suggest that we take our 18 break now, a fifteen minute break, coming back at 10:50, and 19 we will hear the NUMARC presentation then.
20 (Whereupon, a recess was taken.)
21 DR. REMICK: We will continue our discussion of 22 the access authorization proposed rule. And we will 23 continue at this time'with Mr. Joe Colvin, who is executive 24 vice president from NUMARC. Mr. Colvin, we welcome you back l:.
25 before the subcommittee.
Heritage Reporting Corporation O- (202) 628-4888
h 1 MR. COLVIN: Yes, sir. Thank you, Doctor. What I 2 would like to do is provide you very briefly in the interest 3 of time with an overview of the industry development of the 4 guidelines. I would like to clarify a few of the reasons l 5 behind the development of the guidelines and the commitments 6 that we have discussed, and answer a few of your questions 7 perhaps on the appeal review process or in the areas where 8 there were questions from the subcommittee.
9 And I will have Rick Enkeeboll, our senior project 10 manager on-this issue, provide you with an overview of the 11 guidelines themselves in a very brief fashion or be able to 12 try to respond to any questions that you might have in that 13 area.
() 14 With us today we have some other industry 15 representatives who are really the experts behind these 16 programs. And the representative of the industry, let me 17 introduce him if I could. Chuck Sprunk from Pennsylvania 18 Power & Light, Director of Corporate Security. And Chuck 19 goes back a long way. There is a little bit of deja vu in a 20 lot of the questions and comments. He worked with us on the 21 working group on security that was one of the two groups 22 that developed the guidelines.
23 Jerry Ellis is the Security Manager from System 24 Energy Resources, Inc. John Ross is Security Planning 25 Specialist from Baltimore Gas & Electric. And on John's Heritage Reporting Corporation O (202) 628-4888
98 O~ 1 right 'is Bruce Martinis who is a clinical psychologist from 2 Baltimore Gas & Electric Company.
3 So we have brought people who might be able to 4 respond to your questions, and we would try to.have these 5 gentlemen respond in areas where they might' provide the 6 proper answer.
7 MR. MICHELSON: Just for clarification, is that a 8 degreed clinical psychologist?
9 MR. MARTINIS: Yes.
10 MR. MICHELSON: Thank you.
l 11 MR. COLVIN: Let me talk about the background of 12 the guidelines.
13 (Slides shown.)
() 14 MR. COLVIN: In 1984 as it was indicated the l 15 Commission issued a proposed rule. We took a look at the l 16 proposed rule and had some concerns about the rule, some of 17 the requirements there and the impact on the industry. And 18 as a result we formulated a NUMARC working' group that was 19 first chaired by Bud Faye, Vice President of Nuclear at 20 Wisconsin Public Service.
21 We brought together a group of both operating 22 people and security specialists to try to provide a 23 guideline that the industry could utilize. We also brought 24 into that people with experience in ANSI 3.3 and 25 ANSI 18.17, and the people who have been involved in these O. Heritago Reporting Corporation
.U (202) 628-4888 1
i 99
, -)
'w) 1 programs and processes.
2 The thought was that we could provide a guideline 3 that everyone could use that would address the issues before 4 the Nuclear Regulatory Commission. So we petitioned through 5 a meeting with the Commission and received the opportunity 6 to brief the Commissioners and stress that we would prefer 7 that the Commission issue a policy statement in this area 8 that would endorse the industry guidelines.
9 So we started that work in the early January time 10 frame of 1985. Just a few months after that, we had 11 developed some drafts of guidelines. And by August of 1985, 12 we had already gone through seven revisions to that proposed 13 guideline. We were addressing many of the same areas that
() 14 are being discussed today, and we had the endorsement by the 15 NUMARC executive committee. If you recall, NUMARC had a 16 committee at that time and had representatives from the 55 17 utilities licensed to operate or construct nuclear plants.
18 And that was the body of the executive committee that 19 endorsed those guidelines.
20 As the slide shows, we then in January of 1986 21 moved to have not only the committee endorse a later 22 revision to that guideline but to endorse the concept of 23 implementing changes to their security plan incorporating 24 the guideline elements Lito their security plan such that it 25 would be inspectable and enforceable by the Nuclear
(]
11eritage Reporting Corporation (202) 628-4888
.i 100
'O 1~ Regulatory Commission.
2 I think that this was kind of an interesting twist i
Because
~
3 on the typical policy statement versus rule issue.
4 we did not have an organization in the industry to really do 5 the. follow-up like we do in some other areas, like INPO does 6 in the operator training and in areas such as operating 7 performance, maintenance and so on. So we wanted to look at 8 a way where we could get the guidelines incorporated and 9 have the follow-up to ensure that they were indeed 10 implemented properly.
11 So the intent was'to have the industry commit to 12 implement the elements of the guidelines into their security 13 plans through an amendment letter and by increasing the
() 14 level of detail in the security plans, an increase in the 15 requirements, and you can do that without an amendment to 16 the license. And because the security plan is part of the 17 license, that would be inspectable and enforceable by the 18 NRC.
19 I might say that we had a lot of momentum working 20 at that time. Between the staff and the industry, we had 21 the objective to get the guidelines in place and utilized, 22 but we really never accomplished that objective even to 23 date. So you might say that there is a lot of interest in 24 taking this issue and getting it resolved and moving on from 25 both the staff's and the industry's side.
Heritage Reporting Corporation O- (202) 628-4888
.g~g 101
(l 1 In June of 1986 as you are aware, the Commission 2 did vote to direct the staff to develop a policy statement 3 obtaining the commitment again from the industry and 4 incorporate the guidelines into the security plans, and that 5 was publicly noticed. Comments were received on both the 6 policy statement and the guidelines, and then we tried to 7 take those into account.
8 And as a result of that what you have is an 9 updated Revision 8. We will move on through these slides, 10 ., since we have covered most of these points already. In May 11 of this year, we met with the staff and we talked about the 12 public comments that were received and tried to incorporate 13 as many of those public comments as were reasonable into the f( ) 14 guidelines. We then modified the guidelines and published 15 those in the document that you have in front of you. And 16 now we are here meeting with the ACRS Subcommittee on this 17 proposed rule, abbreviated proposed rule.
18 Okay. Let me talk about the previous initiatives 19 for such a second. I would like to come back to that, since 20 in reality we have no commitment on behalf of the industry 21 to utilize the guideline as it has been developed at this 22 time. Again I want to stress that the commitment that the 23 industry made to the NRC was that if the NRC endorses the 24 guidelines in a policy statement that each utility would 25 incorporate those guidelines into its security plan through Heritage Reporting Corporation (202) 628-4888
I: I L j i
I 102 i
.c .
1 an amendment and move on.
2 Since the latest decision by the Commission is to 3 issue a rule, that commitment is no longer valid. 'We think
]
1 4 however that the majority of the industry has been utilizing 5 the guidelines in one way or the other as the reference 6 . document for their access authorization program. So we 7 think that there is a lot of validity to continuing to 8 pursue the completion of these guidelines, and we certainly 9 are agreeable since the Commission has decided to.go forth 10 with the rule.
11 We are agreement to endorsement of these 12 guidelines through a regulatory guide, and obviously we 13 would like to minimize the number of exceptions, additions
() 14 and deletions to those programs, and we are working on that 15 with the staff.
16 Let me move on and talk just about a couple of 17 other issues that came up before I turn it over to 18 Rich Enkeeboll to talk about the details on this. But there 19 has been a question on the appeal process and the review 20 process, and I would like to give you perhaps a perspective 21 on that.
22 As was indicated, when you get attorneys in there 23 that you can get attorneys to come up with whatever position 24 that they want to come up with in areas where the law is not 25 very clear. This is an area where I think that there is a Heritage Reporting Corporation (D/ (202) 628-4888
- 1. lot of clarification that is questionable as to what the 2 case law is on constitutional rights.
3 And quite honestly we started back in the days 4 with the proposed rule such that the appeal requirement in 5 the rule provided rather onerous provisions for a person who 6 was granted or not granted unescorted access, or where his 7 employment was terminated by the fact that he or she was not 8 granted unescorted access.
9 I mean it would even have led to perhaps an 10 adjudicatory hearing before a judge and a court and other 11 things. So we were very concerned about that and did some 12 legal analysis as to whether the NRC was required to protect 13 the individual's rights when it issued rules and so on. And
() 14 . we provided those legal analyses to the staff in the past.
15 When the decision was made to go forth with the 16 policy statement, then that aspect of it became a 17 non-issue. And although the protection of an employee's or 18 an individual's rights was important, by virtue of the fact 19 that we were going to put that in a policy statement and 20 that it would not be a rule issued by the Nuclear Regulatory 21 Commission, we worked out a process that we called the 22 review process, such that for permanent employees of a 23 utility that we would provide a process by which they would 24 be able to appeal up through some other higher level of 25 management and outside the chain of management of those who O Heritage Reporting Corporation (202) 628-4888
1 104-0 1 made the decision as to why the unescorted access was not 2 granted, otherwise to provide a vehicle for those employees 3 to understand why they were not granted unescorted access.
4 And when you get into the question of whether it 5 should apply to permanent employees versus all employees or 6 temporaries, you really come up with some questions from a 7 common. sense perspective that just do not make sense in our b opinion.
9 Simply by virtue of a individual requesting 10 employment at a utility company and that employment being 11 denied for reasons of some element in the criminal history 12 check, some character reference that was not developed, or 13 some other possible piece of information that the utility
(). 14 understands, that individual could then be if there was a 1S rule requiring that be given rights to employability that he 16 or she would not otherwise have.
17 If I went to apply at Rich Enkeeboll's plant and I 18 was denied employment because I was not able to get 19 unescorted access to the power plant to do a particular 20 craft function or work function that you were trying to hire 21 me for, then you have now denied me employability, and I 22 have the right to appeal through this level of management.
23 And if it went to some grievance or some court hearing or 24 something else, then the utility could be directed through 25 that body to either grant unescorted access to the O Heritage Reporting Corporation (202) 628-4888
l 105 1 individual'or'to grant them employment.
I 2 Now if you cannot use the individual in the power i
3 plant with unescorted access and you are required to employ 4 them, then you are left with very few options. You have to 5 escort them through there. And when you look at these kind 6 of common sense approaches, then it raises a lot of 7 questions.
S ?o we limited the employability issue or the issue 9 of appeal or review process to permanent utility employees.
10 Because the real concern was what about the employee that 11 you take from one of your fossil stations and you are going 12 to transfer them to one of your nuclear stations, and he or 13 she has been an employee with your company for a number of
() 14 years. Now you find something in their background that 15 makes it questionable as to whether the unescorted access 16 should be granted. So now that employee ought to be able to 17 understand why that is the situation.
18 The contrast to that is if you hire for example a 19 NSSS vendor to do outage work perhaps, to hire General 20 Electric Company to bring in a team of people to do pipe 21 replacement on the recirculation pipe. If you have that 22 review process or appeals process in place, then if you deny 23 one of their employees access for whatever reason, then that 24 employee now has rights under your program to employability 25 where he might not otherwise have within his own company.
O Heritage Reporting Corporation (202) 628-4888 u__--_-_-_---
rw 106 b
- l. So you get into a dual employability issue and you can get 2 tied up' legally.
3 So it is a very complex issue, and we are going to 4 continue to work on it with the staff. And I just wanted to 5 give you that, because perhaps that might answer the 6 questions that arose with respect to exactly what that issue 7 is.
8 MR. MICHELSON: I am not sure, but are you arguing 9 that you need a' rule for this particular point?
10 MR. COLVIN: I think that rule or no rule. We 11 argue that you did not need the review process in the rule 1
12 or without a rule that would apply to all employees. That 13 it should not give rights to employment to an' individual'
() 14 that they would not otherwise be eligible for through l 15 collective bargaining agreements or through other l
l 16 . contractual agreements.
l l 17 And I am not arguing for a rule or against a rule.
l le The decision has been made on the rule by the Commission.
19 And what we are trying to do is work out the details.
20 MR. MICHELSON: Given that there will be a rule, 21 you want better provisions to cover this. I think that you 22 are arguing that you do not want it to cover except their 23 own permanent employees.
24 MR. COLVIN: From the industry's position, we 25 would request the appeals rights provided in the rule to not
() Heritage Reporting Corporation (202) 628-4888
107 1 apply to all employees or all people that the utility 2 company would hire. That it would only apply to the i
3 permanent employees of that utility consistent with the 4 approach in the guidelines.
5 MR. MICHELSON: And you want that written into the 6 rule?
- 7. MR. COLVIN: Well, the Commission has directed the.
8 staff to include the appeals process or review process in 9 its rule making activities. So the Commission has made the 10 decision that that is an important element and so that is 11 going forth. The question is as to what scope of people
- 12. does it apply and exactly what the provisions of that 13 process are.
l
() 14 MR. MICHELSON: If it does not apply to other than 15 permanent employees, that means that if you reject a 16 temporary employee that he has no repeal process rights, is 17 that what you are saying?
18 MR. COLVIN: There is a possibility that a 19 temporary employee, perhaps you have hired a temporary 20 secretary. The case in point is that the temporary 21 secretary that you want is to work in your administration 22 building that is inside the protected area. And she applies 23 for employment on a temporary basis, and you intend to l
24 provide that service for an outage, and you go through and l
25 do the fingerprint checks and you do the credit reference
() Heritage Reporting Corporation (202) 628-4888 l
m 108 iv) 1 checks and other things, and you draw a conclusion at that '
2 point that there is something in that person's background 3 that you are not comfortable with granting unescorted 4 access.
5 And remember that the decision to grant or not 6 grant unescorted access is made by the licensee and solely 7 by the licensee, and is based upon as pointed out a myriad 8 of items and not one single thing. Once you make that 9 decision under the rule at least as proposed, that 10 individual would have rights to appeal that decision. And 11 yet you do not have that right to appeal it if you hire the 12 individual and you verify a background reference currently.
13 I mean if I hire a secretary today, we do
() 14 reference checks on employees. And if there is some comment 15 that comes out of those reference checks, we may make the 16 decision not to hire that particular individual. But then 17 that individual has no right to employability, and that is 18 where it gets very complex.
I 19 MR. MICHELSON: That is strictly a legal gaastion 20 then.
21 MR. COLVIN: Well, I think that it is a legal i
22 question, but it is a cost impact question on the industry.
23 MR. MICHELSON: But not a question of the insider.
24 MR. COLVIN: I do not think that it is a question 25 of safety or increased threat of radiological sabotage.
Heritage Reporting Corporation
() (202) 628-4888 l
f -
L
- 109 v
l'. From that' standpoint, you are correct.
2 MR. CARROLL: What do the construction trade 3 unions sayabout-this?
4 MR. COLVIN: Well,'in most of the cases where you 5 have a collective bargaining agreement in place at a 6 utility, then those bind you anyway. And in fact-in many-7 cases, almost every employee that you have in one way or.
8 another you have a grievance in place that covers those-9 employees. There are some examples like the one that I,just 10 described where that employee or potential employee does not 11 have any rights.
12 MR. CARROLL: My question was not the utility 13 employees or temporary employees, but it was construction
-( ) 14 crafts and their unions.
15 MR. COLVIN: Right. 'If you have a union contract, 16 a collective bargaining agreement.
17 MR. CARROLL: No. I am talking about that I want 18 to bring in some boilermakers.
19 MR. COLVIN: It is a possibility that if you take 20 someone directly out of the craft hall to bring them in, and 21 for that trade that you do not have a collective bargaining 22 agreement in place, then you would either fall under the 23 general requirements of your company policies or state and 24 local laws, which in many cases do provide rights for those 25 individuals.
O Heritage Reporting Corporation (202) 628-4888
C) 1 And we had some examples that we provided a number 2 of years ago wherein the utility companies had their 3 different state and local requirements that provided the 4 basis for covering the rights of employees that were not 5 covered by federal law.
6 So there is a possibility that you can come up 7 with groups of individuals that would not have a right to 8 understand why they were denied employment in those cases.
9 But there is no other case except this and fitness for duty 10 in which we are providing that information to the individual 11 by law. And we do not have to provide it by law in any 12 other situation.
13 What I would like to do now if I can is turn this
() 14 over to Rich Enkeeboll, our senior project manager, and let 15 Rich cover very briefly the guideline elements, and we will 16 try to respond to any questions that you might have on some 17 of the other comments raised.
18 MR. ENKEBOLL: Thank you, Joe. As Joe mentioned, 19 I am Rich Enkeboll, senior project manager for access 20 authorization. In the interest of time, I would ask you to 21 turn to the last side, and we will talk very briefly about 22 that.
23 And then I can put up the next to the last or the 24 third from the last slide, which will give you a listing of 25 the things that are in the access authorization guidelines, Heritage Reporting Corporation f]
(202) 628-4888
i.
. . 111 1 and you can try to use me to get to the industry 2 representatives to answer the questions which you still have 3 remaining, and I will not belabor the details of the access 4 authorization program.
5 (Slides shown.)
6' MR. ENKEBOLL: As Joe mentioned, the licensee is 7 responsible to grant unescorted access. He does that 8 through a very comprehensive process, and that is the 9 process that 3 s included in the guidelines. The judgment
- 10. that he uses is based entirely on the information provided 11 in that collection process. He does not take any particular 12 feature and throw it out. He uses that as an entire package 13 to decide whether that person should have unescorted access.
() 14 We believe that this process provides a reasonable 15 assurance that that person is trustworthy. Jmd the 16 follow-on after he is granted unescorted access is the 17 continual behavioral observation program to make sure that 18 he continues to be trustworthy.
19 The transferability was discussed earlier. We 20 think that all of the utilities will in fact follow this l 21 program. Most of them are very anxious to get on this 22 program, but they are not in a position to implement this 23 right now, because they do not know exactly what the NRC is 24 going to do. They do not want to implement it and have the-25 NRC come out with something new. So they are ready, willing l
'( ) Heritage Reporting Corporation (202) 628-4888 l
l
y.- 112 t
1 and able to do this whole program, but they are not going to 2 until they get the final resolution through this rule making 3 process.
4 The-audit program, the last four pages in the 5 book, says this is how you audit a contractor or a vendor.
6 That audit program will be used for each utility in the 7 first year after the implementation of this process to 8 determine whether in fact it is effective and whether their 9 process does determine that the people who are selected are 10 trustworthy and will continue to be trustworthy. It is not 11 just a process check. They will check the kinds of 12 questions that you are asking.
13 Will they go back and if someone has failed or D)
(_ 14 been thrown out of unescorted access, will they review what 15 got him in there first place and plug that hole? The answer 16 to that question is yes. Those are the tenants of the 17 program.
18 MR. MICHELSON: This slide is what an access 19 authorization program for the future would look like under 20 the guidelines.
21 MR. ENKEBOLL: Correct. Yes, sir.
22 MR. MICHELSON: It does not deal with what we 23 necessarily have today or had a year ago or so.
24 MR. ENKEBOLL: That is correct. Yes, sir.
25 MR. MICHELSON: Are you the one to ask about
() Heritage Reporting Corporation (202) 628-4888
113
'O 1 grandfathering or do you want to leave that for someone.else 2 later?
3 MR. ENKEBOLL: Let me put up this next slide.
4 MR. MICHELSON: Pertinent to this. This is what 5 we are proposing and not what we have.
6 MR. ENKEBOLL: I am the right guy to ask the 7 question on grandfathering, because I am going to pass it on 8 to the guys in the back row who in fact do this. But in 9 fact grandfathering takes into consideration all of the 10 things that were in this program. A licensee does not just 11' automatically say that he worked for me, therefore I do not 12 have to look any further.
13 He is going to take this program when he
() 14 implements it and he is going to say I think that our 15 program has been a little short, and I am going beef this up 16 or that up or something else.
17 MR. MICHELSON: But there is no requirement to do 18 that.
19 MR. ENKEBOLL: We are not going to require that 20 through the rule process.
21 MR. MICHELSON: But there is no requiremen,t to 22 beef it up for people who have been with you for more than a 23 year.
24 MR. ENKEBOLL: There is no requirement. However, 25 licensees understand their responsibility.
Heritage Reporting Corporation O' (202) 628-4888
I em 114 1
')~'
l 1 MR. MICHELSON: You would not object then I guess 2 if we made certain things that go a little further back, at I
3 least like behavioral observation. That somebody observed 4 him for at least three years or alternatively do a 5 background check and so forth. If people have been doing 6 all of this, then there would be no objection to the time 7 frame being changed a little bit, if you really have been 8 doing all of this.
9 Because there is no cost element. You already 10 have it on your records. However if you have not don' any 11 of this, it could get fairly expensive to go backfit this to 12 every person that you have got.
13 MR. ENKEBOLL: That is correct. I do not think
() 14 that we want to buy into any greater time frame. But if 15 that was laid on the industry, I am sure that it would 16 be --
17 MR. MICHELSON: In your judgment from your 18 experience and so forth, do you think that one year of 19 behavioral observation is adequate for an employee who has 20 been over one year, let's say a year and a half or something 21 like that, and for which you did no background check and no 22 psychological testing, would a year and a half of 23 observation be equivalent to those other two elements?
24 MR. ENKEBOLL: I would make that judgment based on 25 how closely I worked with that incividual. If I did not Heritage Reporting Corporation
(' ) (202) 628-4888
i l
115 v
1 work with him closely, I might say that a year is not l 2 enough, I really do not know this guy. But I think that a 3 year is enough if I have worked with him every day and I 4 know what he is doing. Yes, sir, a year is enough.
5 MR. MICHELSON: I think that is true, but we are 6 trying to write rules that can be generally applied.
7 Clearly some people you know within six months even.
8 MR. ENKEBOLL: Yes, sir. But we do not want to 9 write a rule that puts a burden on the industry and does not 10 give the industry credit for having managers who have been 11 doing this for many years. And I do not think that we have 12 to put a number in there that says you do not know what you 13 are doing, manager, we have to have three years rather than
() 14 one year.
15 MR. MICHELSON: Well, I think rather you would say 16 it this way. That if you have not done a background check 17 and you have not done psychological testing, then you must 18 have had a behavioral observation for at least three years 19 for instance.
20 MR. ENKEBOLL: You could say that, because 21 everybody has had a background.
22 MR. MICHELSON: And if everybody has done this, 23 then there is no cost or anything, and everybody is happy 24 and we kind of now what frame we are talking about.
25 MR. ENKEBOLL: Yes, sir.
Heritage Reporting Corporation
(~} --
(202) 628-4888
i 116
~O l'~ )GR. MICHELSON: But right now we do not. Right 2- now I have to say I do not know. I have not done a.
3 background check, and he has only been here a year, I have 4 not done a background' check, and I have not done a
- 5 psychological test, but I have observed his behavior for one Ei year. That is all that I done. As a minimum that is all
'7 you can say.
8 MR. ENKEBOLL: Jerry, or John, or Joe,. do you.want 9 to comment on that?
10 MR. COLVIN: Let me make one comment. The 11 decision as to hov long you could go and grandfather and so 12 on is based on a. lot of discussion with the people in NSSS 13 and the rest of the' staff as to what you need in their
() 14 opinion to provide this reasonable assurance or high 15 assurance that the individual is trustworthy. I do not 16 think that Rich or certainly I am not capable of telling you 17 that one year is okay or one and a half is not okay.
18 I think that that is a. decision that each of the 19 companies has to make. But when you get into the other 20 things that you are doing and the presumption that the 21 employees that you have now are trustworthy or otherwise 22 they would not currently have unescorted access, that 23 coupled with the fact that the programs have been inspected 24 and are continuing to be inspected by NRC staff, I think 25 that when you provide that and look at the total picture-Heritage Reporting Corporation
({} (202) 628-4888 l
g3 117 ij -
1 that we are comfortable and I believe that the staff is j l
2 comfortable with that approach.
3 Perhaps John Ross could give us something.
4 MR. ROSS: I am John Ross from Baltimore Electric, 5 and I am one of the advisors to NUMARC. I think that we are 6 overlooking one key factor. And that is that we have 7 supervisors that are trained in CBO who have all had four, 8 six, eight, ten hours of training on what to look for. I 9 think that that helps us on the one year thing. I would 10 probably agree with you if we had untrained supervisors that 11 we ought to consider a three year time.
12 MR. MICHELSON: Have they all been trained more 13 than a year ago for this purpose?
O)
( 14 MR . ROSS: Sir?
15 MR. MICHELSON: Were they all trained more than a 16 year ago for this purpose, in other words one year of l 17 trained observation? If they were trained yesterday, you do 18 not have a year of trained observation as with a person just 19 365 days back.
20 MR. ROSS: That is correct. But most of our 21 supervisors have been there some five, six or seven years.
22 MR. CARROLL: Where did the training occur though?
23 MR. ROSS: Sir?
24 DR. REMICK: Carl's question is when did the 25 training occur?
() Heritage Reporting Corporation (202) 628-4888
118
~
l' MR. ROSS: I do not have those kind of statistics.
2 But if you would like, I can get them for you.
~3 MR. MICHELSON: I would be surprised if it was six 4 or seven years ago. Although we have talked about this 5 problem for fourteen years now, you have not been really 6 getting to it until the last couple of years.
l 7 MR. ROSS: After awhile I will give you some other 8 statistics to show you who has implemented 18.17, 3.3, and 9 the NUMARC.
10 MR. MICHELSON: Would you have a problem if 11 everybody was cleared at least through those standards or at 12 least to those standards without the one year 13 grandfathering?
() 14 MR. COLVIN: I cannot say whether we would have a 15 problem or not, Mr. Michelson. The real question is whether 16 that is something that is necessary to provide this 17 reasonable assurance. And I think that we have to defer to 18 the staff in that area. Not having a full understanding of 19 what each licensee does, we cannot make a decision as to the 20 potential impact of that. And that is something that we 21 probably want to discuss with the staff.
22 MR. MICHELSON: I think that the staff should at 23 some time, maybe the next time that we have a meeting, make 24 a presentation as to the basis to believe that one year of 25 trained psychological observation is adequate in lieu of any Heritage Reporting Corporation (202) 628-4888 1
- i. -4 119 L E background check and any psychological testing which is what L 1 l- 2 the rule presently allows. If they can provide that basis, L.
3 then I think that we could buy it.
4 DR. REMICK: I wonder if the people here from PP&L
! 5 and Baltimore Gas & Electric might hazard a guess as to what l
6 fraction of the people who might be. grandfathered under this 7 rule would have just slightly over a year experience on 8 unescorted access.
9 What percentage of the employees, would you hazard 10 a guess?
11 MR. ROSS: I would think that at Baltimore Gas &
12 Electric that those people would probably be about 13 80 percent.
() 14 DR. REMICK: That is 80 percent with just a little 15 over a year?
l 16 MR. ROSS: That have more than a year.
1 l 17 DR. REMICK: That have more than a year, okay.
1 18 MR. MICHELSON: If it is less than a year, then it 19 is no problem. They get the full blown treatment. It is 20 not clear in the document by the way exactly when 21 grandfathering starts. In fact it is very unclear.
22 Whenever the amendment comes along is when grandfathering 23 starts. And I do not know when that is, or whether every 24 time you amend it that you get a new grandfathering, I do 25 not know.
() Heritage Reporting Corporation (202) 628-4888
J 120 1 DR. REMICK: How long has industry in general been 2 doing observation training? You have had the fitness for 3 duty requirement there to do that. It is the same 4 observation basically. There might be a few different 5 characteristics that one would look at.
6 MR. COLVIN: Dr. Remick, that is a good point. On 7 the fitness for duty side, we had in place continued 8 behavioral observation, the EI guidelines in the programs, 9 two of those guidelines by January 1, 1985. So that does fi 10 not menn that every supervisor had been trained by that 11 time, but those programs were in place and that was 12 backfitted in the sense of looking retrospectively.
13 So those programs have been in place for some
. ( )- 14 time. I guess what we cannot assure ourselves of or do not 15 have knowledge of is the fact as to how many people of the 16 total population have been trained and also the number of 17 employees that have gone through that.
18 DR. REMICK: Please continue.
19 MR. ENKEBOLL: This is the basic list of the 20 things that the utility goes through to determine the 21 trustworthiness of the individual. I think that we had some 22 questions on military service. The r ilities treat military 23 service as another job. Therefore we say five years like 24 other employment is all you need to check.
25 Now listening to some of your concerns, we might Heritage Reporting Corporation
( )' (202) 628-4888
121 have to go.back and check anybody who was in the military.
1-2 I think that we would have to go even further than that if .
I 3 we were'to'do more than five years. Because if an l 4- individual lies and tells you that he was not in the 5 military when'in fact he was and you did-not check that, you 6 would have a problem.
7 So you have to check every single employment 8 applicant for his military record if you go beyond what we 9 are taiking about here. And we thought that that was rather 10 excessive. We also like to think that if we are only 11- interested in five years for other employment what is 12 special about the military employment.
13 If somebody had a problem in Vietnam in 1973, is
() 14 that going to keep him from having unescorted access. Do we 15 really need to go dig out that far. We do not think so.
16 And that is why we had limited it to the five year period.
17 -MR. CARROLL: Well, as I understood the staff, it 18 had more to do with the criminal. history. 'If someone during 19 his milite.ry service ten years ago was convicted of a crime, 20 it does not end up in the FBI files, and you do not get that 21 out of there with a FBI fingerprint search.
22 MR. ENKEBOLL: If it is not in the FBI file, you 23 would not get it out, that is correct. But in order to get 24 that out, and we thought for awhile that the military was 25 providing that information to the FBI, and I understand as
() Heritage Reporting Corporation (202) 628-4888
)
-v 122 We do not have 1 you are saying it is an informal agreement.
2- the information of whether it is in there or not.
3 But in order to do that, and if a person has a 4' ' criminal record in the military, they will have served their 5 time. Now at what point do you take a person who has been 6 in prison and has given his time, has he not paid for that 7' crime or is that crime going to follow him for the rest of 8 his life.
9 MR. CARROLL: I think that if I am going to 10 evaluate him as a potential employee, that I would at least 11 want to be aware that he has committed a crime.
12 MR. ENKEBOLL: We would have to check everybody.
13 We cannot check just the people who say I was in the
() 14 military, because that is the only way that we are going to 15 find out, and that puts quite a heavy burden on us.
16 HR. WARD: I do not understand it. I thought that 17 the military service issue was coupled with the criminal 18 history and not the employment history. The criminal 19 history goes back to beyond five years.
20 MR. ENKEBOLL: That is correct.
21 MR. WARD: So for that reason to cover that for 22 someone in the military, the military service record has to 23 go back beyond five years.
24 MR. ENKEBOLL: That is correct. The point that I 25 am making is though that you cannot just ask the question
() Heritage Reporting Corporation (202) 628-4888
123 1 has he ever been in the military and just send out requests 2 for DB-214s for those people. Because the guy who had a 3 criminal record thirty years ago is going to lie to you.
4 Because you are going to take that into consideration and 5 you are not going to give him a job.
6 So he says I have never been in the military. You 7 cannot believe him. You are going to have to request a l
8 DB-214. So everybody who comes into my employment office, I 9 request a DB-214 no matter what they say on their form.
10 Otherwise I am going to miss those people. And we are 11 saying that that is beyond what we need, that we have not 12 been doing that. And we have been successful for a lot of 13 years in determining people's trustworthiness, and why-() 14 should we start going beyond five years now.
15 MR. WARD: Now you have told us why it is hard to 16 do it, but you have not told us why it is not necessary to 17 do it.
18 I meaa how do you check on the criminal history of 19 someone who has been in the military for a significant 20 amount of time?
21 MR. ENKEBOLL: You do not unless you go get a 22 DB-214 and have it recorded on that DB-214. That is 23 correct.
24 MR. WARD: So there is a gap there. You have more 25 information on someone who has not been in the military than O Heritage Reporting Corporation (202) 628-4888
124
( 1 you have va someone who has been in the military.
2 MR. MICHELSON: As far as' criminal.
3 MR. WARD: As far as criminal.
4 MR. ENKEBOLL: No, sir. You are going to have the 5 same amount. Because if they have done anything in the 6 civilian world, it is going to be in their civilian criminal 7 history. If they have done it in the military, you are not 8- going to know it.
9 MR. MICHELSON: That is also on the assumption 10 that he is also a criminal in his civil life as well.
11 MR. ENKEBOLL: Yes, sir.
12 MR. MICHELSON: Presently I guess that you do not 13 go back and ask for military history then?
() 14 MR. ENKEBOLL: You ask for military history.
15 MR. MICHELSON: Presently I am talking about.
16 MR. ENKEBOLL: You ask an individual whether he 17 has been in the service. And if it has been within the last 18 five years.
19 MR. MICHELSON: Presently you are asking about the 20 last five years.
21 MR. ENKEBOLL: The last five years. Now some 22 utilities ask forever, but we are not listing that as a 23 requirement.
24 DR. REMICK: Mr. Ross.
25 MR. ROSS: Let me say one thing. One thing as far Heritage Reporting Corporation (202) 628-4888
125 V
1 as the military going back forever is that there are a lot 2 of military crimes that are not transferrable into the 3 civilian sense, sleeping on post and a few things like this 4 that are going to show up in there. If a person commits a 5 rape or something, that is going to wind up in a civil case 6 -with something to be reported to the FBI, particularly 7 anything that they are doing off station. Those would be in 8 the civilian justice system anyway.
9 MR. MICHELSON: If he sabotaged equipment, that 10 would be in his military record too, and you might be 11 interested in knowing about that.
12 MR. ROSS: That is correct.
13 MR. MICHELSON: But you would not if you did not
() 14 ask for it.
15 MR. ROSS: That is right, but we do not go back 16 for the life of a person.
17 MR. MICHELSON: How on the grandfathering 18 provision, there is nothing said that a given utility who 19 does not ask for military service will still not ask for it 20 for people who have been there for more than a year.
21 MR. ROSS: That is correct.
22 MR. MICHELSON: I think that that is kind of 23 difficult to comprehend. At least five years I can 24 comprehend, but you will not even have that degree of 25 information for those people.
O Heritage Reporting Corporation (202) 628-4888
t.
- g. 126 MR. ROSS: That is correct, sir.
2 But let's put.
2 some faith in the CBO training program which we have reason 3 to believe, and I cannot give you statistics today, that 4 most of these supervisors have been trained for four or five 5 -years to look for this type of stuff. And if the guy has 6 been there for three to five years, he hopefully has been 7 observing this because he gets an anneal retraining session 8 on the CBO.
9 MR. MICHELSON: Well, the basic argument is of 10 course that we would like to have a multiple facet attack to 11 this problem, and not just a single element attack. That is 12 why the three basic elements were defined. They were just 13 going to ignore those in this case for more than one year if.
() 14 they were not in the utilities program to begin with.
15 MR. ROSS: Yes, sir. But we do not consider that 16 one of those things that he has on the board is an 17 individual thing by itself.
18 MR. MICHELSON: That is right. That is my 19 argument.
20 MR. ROSS: We collectively look at all seven of 21 them.
22 MR. MICHELSON: Yes. I think that all seven of 23 these are very important, and that we ought to have some 24 knowledge of them beyond one year.
25 MF, ENKEBOLL: Quite frequently the one below
() Heritage Reporting Corporation (202) 628-4888
127 O '1 military service, character and reputation, that is where 2 you check on sources. And frequently that will lead you to 3 the kinds of questions that you are now asking. It may or 4 may not bring it out. But when you do a background 5 investigation and go out and check the neighborhoods and 6 check the sources as the military does, you find out most of 7 what is going on in that person's background.
8 Are there any other questions?
9 DR. REMICK: Yes. Has NUMARC given any thought to 10 at least in the initial audit of perhaps having one group of 11 individuals do that for all utilities, so that you have some 12 common standards?
13 MR. ENKEBOLL: Yes, sir, we have given it some
() 14 thought, but we do not now have an official program.
15 Joe, you might want to comment on that.
16 MR. COLVIN: Let me comment on that. In the 17 procurement area to digress a second, we are working with 18 two organizations that are now merged, the CASE and NSQAC 19 are the acronyms, now into an organization that is 20 temporarily titled the Nuclear Procurement Issues Council, 21 which is an utility based audit function for vendors and 22 suppliers that really consists of a sharing of resources 23 between utilities and the sharing of audit results. All of 24 the utilities have agreed to support that.
25 Now there is a meeting in October of this
() Heritage Reporting Corporation (202) 628-4888
q 128
%J 1 organization. It la the first organizational meeting. And 2 that is one of the topics on the acenda, in effect a 3 combined third party audit that could be utilized by 4 utilities to go do the vendor audits including audits of 5 organizations that have people on unescorted access like 1
i 6 INPO, where they only get audited by one organization, and 7 then the results of that are being used.
8 So we are trying to look at methods where we can 9 do that, and that is basically the current status at this 10 time.
11 DR. REMICK: That bit of being audited can be 12 important. Just purely by chance I have some colleagues 13 that run a small company which has been in existence for
() 14 some years. And they make visconometers to measure the 15 viscosity of oil. And they get audited by a number of 16 nuclear utilities independently. And for a small company 17 with just basically two technical people being tied up a 18 couple of days by different people coming in doing the same 19 thing.
20 They are seriously considering whether they want 21 to sell to the nuclear industry. They feel that it is just 22 not worth their time. There is no way to recover that and 23 so forth. So it can be important to some companies.
24 MR. ENKEBOLL: We are sensitive to the small 25 contractor who has that problem. We are trying to do the
() Heritage Reporting Corporation (202) 628-4888
( same thing with the fitness for duty arena.
1.
2 MR. MICHELSON: Just for clarification on these 3 elements on your slide, the criminal history I think you 4 said goes back however far.
5 HR. ENKEBOLL: Correct.
c 6 MR. MICHELSON: Does education also go back 7 however far?
8 MR. ENKEBOLL: The education is for the last five 9 years, unless he were pursuing a post-high school education 10 leading to a degree, and then that goes back for however 11 long.
12 MR. MICHELSON: So if he did not pursue anything 13 like that in five years, you do not go back and check the
() 14 education history, is that what you are saying?
15 MR. ENKEBOLL: That is correct. Yes, sir, that is 16 correct.
17 MR. MICHELSON: In the case of your more 18 professional types where you are worried a little more about 19 their competence, do you go back and check to see if they 20 really did get the degree that they said they got?
21 MR. FNKEBOLL: Yes, sir.
22 MR. MICHELSON: Even if it was ten years ago?
23 MR. ENKEBOLL: Yes, sir.
24 MR. MICHELSON: But not if he just said that he 25 was a high school graduate?
O Heritage Reporting Corporation (202) 628-4888
130 MR. ENKEBOLL:
1- Correct.
2 MR. MICHELSON: Okay.
3 MR. ENKEBOLL: But if he is hired on because he 4 has a degree. _
5 MR. MICHELSON: Is this generally practiced do you 6 know, or is this just your particular knowledge?
7 MR. ENKEBOLL: This is what is required by the 8 guidelines.
9 MR. MICHELSON: The education history then does go 10 all the way back except for high school graduates by.the 11 guideline. I did not look be.ck at it.
12 MR. ENKEBOLL: Check the last five years. And if 13 he has done any post-high school education leading to a 1(]) 14 degree, check that no matter how far.
15 MR. MICHELSON: No matter how far back.
16 Now are those the only two elements that you check 17 any further back in time on, in other words all of the way 18 back to the beginning?
19 MR. ENKEBOLL: Yes, sir, although there might be 20 some credit checks. I do not know how far.
21 MR. MICHELSON: But normally you do not ask for 22 employment thirty years ago if a guy is older or twenty 23 years ago or whatever?
24 MR. ENKEBOLL: That is correct, we do not.
25 MR. MICHELSON: And none of these others except O Heritage Reporting Corporation (202) 628-4888
E f 131 for education and' criminal?
11' 2 MR. ENKEBOLL: Yes, sir.
. 3 MR. MICHELSON: Okay. Thank you.
'4 DR. REMICK: If I recall, I think that_the Q 5- clearance goes back fifteen years.
6 MR. MICHELSON: Something like that, except.for 7 education.
8 DR. REMICK: I think residence and employment are 9 the ones that go-back fifteen years.
10 Are there any further questions from the 11 subcommittee?
12 (No response.)
13 DR. REMICK: Does that conclude the NUMARC 14 presentation?
'15 MR. COLVIN: Yes, sir, that concludes our 16 presentation. We would be happy to answer any questions.
17 MR. MICHELSON: I thought that you were going to 18 tell us a little bit more about your position on access 19 during shutdown.
20 were you going to tell us anything more?
21 MR. COLVIN: I would be happy to try to respond to 22 that. This has been an issue that has been discussed on the 23 concept between the' staff and the industry for many years, 24 the concept of devitalization or establishing vital islands 25 within a non-vitalized area.
Heritage Reporting Corporation
' f~ '
(202) 628-4888
z 132 0 1 In our discussions with the staff in preparation 7 of the guideline, it was concluded that there might be l
3 benefit to provide a process by which you could devitalize 4 an area during a major outage or event. For example if you 5 were going to go into steam generator replacement and 6 containment, you might devitalize the containment. And then 7 _you ask yourself if you want to do that, what steps would 8 you have to put in place.
9 And what you see in the guideline are the elements J that were already discussed by the staff. In the guideline 11 is the result of those discussions. I think that in 12 practicality when you ask the question of how many utilities 13 have actually taken advantage or have done that, I think
() 14 that the number is very small at this point in time.
15 I know of one utility in the steam generator 16 replacement that modified its access authorization 17 requirements.
s To be honest, I am not sure whether they 18 exempted all of the current requirements or whether they 19 just modified a few of them.
20 But this was put in there to allow a utility to 21 put in place some reasonable requirements when there was not 22 a need for the deta d led requirements contained within the 23 guidelines. I think because of the questions raised by the 24 . staff and that were raised by CRGR and now that the staff is 25 wrestling with that I think that we need to have some
() Heritage Reporting Corporation (202) 628-4888
133 l' further discussions on this. -
2 MR. MICHELSON: Clearly there are cases where you 3 could devitalize it with reasonable certainty. There are 4 other areas that would be very difficult to devitalize 5 because-of the inability to reverify that everything is 6 okay. And maybe with that thought in mind that it does not 7 look quite so bad as it did on the first reading.
8 DR. REMICK: Is it my understanding that the 9 industry position is you would prefer a policy statement, 10 and that if there is a policy statement with the industry 11 commitment that they would modify their physical security 12 plans to incorporate the NUMARC guidelines in that, that if 13 the Commission comes out with a rule that that commitment 1
() 14 does not stand but it is moct anyhow I guess because you L 15 would have to meet the requirements.
16 But if they do come out with a rule that you would 17 like to see the regulatory guide endorse as closely the 18 NUMARC guidelines as possible. And there are several of 19 those I guess. The appeal process you feel very strongly 20 about.
21 MR. COLVIN: Yes, sir. Let me clarify the policy 22 statement versus the rule issue. I think that up until-the 23 point that the Commission made the decision to go forth with 24 the rule that we had a very well established position that 25 we would prefer the policy statement and utilizing the s
O Heritage Reporting Corporation (202) 628-4888 E____________
134 g)
\
)
1 commitment to implement the guidelines and amend the 2 security plan.
3 Once that decision has been made, we have now 4 gotten to working with the staff to get the guideline 5 incorporated or a reference in a reg guide and proceed 6 onward. To be honest, we have not revisited the question of 7 should there be a rule or not a rule, because that decision 8 is behind us. So we would like to move on. Should the 9 Commission desire an opinion on that, we would be happy to 10 provide it, but I think that that is already behind us.
11 DR. REMICK: Are there any further questions? j 12 MR. MICHELSON: Yes, I have got a couple of 13 questions.
/~ N
() 14 On the grandfathering, what is your understanding 15 of your grandfathering provision on page 13 which says 16 within 365 days prior to the date of amended security plan 17 approval, what does that mean?
18 MR. COLVIN: The guideline was written to commit 19 to the NRC to amend the security plans to incorporate the 20 elements of the guideline. And once those plans were 21 amended by a letter to the NRC, then that would change the 22 licensing basis of the plan. So that was the 365 days from 23 that date.
24 MR. MICHELSON: It is the date of the licensee's 25 letter to the NRC saying what they are doing?
Heritage Reporting Corporation f]
(202) 628-4888
es 135 i) #'
1 MR. COLVIN: Yes, sir. In the case of the rule, 2 it would coincide with the data implementation date required I 1
3 in the rule and 365 days from that date is our !
4 understanding.
5 MR. MICHELSON: And that would be one that I could 6 tie down. Your other one I am not sure what it means.
7 Because then the NRC comes back and says change this or 8 that, and I do not know when the clock really starts.
9 MR. COLVIN: I think that that is a point that 10 obviously needs clarification within implementing guidance 11 to the rttle.
12 MR. MICHELSON: Now I have one question for the 13 staff yet.
() 14 Apparently the staff took no exception to the 35 devitalization of vital areas, is that correct?
16 MR. MORRIS: In our package that we provided to 17 you, we did not. Because the question has come up in the 18 CRGR meeting, and there is uncertainty in our minds about 19 what these start-up procedures would be like to give you the 20 assurance that you need. And we do not quite know what they 21 would be like. So we think that we need to look into that 22 further.
23 MR. MICHELSON: So we may see further words on 24 it?
25 MR. MORRIS: Yes. And I would hope that maybe t
Heritage Reporting Corporation k' ]' (202) 628-4888
l
- . 136 1 NUMARC will give'some thought to what they think those 2 procedures are like in some specific cases. And if they 3 could provide us some insights, it would help us greatly in 4 figuring out how we could answer questions about what that 5 procedure would be.
6 MR. MICHELSON: I can see how devitalization might 7 work, but I was wondering to what extent would the NRC 8 review the utilities' proposal to devitalize for the purpose 9 of carrying out a certain operation, would that be reviewed 10 or is it just up to the local resident inspector to decide 11 if it is okay?
12 MR. BUSH: I believe that that would he up to the 13- inspector on a case basis if the licensee indicated that
() 14 they were going to go into an outage and that this is how 15 they plan on bringing people in and that sort of thing.
16 MR. MICHELSON: There is nothing in the guidelines 17 that led me to believe that there really was anything that 18 was necessarily reviewed or whatever. It is jus that there 19 does have to be appropriate procedures and that they have to 20 be followed, whatever that means. And I did not know if i
l 21 those procedures were reviewed by anybody or had any 22 limitations on them.
23 I think that it is a sufficiently critical thing 24 that somebody ought to take a look at whenever you 25 devitalize a vital zone, and not just the local resident O Heritage Reporting Corporation (202) 628-4888
L 137 L($).
-1 inspector.necessarily who may not even be well qualified to 2 make that particular judgment.
3 MR.. CARROLL: I would change my terminology from 4 just' start-up procedures to inspection and start-up 5 procedures. . Inspection is a very important element of that.
6 MR. MORRIS: The guidelines include a visual 7 inspection plus the procedures. So I do not know whether 8 that meets what you had intended or not.
9 MR. CARROLL: 'It does.
10 MR. MICHELSON: A thorough visual inspection.
11 DR. REMICK: How about very thorough.
12 MR. MICHELSON: No, it just says thorough.
13 DR. REMICK: Are there any further
() 14 questions?
15 MR. CARROLL: Joe, I guess one philosophical 16 question.
17 Is it not better from the industry's point of view 18 in terms of dealing with unions to be able to say hey, we 19 are doing this because there is a rule here. I guess having 20 sat across the bargaining table for many years that whenever 21 I was saying I am conforming to a reg guide or my perception 22 of what the NRC wants, the union turns it around on me and 23 says a. reg guide is not mandatory, let's negotiate something 24 and im Y9u prtaent it to the NRC as an alternative that 25 makes us happy.
() Heritage Reporting Corporation (202) 628-4888
l I
73 138 l N-] Is the industry that firm that they really want to l 1 I
i 2 see something as squishy as a policy statement? {
3 MR. COLVIN: Well, you know we have never 1
4 addressed that to be quite honest with the industry .
1 5 leadership as to whether a policy statement brings with it 6 the problems on one side but it also alleviates problems on 7 the other side, and I am not sure where the balance comes 8 out to be honest.
9 But I think that it is clear that if you only 10 isolate the negotiations with the bargaining unit or with 11 the state or others that when you issue a NRC rule that of 12 course federal exemption applies and therefore there is no 13 more argument.
() 14 So if you have those kinds of arguments, that 15 would make it easier on you possibly. But I think that when 16 you look across the whole industry and look at the union 17 involvement and look at where that really creates problems, 18 I am not sure that that is a large issue. That narrow 19 element of it is just a question in your question.
20 DR. REMICK: Okay. We have the open ended 21 question of what the subcommittee action should be. I think 22 that I indicated my preference in the mid-morning. I did 23 not realize that the staff was considering modifications. I 24 was under the impression that the staff was asking for a 25 letter from the committee this month.
(]
b Heritage Reporting Corporation (202) 628-4888
1 l
19 l
i n/
s_
1 In talking with the staff during the break, they 2- likewise do not recommend that the ACRS consider a letter 3 this morning, because they are considering some changes. So 4 if the subcommittee agrees, I would propose that we drop the L 5 presentation next week to the full committee, wait for the l ~
l 6 staff to make any modifications that they wish, and at that 7 time then decide whether that calls for another subcommittee 8 meeting or that we would then if the modifications are not 9 great over what we have heard today take it to the full 10 committee at a later full committee meeting.
11 Is that acceptable to the subcommittee?
12 MR. CARROLL: That sounds reasonable to me.
13 DR. REMICK: All right, fine. I assume that that
() 14 is satisfactory with the staff. I want to thank the staff, 15 and the NUMARC representatives, and the industry 16 representatives'for coming in and helping clarify this. We 17 appreciate it.
18 I am now faced with the embarrassing predicament.
19 We urged the staff to come down and be ready to talk about 20 performance indicators earlier. But it seems like it might 21 be a logical point to break for lunch.
22 Would that give staff here for performance 23 indicators any particular problem if we break for lunch now?
24 (No response.)
25 DR. REMICK: All right. Let us then break for Heritage Reporting Corporation O' .
(202) 628-4888
V !
140 0: .1' lunch and returning at 12:45, and we will take up the 2 discussion on the performance indicators.
3- (Whereupon, at 11:45 a.m., the subcommittee was 4' recessed, to reconvene at 12:45 p.m., this same day.)
5 6
-7 8
9 10-11-12 13 O 14 15 16 17 18 19 20 21 22 23 24 25 O Heritage Reporting Corporation (202) 628-4888
- t. . - _ . - - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ .
141 Q- 1 AEIEEHQ2H EEEE12H 2 (12: 45 p.m. )
3 DR. REMICK: Good afternoon, ladies and gentlemen.
4 We will continue the meeting of the Subcommittee on Human 5 Factors.
6 The next topic we have is the' briefing by the NRC 7- staff on performance indicators. There was no handout for 8 the subcommittee in advance on this, but you do have a 9 handout of the viewgraphs.
10 I assume, Dr. Ryan, you are going to take the 11 lead? Okay. do you want to proceed, please?
12 (Slides being shown) 13 DR. RYAN: My name is Tom Ryan and I am with the.
R( ) 14 Human Factors Branch within the Office.of Research. And my 15 purpose here this afternoon is to introduce the main 16 presentations which will be made by Mr. Carl Johnson and Dr.
17 Brasmia Lois, also of the Human Factors Branch.
18 And on behalf of the branch, I would like to 19 personally thank the subcommittee for allowing us to come 20 down and to appear before you and talk to you a little bit 21 about our performance indicator research.
22 The handouts which you have in front of you are 23 complete and they cover both my introductory remarks and the 24 two main presentations.
25 The purpose of the briefing is first of all to Heritage Reporting Corporation (202) 628-4888
_. _ _ _ _ _ _ _ _ _ _ _ _ . 1
q 142 Y-)
1 familiarize the subcommittee with risk base and programmatic 2 performance indicators research that we have completed, and 3 that is ongoing, and secondly to seek the subcommittee's 4 comments on the research that has been completed and your 5 suggestions regarding some of what might be our future 6 activities.
7 In August, you were sent three documents as sort 8 of background to all of this. The first was a report titled 9 validation of risk-based performance indicators. And that 10 was preparatory to the presentation that Carl Johnson is 11 going to make on risk-based indicators.
12 The other two documents pertain to the 13 programmatic performance indicators. The first was titled
() 14 the development and evaluation of programmatic performance 15 indicators associated with maintenance at nuclear power 16 plants, and the second was titled management, organization 17 and safety in nuclear power plants. And these are 18 preparatory to the presentation that Dr. Erasmia Lois will 19 make a little bit later.
20 The purpose of the research. Our research is 21 intended to complement and build on the work that has 22 already been done within AEOD, and to the extent we can, on 23 work that has been done in the industry.
24 At the request of the Commissioners and AEOD, we 25 basically have focused on two general areas, but two other
() Heritage Reporting Corporation (202) 628-4888
p, 143
.V. 1 sort of sub-areas that also file under what we call i
2 programmatic performance indicators.
! 3 The first are the risk-based indicators, and we 4 have been focusing on safety system function trends.
l 5 In the programmatic performance indicator area, we 6 are focused on technical support programs, as you see up 7 here maintenance, and we are getting into the training area.
8 We have also been working in the area of what we 9 call-top-down indicators, organizational type indicators.
10 And we also have started some research on sort of a bottom 11 up approach. That is how do we take component indicators 12 and make some kind of realistic composites out of them.
13 Thus far, in the approximately year and a half we
() 14 have been involved in this work, the work has directly 15 involved members of the Human Factors Branch staff. And we 16 have had people from AEOD and NRR who have been actively 17 advising us in this research.
18 In addition to that, we have had involved a total 19 of eight contractor entities: the Brookhaven National 20 Laboratory; two commercial contractors (Science Applications 21 International Corporation and Communications Technology 22 Associates from McLean, Virginia); the Human Affairs 23 Research Center of Battelle Memorial Institute in Seattle; 24 and four universities (the University of Minnesota, the 25 University of Maryland, Wayne State University in Detroit, O Heritage Reporting Corporation (202) 628-4888
\
' 144
~
l 1 Michigan and American University here in the District).
2 Both of the presentations this afternoon are going 3 to focus around five basic issues or questions to the degree 4 that the individual research activities have progressed to
- 5 those points.
1 6 The first is a conceptual framework, a means by 7 which initial' candidate indicators were identified in the
, 8 various areas; secondly, criteria that were used to screen 9 them, to refine them; thirdly, criterion measures that were 10 .used in attempts to evaluate, validate them; the validation 11 techniques which involve both statistical analyses and 12 engineering analyses; and of course, then, the results that 13 have emerged from those particular evaluations.
() 14 What I would like to do before I turn the meeting 15 over to Carl Johnson is to just take a moment and try to 16 introduce what we mean by a risk-based indicator, the 17 concept itself.
18 And basically this involves two inputs. First of 19 all is the ratio per unit time, and in this case three 20 months, of the total hours that a specific safety system is 21 out of service to the total hours that the plant is 22 critical.
23 The second input is an upper bound established by 24 the staff using existing PRA data, using the formula that 25 you see on the bottom that is familiar to PRA people.
O Heritage Reporting Corporation (202) 628-4888
__a
145 1 Finally, the measure of the risk-based indicator 2 is the relationship between the ratio and the bound and the 3 direction of the movement of the ratio toward the bound, 4 parallel with the bound or away from the bound, when we are 5 trending the indicator.
6 Most of the information that Mr. Johnson will be 7 presenting to you will show these trends.
8 So that is basically what we meen by a risk-based 9 indicator.
10 MR. MICHELSON: What is the anticipated source of 11 the hours the system is out of service? Where do you get 12 those numbers?
13 DR. RYAN: Mr. Johnson is going to get into all
() 14 the details. I just am trying to make sure that we all kind 15 of are on the same sheet of music with regard to what we 16 mean by a risk-based indicator.
17 MR. WARD: You say Carl is going to define them?
18 DR. RYAN: That's right. He is going to go into 19 all the details. I'm just trying to give you sort of a 20 sense of what we mean when we say risk-based indicator.
21 Lastly, I would like to turn to what we mean by 22 programmatic performance indicator.
23 Ideally what we would very much like to be able to 24 do is to relate the attributes of a technical support 25 program or organizational factors, what have you, over here, f,]'
Heritage Reporting Corporation (202) 628-4888
s 146
- ~
1 with a criterion measure.
2 Unfortunately, we don't have that kind of data.
3 So basically in the short term what we are focusing on are 4 these intenaediate outcomes. And they are determined by 5 either through an engineering analysis or what we find in 6 the data source itself. 'For example, in the LER system, 7 sequence coding attributes certain causal factors to these 8 outcomes by relating those to and then attempting to 9 establish a statistical relationship between those outcomes 10 and a criterion measure, is what we today are calling our 11 programmatic performance indicators.
12 So.they are not process indicators. That is the 13 attributes of the programs itself. They simply take on
() 14 their meaning through the relationship of the outcome, 15 whatever it attributed as a causal factor to that outcome, 16 and then trying to relate those outcomes to some criterion 17 measure.
18 MR. WARD: Are we going to hear more about that?
19 DR. RYAN: All details. I am just simply trying 20 to introduce the two concepts.
21 MR. WARD: I didn't understand that.
22 MR. MICHELSON: I don't think I did, either.
23 MR. CARROLL: I don't, either.
24 DR. RYAN: What we would very much like to do in 25 an ideal world would be to go in and take a look at the Heritage Reporting Corporation (202) 628-4888 i
1 7 .147 1 attributes, say, of a maintenance program, and be able to 2' relate those to the far right-hand column, to some measure 3 .that we call a safety measure.
4 We don't have that kind of data available to us, 5 especially for the left-hand column. So basically, we are 6 focusing on intermediate outcomes. And it may take on their 7 definition as a maintenance indicator, for example, in the 8 sense that somebody has attributed to them as a causal 9 factor or by means of an engineering analysis,-that 10 something in maintenance led to this.
11 We may not be sure, since~we are accepting what is 12 in the LER report, for example, or something.
13 But we simply establish the measures on this in
.( ) 14 terms of counts that arc associated with, say, maintenance, 15 for example. It's done through an engineering analysis or 16 simply has been identified from the source of the 17 information.
18 MR. MICHELSON: Is somebody going to define 19 criterion measure later on?
20 DR. RYAN: That is correct.
21 MR. MICHELSON: Okay.
22 DR. RYAN: The other step in the process then is 23 attempting to associate, using a statistical approach, a 24 relationship, a systematic relationship, between the outcome 25 scores and a criterion measure, of which some are
() Heritage Reporting Corporation (202) 628-4888
148 g-)
V 1 illustrated in the right-hand column.
2 MR. MICHELSON: Those are outcomes, is that what 3 you are really saying?
4 DR. RYAN: Well, these are both outcomes. These 5 are intermediate outcomes, final outcomes.
6 MR. MICHELSON: Final outcomes.
7 DR. RYAN: And the intermediate outcomes take 8 their meaning from this right hand, or the left hand column.
9 The other point I would like to make in closing, 10 in my part, is basically the criterion measures we look at 11 are time delay, in the sense that the assumption underlying 12 some of this is that the intermediate outcome occurs and it c 13 has its effect downstream somewhere:,
(G_) 14 So Dr. Lois will be discussing that.
15 So again, I am just trying to sort of introduce 16 the concept.
17 DR. REMICK: And daily power loss, is this 18 something like capacity factor or unavailability or 19 something?
20 DR. RYAN: She is going to operationally define 21 them, Dr. Remick.
22 DR. REMICK: Okay.
23 DR. RYAN: She is going to go through all of this.
24 I'm just trying to sort of lay out the concept.
25 DR. REMICK: We' re waiting with anxious breath, t Heritage Reporting Corporation (202) 628-4888
f 149 1 MR. WARD: You have information on the right hand 2 column.
3 DR. RYAN: Right.
4 MR. WARD: But you're going to relate the left 5 hand celumn to the right hand column.
6 DR. RYAN: Right.
7 MR. WARD: If you have information in the right 8 hand column, and you don't have any related quantifiers --
9 DR. RYAN: That's right. We don't have --
10 MR. WARD: But then, why are you -- so you are 11 giving up on the right hand column, you are going to relate 12 the program elements to the intermediate outcome?
13 DR. RYAN: No. The only thing we are doing ins we
() 14 are saying, let's take the LER system, we go back and say 15 how many ES actuation over a given period of time? We go 16 back and say okay, what might have the cause been? We see 17 the word " maintenance" so we count that one.
18 . Okay. We will accept that. In some cases, some 19 kind of retrospective analysis has been done in looking at 20 what might have been the causal factors.
21 So by simply defining these in terms of this we 22 say okay, we have some data on an outcome that we are 23 associating with maintenance.
24 We have data on these things over here, and this 25 is just illustrative.
l l
(} Heritage Reporting Corporation (202) 628-4888
7
,73 150 )
L'~J 1 MR. WARD: You are going to substitute --
2 DR. RYAN: That's right.
3 MR. WARD: -- intermediate outcomes for the 4 program elements.
5 DR. RYAN: That is exactly right. They are simply 6 being used now and taking on their meaning in terms of the 7 causes that it had been said had led to them. And in many 8 cases, they are maintenance.
9 Now, ideally, as I started out to say, we would 10 very much like to have lots of data on the attributes of 11 maintenance programs, and simply working this left hand 12 column to the right hand column.
13 In the short term we don't have that kind of
() 14 information.
15 So we are saying okay, we are going to focus now 16 in intermediate outcomes as the indicators which simply take 17 on their meaning from the left hand column. Is that -- ?
18 MR. WARD: Yes, I think I'm getting it.
19 DR. RYAN: Okay.
20 If there are no questions from me, I would like to 21 turn it over to Carl Johnson who then is going to start a 22 brief on the risk-based indicators.
23 (Slides being shown) 24 MR. JOHNSON: My name is Carl Johnson, NRC, 25 assigned to the Research office. And I would like to
() Heritage Reporting Corporation (202) 628-4888
e, 151 f L-) i discuss our research on risk-based indicators.
i 2 The work that I am going to be doing was done by, 3 primarily by our two principal investigators who are sitting 4 right behind you, Dr. Azarm and Dr. Vesley from Brookhaven 5 and SAIC.
l l 6 First, I would like to sort of go into how this 7 candidate indicator fits into our scheme of performance 8 indicators.
9 The existing set of performance indicators was 10 selected to represent at least qualitatively things that 11 relate to risk, things that relate to initiating events, 12 unavailability of safety systems, programmatic causes, or 13 potential for causing common cause failures. And when the r '
( 14 existing set of indicators was implemented, then we were 15 asked to continue some work to see if we could come up with 16 some indicators that were more directly related to risk.
17 So the first one that we undertook is to do a 18 better job on unavailability of safety systems.
19 The present indicator, the best, most direct 20 indicator of unavailability of safety systems that we are 21 using now, is safety system failures. And this candidate 22 indicator then would be intended to be a replacement or 23 augmentation of that information.
24 The approach is to use a, instead of an indicator 25 based on complete safety system failures, to use an Heritage Reporting Corporation O' (202) 628-4888
152 1 indicator based on train level or component level data, on 2 unavailability of individual trains.
3 The idea then is to come up with an indicator 4 which would allow more easy trending and that we could 5 compare a magnitude of this indicator against a standard of 6 what you would' expect to see for the particular safety 7 system involved.
8 The safety systems selected were, for PWR is aux 9 feedwater, diesel generators and high pressure injection, 10 and similar systems for BWRs.
11 They were selected on the basis of being important 12 safety systems which fail often enough or which have 13 component or train level failures that fail often enough to
() 14 make it worthwhile tracking them.
15 So the approach is to collect as input data number 16 of train-level failures, loss of function of individual 17 trains and the hours that trains are out of service for 18 maintenance or surveillance testing, work up an estimate of 19 unavailability of individual trains and aggregate this to 20 the system level and compare that with expected values of 21 unavailability.
22 MR. WARD: So for the train failures, do you 23 associate a time with that?
24 MR. JOHNSON: Yes. Can we defer that for two 25 minutes, and I'll show you exactly how we do it. We're
() Heritage Reporting Corporation (202) 628-4888
153
.O 1 going to do it right.here.
2 MR. MARD: All right.
3 MR. JOHNSON: The way this indicator is formulated 4 is like this. The indicator is the fraction of time that a 5 piece of equipment, in this case a train, is out of service, 6 or cannot respond to a demand, either because it is out of 7 service for maintenance or testing or because it is already 8 in a failed condition, hasn't been recognized yet as being 9 failed.
10 MR. MARD: Okay. So if it is found to be failed 11 or fails on demand, you go back and look at-the time, or 1
12 infer a time that it has been out of --
13 MR. JOHNSON: Exactly. So here is the equation l
() 14 that is used in the indicator.
l 15 MR. MICHELSON: Before we get to that, though, how 16 were you defining failure, for instance?
17 MR. JOHNSON: Yes. This is the most difficult 18 thing in plugging in the data. The biggest source of 19 uncertainty is the number of failures.
20 MR. MICHELSON: What even constitutes a failure.
21 MR. JOHNSON: Yes. And so the criteria that --
22 MR. MICHELSON: The other thing I was wondering 23 about, and I just want to leave the questions on the table 24 and you can cover them later, the other question I have is 25 where are you getting the information about fractions of Heritage Reporting Corporation O (202) 628-4888
154
) time, since the LERs may or may not have that information.
1 2 But the sequence coding does not have time.
3 MR. JOHNSON: Right. These data are not currently 4 reported to the NRC. And if I could defer it a little bit, 5 I would like to show you some candidate sources of the 6 information.
7 MR. MICHELSON: Oh, okay. Thank you.
8 MR. JOHNSON: So the way the indicator is 9 formulated is, this is the hours of time that the train was 10 out of service this particular quarter for maintenance or 11 whatever reason. Then this is the number of failures of the 1
l 12 train during this quarter. This is with a loss of function.
13 There are some failures that could occur for example, in an
-( ) 14 instrumentation that would not cause a loss of function of 1
15 the train.
-16 So this is number of times that the train failed 17 or was unable to function during the quarter. And then the 18 time that is assessed to that, the most accurate way would 19 be to take half of the interval between when it was last 20 operable and when the failure was discovered.
21 That would, however, require collecting data on 22 each individual failure. Sort of event-based data. So in 23 order to simplify the data needs, the idea is to simply take 24 half of a surveillance interval as a surrogate for the 25 amount of time between when the thing was last operable and O Heritage Reporting Corporation (202) 628-4888
155 I
1 when it is discovered.
2 MR. MICHELSON: Now, what was the definition of 3 failures? Is that hours?
4 MR. JOHNSON: No , that's number.
5 MR. MICHELSON: Numbers.
6 MR. JOHNSON: Numbers of failures. How many i
7 times, say, did the train turn up, say, ran a surveillance 8 test for example, and found that it did not pass the 9 surveillance test, would not serve the function that it is 10 intended for.
11 MR. MICHELSON: Are hour critical? The total 12 number of hours that you thought you needed the equipment?
13 MR. JOHNSON: That's right. So again, the lll 14 accurate denominator would be the number of hours that the 15 component or the system is needed in the quarter.
16 MR. MICHELSON: Yes. But that isn't always 17 related to criticality. Some systems you need all the time.
18 MR. JOHNSON: That's right. But there are systems 19 that -- the systems that we selected here. The primary, 20 the most restrictive requirements in the tech specs are 21 during operation Modes 1, 2 and 3 and'during shutdown they 22 are allowed much longer, or different allowed outage times.
23 MR. MICHELSON: Now, LPCI/RHR meant only the LPCI 24 mode of RHR?
25 MR. JOHNSON: It means both. It's that system.
Heritage Reporting Corporation 9 (202) 628-4888
156 p, h. ~
1 MR. MICHELSON: Okay.. Then the RHR system is 2 needed when you are critical.
3 MR. JOHNSON: That's true. That's true.
4 So this is a way of simplifying the data needs so 5 you don't have to go in and parse it out, and it is pretty 6 close.
7 Then, the information on train unavailability can 8 be aggregated to a system level by straightforward 9 reliability relationships, either on a fairly simplified 10 basis,Leither as an average train unavailability, or by the 11 product of the two trains or by more accurate methods, 12 depending on how accurate you are interested in'getting the 13 final readout. l
() 14 ,
MR. WARD: Would you take into account whether the l 15 uses for example are 100 percent trains?
16 MR. JOHNSON: Yes. And the example I gave there 17 would be for 100 percent trains.
18 MR. MICHELSON: Now, failures is going to be 19 related to loss of function, then, is that right?
20 MR. JOHNSON: Loss of function.
21 MR. MICHELSON: Of the system.
22 MR. JOHNSON: Of the system.
23 So then we go into what does the indicator mean.
24 Well, since rish is a product of things including 25 unavailability of safety systems, then one indication of O Heritage Reporting Corporation (202) 628-4888
157 l
1 -flagging performance of these selected risk-significant
? systems-is unavailability or unreliability of the system l'
3 worse than the NRC standards for that system.
l 4 DR. REMICK: That assumes there are such things as 5 NRC standards.
6 MR. JOHNSON: That's right. And that is only true 7 for example with aux feedwater. With most of these systems, 8 these standards do not exist, but can be set. They have not 9 been set. That's right.
10 And so the indicator magnitude and trend are both 11 significant from a risk standpoint.
12 Also, I make this hypothesis, namely, that NRC, 13 namely.that this unavailability is related or reflects
() 14 maintenance effectiveness.
15 The agency's definition os maintenance is that 16 maintenance is the aggregate of systematic actions to 17 prevent degradation of or failure of and promptly restore
- 18. the intended function of systems, structures and components.
19 And so therefore, one indication of ineffective 20 maintenance is a high level or a degrading trend in failure 21 rate or unavailability of these selected safety systems.
22 Now, this is not a broad-based sort of indicator 23 like the maintenance effectiveness indicator that the agency 24 is pursuing. This is some supplemental information on these 25 particular risk-significant systems.
Heritage Reporting Corporation (202) 628-4888
158
'l MR. MICHELSON: How do you sort out the operator-2 type errors that cause the loss of function?- Sometimes 3 these occur and they cause loss of function for months, 4 even, because they didn't realize they had done it.
5 Is that still considered maintenance, then?
6 MR. JOHNSON: Yes. In this definition.
7 HR. MICHELSON: It clearly is not maintenance.
8 MR. JOHNSON: It is not maintenance in that it is 9 not the maintenance crew.
10 MR. MICHELSON: Well, it is not maintenance in the 11 sense that there was nothing wrong with the equipment. It
.12 was something wrong with the operator.
13 NR. JOHNSON: That is true.
14 MR. MICHELSON: .He closed the valve when it was 15 supposed to be open.
16 MR. JOHNSON: That's true. But the objective of 17 the maintenance program is supposed to :a to keep up the 18 unavailability, to keep up the availability of these safety 19 systems. And so that is one indication of ineffective 20 maintenance would be the degrading of unavailability of l 21 these safety systems.
t 22 The indicator was developed initially using 23 computer simulations to evaluate trends, the ability of the 24 indicator to recognize trends, to set levels for comparison 25 for looking at the response time of the indicator.
() Heritage Reporting Corporation (202) 628-4888 l
159 O- . '
1 And this is simply an example of one of these 2 thousands of simulations that were done.
3 This particular thing happens to be on the Y 4 axis's unreliability or failure rate of a system versus 5 quarters where the simulator put in a failure rate for a 6 train, for a two-train system, and increased the failure 7 rate in a stepped manner like.this, so this is truth coming 8 along here like this with a dotted line. And the stars are 9 simply the response observed in this particular run.
10 And the point of it all is that you can typically 11 see an increasing trend sconer with this sort of an 12 indication based on train level information, sooner than you 13 can if you wait to simply count system failures.
H( ) 14 Nothing remarkable. It simply illustrates one of 15 the reason the indicator was developed in the first place.
16 MR. MICHELSON: What is that input failure rate 17 again?
18 MR. JOHNSON: This is train failure rate. Along 19 here it was increased to this level.
20 MR. MICHELSON: Now, is that a measured failure 21 rate or that is just a number being used for calculation?
22 MR. T'HNSON: This is typical of failure rates for 23 the kinds of system we use.
24 But this is an input in a computer simply to 25 evaluate response times using --
Heritage Reporting Corporation
(} (202) 628-4888
.. i 160.
1 MR. MICHELSON: But it is a historical. record of
'2 something, isn't it?
3 MR. JOHNSON: This is a computer simulation. This
.4- 'is not a historical-record.
5 They want to find the response rate of the 6 indicator to step changes in the underlying performance
.'7 reliability of the equipment or ramp increases like with 8 aging, things like this.
9 So there are a lot of games played with the 10 computer to see how fast would this kind of an indicator 11 respond to -- how soon would you be able to see it compared-
.12 with instead of using a train level indicator, waiting for a 13- system failure.
( )- 14 MR. MICHELSON: You are saying this whole thing is 15 a simulation?
16 MR.. JOHNSON: Yes. .This.whole thing is a 17 simulation.
18 MR. MICHELSON: There is no real data being 19 exemplified there.
20 MR. JOHNSON: There is no real data but it is put 21 in there in the ballpark of the failure rates that are 22 observed in systems with the sort of distribution expected.
23 MR. MICHELSON: And you just put in an artificial 24 step.
25 MR. JOHNSON: Yes. To observe, what would you
) Heritage Reporting Corporation (202) 628-4988
161 El' .
1 see, if you used the indicator, this candidate indicator, 2 versus if you use the existing indicator, which is waiting 3 until the failure-occurs, the system failure.
4 MR. MICHELSON: So you must have simulated 5 something else besides. I don't know how you did this, but 6 it is interesting.
7 MR. JOHNSON: It is all described in this report 8 that we sent-to you.
9 MR. MICHELSON: Well, yes. A lot of reports are 10 sent to me.
11 MR. WARD: Let's see. Why did the peak values of 12 the observed indicator go se high?
13 MR. JOHNSON: This is two failures here and this 14
-( ) is simply a way of counting what happened here.
15 MR. CARROLL: As opposed to a train failure.
16 MR. WARD: Oh, yes.
17 MR. JOHNSON: So the point I am trying to make is 18 that back here, when the underlying unreliability of the 19 equipment was increased arbitrarily in the machine, the 20 indicator would have shown it back here a few quarters after ;
21 it happened. If you wait until you observe failures you 22 have to wait longer. 1 23 MR. MICHELSON: Why do they generate up to a large 24 peak finally? This is not a random process.
1 25 MR. JOHNSON: Yes, it is. Failures are random and Heritage Reporting Corporation (202) 628-4888 t- i
1 f1 162 1 they are randomized in this thing here.
2 MR. MICHELSON: It is just as likely the peak l
3 could have been in an earlier quarter than.later.
4 MR. JOHNSON: That's right.
5 MR. MICHELSON: And you made thousands of runs to 6 get this particular answer.
7 MR. JOHNSON: No, not to ge". this one. This is 8 one of those thousands, a typical one of the thousands.
9 HR. MICHELSON: All right, then.
10 MR. JOHNSON: This is not a smoothed out result.
11 MR. WARD: Okay. So this just happened to be the 12 big --
13 MR. JOHNSON: Just a typical one, yes.
( 14 MR. WARD: -- combination of where you got enough 15 to get a system failure.
16 MR. JOHNSON: And the point is that you can see a 17 ramp faster if you are dealing with component or train level 18 data than you can if you wait for system --
19 MR. WARD: Yes, we got that point, but we just 20 didn't understand.
21 MR. JOHNSON: Okay.
22 MR. WARD: The definition of the ordinate is 23 system unreliability?
24 MR. JOHNSON: This particular one, instead of 25 unavailability, it is unreliability. This is failure rate.
() Heritage Reporting Corporation (202) 628-4888
163-1 MR. WARD: Okay.
2 MR. CARROLL: On demand.
3 MR. JOHNSON: Per quarter.
4 MR. ' WARD : : 'Okay. That's not the same as your term 5 in the equation, but it can be related to it.
6 MR. JOHNSON: I neglected to point out that you 7 can analyze those data, the eame data in two different ways.
8 You can look at failure rate. In other words, just count 9 the failures per amount of. time, and that is'a failure rate 10 or you can also include,.look at it as an unavailability, 11 the fraction of time it would be not available. Both are of 12 interest. .This particular run happened to be unreliability.
13 So we showed this kind of stuff to the Commission,
( ). 14 and they were about as skeptical as you sound. They said, 15 why don't you try this with some real data, real plant data,
- 16. do what they call the retrospective analysis.
17 MR. CARROLL: Back to your figure for just a 18 minute.
19 MR. JOHNSON: Yes.
20 MR. CARROLL: 0.1 system unreliability means?
21 MR. JOHNSON: .1 failures per quarter. Am I 22 right, Bill?
23 MR. CARROLL: .1 failures per quarter. On a 24 system basis?
25 MR. JOHNSON: Yes.
[]' Heritage Reporting Corporation (202) 628-4888
p 164' N^)
1 - MR . MICHELSON: Failure is loss of system 2' function?
3 MR. JOHNSON: Yes. Okay. So the Commission asked 4 us to try this sort of thing out with some real plant data.
5 And basically the idea is to try to validate the-6 indicator, does it make sense with real data.
7 So what we did was analyze some historical data.
8 The Litent was to test whether the indicator can identify a 9 high unavailability of a system, a low unavailability of a 10 system, and a changing trend in a system.
11 We have the results summarized on this page here.
1 L
12 This one chart is basically what the report we 13 sent to you describes.
) 14 For the question'of could the' indicator 15 distinguish a high unavailability of a safety system, we 16 selected a plant that had had a loss of all feedwater and 17 postulated that we would expect to see a high unavailability 18 of aux feedwater for some time, high or degrsding 19 unavailability for some time before the event.
20 And that ir what the analysis showed, that for a 21 year or two before the event, as far back as we went in 22 collecting the data, the unavailability indicator was higher 23 than you would expect to see.
24 And so I concluded that this is a valid indication 25 that yes, you can see high unavailabilities.
Heritage Reporting Corporation O- (202) 628-4888
(
165
~O. 1 MR. MARD: So you picked a plant that had had a 2 loss of --
3 MR. JOHNSON: Yes.
4 MR. WARD: And went back and examined the-5 failures.
6 MR. JOHNSON: Examined the historical data.
7 MR. WARD: Okay.
-8 MR. MICHELSON: You didn't plot that up for us 9 nicely, did you?
10 MR. JOHNSON: I've got a picture.
11 MR. MICHELSON: I mean comparable to that earlier 12 slide.
13 MR. WARD: It's in the report, right?
() 14 MR. JOHNSON: It's in the report. But I can show 15 you what it looks like.
16 MR. MICHELSON: Those were' thick reports.
17 MR. JOHNSON: Here's what it looks like. This is 18 this indicator. This is a smooth indicator. In other 19 words, it's a running average of three quarters. The event 20 occurred in this quarter here. And you can see for the 21 period over which we collect the data, before that it was 22 pretty high, running around 10 percent. And the NRC 23 standard for this is an order of magnitude below this, and 24 if you add on something for a variability of the data, like 25 on a statistical quality control approach where you put an f Heritage Reporting Corporation (202) 628-4888
166
.ri-),
t .
1 ' upper band, an upper band would be in this neighborhood j 2 'here.
3 This is a pretty rough approach. There are some ,
4 better ways to do it. But this is one way.
5 So yes, you can see high unavailability.
6 MR. CARROLL: What are the points on the right?
7 MR. JOHNSON: That is after the restart. This is 8 again, trending down, and we have not collected the data out 9 here.
10 MR. CARROLL: So you don't really know on this 11- basis whether the corrective action they took was effective.
12 It seems to be trending in the right direction, but you 13 still haven't --
(f 14 MR. JOHNSON: We haven't collected the data out in 15 this period --
16 MR. CARROLL: You haven't got points down to where 17 you'd like to see them.
18 MR. WARD: But do those high points there, .17 for 19 example --
20 MR. CARROLL: That's because you have a failure.
21 MR. MICHELSON: Is that the second failure?
22 MR. WARD: That's what I'm trying to find out. Is 23 that associated with operation after the failure? i 24 MR. JOHNSON: Let me say, please don't place too 25 much emphasis on this. I don't know whether that particular Heritage Reporting Corporation (202) 628-4888
167 p.
L A=
t.Q 1 point includes -- over here it is'a smooth data. But I
- 2' don't know if this includes the two quarters prior to this 3' or if it was restarted after this.
4 DR. VESLEY: No, it does not.
5 MR. JOHNSON: So this is the way it started again Okay.
~
6 after the restart.
7 DR. VESLEY: It does include failures. That is 8 what the cause of that peak was. That train was still 9 failing after startup.
10 MR. WARD: Is that right?
11 DR. VESLEY: Right.
12 MR. WARD: After all the work they did.
13 DR. VESLEY: Well, yes.
() 14 MR. MICHELSON: So what do you conclude from that 15 graph?
16- MR. JOHNSON: Okay. What I conclude is that yes, 17 this indicator -- what we are doing is we are not testing 18 the plant, we are testing the indicator. And we test the 19 indicator and we say gee, yes, the indicator can observe 20 high or flag high unavailability of a safety system in a 21 case where we expected to see it. So we say okay.
22 MR. MICHELSON: Is that a plot of Q?
23 MR. JOHNSON: Yes. It is an indicator of Q.
24 MR. MICHELSON: It's a plot of Q. I thought it 25 was a plot of Q train or Q system, rather.
Heritage Reporting Corporation
(( ) (202) 628-4888
1 168 o- l= MR. JOHNSON: Oh, oh. Okay. Excuse me. This 2 -particular thing is a train average, average of each of 1.
3 'those two trains.
4 MR. MICHELSON: Of what?'
.5 MR. JOHNSON: Of unavailability.
6 MR. MICHELSON:. It-has nothing to do with Q, then?
,; 7- MR. JOH" SON: Q is unavailability. Is that what 8 you mean?
.50 MR. MICHELSON: Q is a modified unavailability
-10 that has, you know, that is your first equation, Q train.
11 MR. JOHNSON: Yes. This.is the average of those
. 12 two trains. This.was plotted up again on an average, train 13 average, on a system, the product of those two, and so (f.- 14- .forth. The one I happen to have'here is the average train'.
15 MR. MICHELSON: Let me try'one more time.
16' In your very first slide, you have Q train.and Q 17 system.
18 MR. JOHNSON: Yes.
19 MR. MICHELSON: Is this a plot of Q train or Q 20 system? Or is it something else?
21 MR. JOHNSON: This is in between.
22 MR. MICHELSON: Okay. It is something else.
23 MR. JOHNSON: This is, as.I mentioned, there are 24 several ways of aggregating the data from the train level.
25 The first, simplest way, is to simply take, for a Heritage Reporting Corporation (202) 628-4888
, 169
,t 1 .two-train system, take the average of the two. That is what 2 this'is.
3 Another way --
4 MR. MICHELSON: That's not what.your equations 5 say.
6 MR. JOHNSON: This one?-
7 MR. MICHELSON: Yes.
8 MR. JOHNSON: That's right. There are several 9 ways of coming up with a way to aggregate it.
10 MR. MICHELSON: If you had given me the equation 11 that you were plotting, I would have understood it. You 12 really are plotting some other equation.
13 MR. JOHNSON: That's true. This is an average.
( ). 14 MR. MICHELSON: I just was trying to somehow 15 follow it.
16 DR. VESLEY: It's an average of the Q individual 17 trains.
18 MR. MICHELSON: Okay.
19 DR. VESLEY: Carl using it to get Q train 1 plus Q 20 train 2 and then he is averaging.
21 MR. MICHELSON: By averaging, you mean just 22 dividing by 27 23 DR. VESLEY: Yes.
24 MR. MICHELSON: So it is sort of a Q system over 25 2, then?
Heritage Reporting Corporation O. (202) 628-4888
I
, 170
~
1 DR. VESLEY: Q train. Yes, that's right.
2 MR. MICHELSON: By using your equation, it is Q 3 system over 2. Is that correct?
4 MR. JOHNSON: No.
5 MR. MICHELSON: That's what I would do if I took Q 6 train 1 and Q train 2 and divided by 2.
7 DR. VESLEY: He's averaging.
8 MR. JOHNSON: Not multiplying.
9 MR. MICHELSON: That's a product in there.
10 DR. VESLEY: Yes, that's the product.
11 MR. MICHELSON: Okay. That's something else yet.
12 MR. CARROLL: I'm interesting in following up on 13 your figure just to understand the right hand side of it.
() 14 Can you put that back up there?
15 MR. JOHNSON: If I can find what I did with it 16 here. Here it is.
17 (Pause) 18 MR. CARROLL: So we had the situation in '83, '84, 19 that was telling us something bad was going to happen.
20 MR. JOHNSON: Well, I wouldn't say that. It would 21 call your attention, the idea is it would flag attention to 22 gee, I wonder what is going on.
23 MR. CARROLL: Okay. So then an event happened.
24 MR. JOHNSDN: Yes.
25 MR. CARROLL: On the right hand side we now have
() Heritage Reporting Corporation (202) 628-4888
l 171
.1 the plant back in operation.
2 MR. JOHNSON: Yes.
3 MR. CARROLL: And I guess what I am still not 4 clear on is --
5 MR. JOHNSON: This still flags your attention.
6 Right? .And that's the whole idea of an indicator, to flag 7 attention so someone can figure out well, what is happening?
8 The indicator doesn't tell you what is happening, it just 9 calls attention to this particular thing.
10 So to me I'd say yes, that is an indicator that 11 the indicator works.
12 MR. WARD: Even though there seems to be a nice 13 trend in the '87 '88 period, I guess you conclude that the
() 14 plant may not have fixed their earlier problem.
15 MR. JOHNSON: No. I would simply.say this 16 continues to. flag attention to somebody, anybody interested, 17 like yourself, to say gee, I think maybe I will, through 18 some other mechanism than reading this graph, try to figure 19 out what happened, what is going on. ;
20 DR. VESLEY: It has not yet come down to an 21 acceptable level as measured by that --
22 MR. JOHNSON: Well, we don't have the data out 23 here, Bill.
24 DR. VESLEY: Yes, that's right. But up to where 25 that graph is, that's right.
Heritage Reporting Corporation (202) 628-4888
fs 172 U 1 MR. JOHNSON: This is almost a year and a half 2 ago.
3 DR. VESLEY: If it has come down to acceptable or 4 gone up, that's where the data was found.
5 MR. CARROLL: This is averaged over three 6 preceding -- ?
7 MR. JOHNSON: Three quarters.
8 MR. CARROLL: Okay. So the effect of the failure 9 isn't in that at all. Okay.
10 MR. MICHELSON: When a train fails, the system 11 does not necessarily lose its function.
12 MR. JOHNSON: That's exactly right.
13 MR. MICHELSON: Is that why you are adding the --
. (~T (,) 14 Why are you adding the Q for one train which is just a loss 15 of function for one train and adding it to the second train 16 and then dividing by 2 to get an average? What does that 17 have to do with loss of total system function?
18 MR. JOHNSON: Nothing. Some people, it is a way 19 of how do you read this information? Some people would 20 rather look at it and say what is the average train 21 unavailability, that means something to them.
! 22 Other people say, what is the system 23 unavailability? They like to see it that way.
24 So we do it both ways. Show them either the 25 average or the product --
Heritage Reporting Corporation I}
' (202) 628-4888
l 7
173 1 MR. MICHELSON: That is neither.
2 MR. JOHNSON: This is the average. Average train 3 unavailability.
4 MR. MICHELSON: It is the average of two. trains' 5- unavailability.
6 MR. JOHNSON: Yes.
7 MR. MICHELSON: But that is not even related to-8 system unavailability necessarily.
9 MR. JOHNSON: It's the square root. In that 10 equation I gave you it is the square root of the system 11 availability. Or close to it. It isn't quite. Because the 12 equation I had down there wasn't the average squared. But 13 this is one way of looking at it.
7() 14 DR. VESLEY: If one train is failing all.the time 15 'and the other train never fails --
16 MR. MICHELSON: The system is never lost if the 17 other one never fails.
18 MR. JOHNSON: That's right. That's right. Yes, 19 you're right.
20 DR. VESLEY: Excuse me. Bill Vesley. In this 21 case, these two trains were similar. In fact, Q train was 22 approximately where Q train 2 was statistically, so that's l 23 why they were averaged.
24 MR. MICHELSON: None of it makes sense.
25 DR. VESLEY: In this case, that is why the average O, Heritage Reporting Corporation (202) 628-4888
I
)
E 174 '
1 -- these two trains were behaving alike. There was no l l
2 difference --
3 DR. REMICK: But I fail to see the significance of 4 this particular indicator.
5 All we are really saying here is if you have some 6 indication of train failure, that presumably will alert you 7 that you might have a system failure.
8 That is just common sense. If you have an 9 increased rate of train failures, whether it is all in one 10 train or half in one train and another train, that 11 potentially you might be leading to a system failure.
12 MR. JOHNSON: It gives you more than just it did 13 or didn't fail. -It gives you the fraction of the time that L () 14 the train was unavailable.
15 DR. REMICK: But with some assumptions about half 16 a surveillance test and so forth.
17 MR. JOHNSON: Yes.
18 DR. REMICK: I guess I don't want to belittle it, 19 but I just don't see that your demonstration tells me much.
- 20 I would assume that if you had an indication of 21 increased train failures, that that should alert you to the 22 fact that you might end up with a system failure. ,
23 MR. MICHELSON: I don't like his equation.
24 MR. JOHNSON: The question is, can you put some 25 numbers on this so that you can trend it?
O Heritage Reporting Corporation (202) 628-4888 i
._______________________D
p
- ,- ,s; 175
\} -1 MR. MICHELSON: You've got a.real problem with 2' that equation, because you do a surveillance'immediately on
'3 the second train when the first one fails. Generally you 4 do..
5, And so I don't know what the meaning of half the 6 surveillance interval is in that case.
7 MR. JOHNSON: Some systems you do that on and some 8 you don't.-
9 MR. MICHELSON: I think on all of those that you 10 were looking at here, if you lose train RHR you have to make 11 sure train'B is okay right away.
12 MR. CARROLL: If it doesn't fail, it's that term 13 to zero.
() 14 MR. MICHELSON: yes, but then why does it take 15 half the surveillance interval?
16 MR. CARROLL: No, no,.no. If failure is zero.
17 MR. MICHELSON: This is per train. Train A 18 failure will be something -- if it is failed, train B bill 19 be zero. !
20 DR. VESLEY: In fact, that is right. If those 21 trains are tested, then it is not half the surveillance 22 interval, but half the time since the last train test.
23 MR. MICHELSON: That's right.
24 DR. VESLEY: That could be just a day or just a 25 couple hours.
I Heritage Reporting Corporation (202) 628-4888 i
176
- l. .1 MR. MICHELSON: That's right.
l l 2 DR. VESLEY: And that would be included.
3 MR. MICHELSON: Yes. l 4 DR. VESLEY: In fact, that happens to be a general 5 formula. But for'these specific cases, as in diesels, too, . !
6 'where one train is tested after the other fails, then that 7 test interval is very small.
8 MR. MICHELSON: It is not just half the 9 surveillance interval.
10 DR. VESLEY: No.
11 MR. MICHELSON: Okay. Then I don't have a 12- problem. Thank you.
13 MR. WARD: It is just common sense, but it seems
() 14 to me that you could actually put a quantity on it so that 15 you can first of all compare that with some. standard and 16 second you can look at the trend more sensitively than you 17 could --
18 DR. REMICK: But I could have total train failures 19 per quarter, and I would have an indicator, and I would have 20 done as good a job I think as they have there. It does not 21 have time factored in. And maybe that adds something more 22 than I think it is. In other words, it is a fraction of 23 time it is unavailable.
24 But I could say just count up the number of train 25 failures in a quarter and plot that and I think it would Heritage Reporting Corporation
-() (202) 628-4888
g 177-
\)
l 1 have shown me, alerted me to the fact that you might be 2 headed toward a system failure.
3 DR. VESLEY: It is interesting that at the end-of 4 that plot, that train was down not due to failure but due to 5 maintenance time.
6 And so the maintenance was dominating, caused that 7 train to have a high unavailability. So that-is the other 8 factor coming in. At the beginning of that high trend it 9 was due to failure, but it was still high because those 10 trains were being dropped down for an inordinate amount of 11 time according to that for maintenance.
12 MR. WARD: That's after the NRC made a lot of 13 inspections.
() 14 DR. VESLEY: Well, whatever it is. So it caused a 15 train to be unavailable in other ways. So it was the 16 maintenance contribution toward the end of that period that 17 was causing it to have a fairly high unavailability.
18 So that is the other factor that comes in and has 19 to be balanced against the figure, that if the train is 20 being brought down too much for maintenance, or is left 21 down, or the failure not corrected, it has the same effect 22 on the unavailability. So those are two factors.
23 MR. JOHNSON: Two other things were looked at.
24 One is a plant with the opposite situation.
25 Namely, a plant with a reputation for running well. We
() Heritage Reporting Corporesion (202) 628-4888 I
i
I E 178 i -
1 expected to see a low unavailability of aux feedwater for 2 that plant. =And this was a three-unit plant. And what we 3- found was in two of the units, yes, there was a low 4' unavailability of aux feedwater but in the third unit the 5' unavailability was low. Then it went up for a short period 6 of time, and then it went back down again.
7 On discussing this with the resident inspector and 8 with the utility, it turns out that in fact there were j 9 problems with the aux feedwater system of that one unit 10 during this time period. And so, and this indicator is 11 simply reflecting those problems.
12 So I concluded that yes, this is a valid 13 indication that you can see low unavailabilities. And it is
() 14 sensitive. And that does reflect intermittent, not-15 intermittent, but the problem that occurred was recognized 16 and resolved.
17 The third case was to look at a plant where the i l
18 region had reported that the plant had installed some 19 improvements in maintenance, surveillance and quality 20 programs. And so I expected to see that the unavailability 21 of aux feedwater at that plant would have improved.
22 And the indicator did not show that. Instead, 23 after these things were installed and these changes were 24 made, in fact the unavailability indicator was higher, 25 degraded in other words, but then it improved, there was an
() Heritage Reporting Corporation (202) 628-4888
179 l' -improving trend. But-initially it was high unavailability 2 after that.
3' ~MR. CARROLL: Can we see that?
=4 MR. JOHNSON: So I don't know what to make of 5 that. Can you see it? Yes.
6 MR. MARD: You don't know whether the indicator is 7 poor or whether the so-called improvements were poor?
8' HR. CARROLL: Or whether there is a time constant 9 associated with it.
10 MR. JOHNSON: That's possible. I don't know.
11 Let's see. Here is the aux feedwater and .are is 12 the diesel generators. I guess you can't see two at once.
13 For the aux feedwater, this again happens to be-( 14 read out as average unavailability for'these trains.
15 And you can see, this is when the improvements.
16 were installed and it was low before that. Afterwards it is 17 high trending down.
18- With the diesel generators --
R19 MR. CARROLL: The O-5 line was the same, 0-5 line 20 --
21 MR. JOHNSON: Yes.
22 MR. CARROLL: -- in that first --
23 MR. JOHNSON: That's right. Same scale.
24 MR. CARROLL: That is the statistical --
25 'R. JOHNSON:
M Yes, that's right.
() Heritage Reporting Corporation (202) 628-4888
f
-y 180
%.J' 1 MR. CARROLL: -- bound, if you will.
2 MR. JOHNSON: Same bound. Yes. Same bound of 5 3 percent. And we didn't see what we expected to see.
4 On the diesels --
5 MR. CARROLL: It couldn't have got better.
6 MR. JOHNSON: Should have got better.
7 MR. CARROLL: Well, it couldn't have.
8 MR. CARROLL: And they couldn't have. That's 9 right. They were already good.
10 On the diesels, the diesels did get better.
11 MR. MICHELSON: Let me ask again, since you didn't 12 label, your ordinate there is nondimensional, which 13 suggested to me it was Q.
() 14 MR. JOHNSON: No. It is unavailability for a 15 train. Average train unavailability.
16 MR. MICHELSON: Okay. So you haven't told me what 17 you are doing with Q yet, which is your indicator, I 18 thought.
19 MR. CARROLL: Before you leave that one, just 20 point out to me where those improvements were made on that 21 plot.
22 MR. JOHNSON: In here.
23 MR. CARROLL: Okay.
24 MR. MICHELSON: Now, are you trying to show us how 25 your indicator works or are you just trying to show us how Heritage Reporting Corporation
[)
(202) 628-4888 l
~
O 1 to plot unavailability?
2 MR. JOHNSON: No. I'm trying to tell you what we 3 found when we tested the indicator against historical plant 4 data.
5 MR. MICHELSON: You didn't plot the indicator, 6 though. That wasn't a plot of the indicator.
? MR. JOHNSON: We did plot it. I didn't bring that 6 version of the plot. There's two plots for each one. One is Q, system unavailability. The other is average train.
9 I 10 happened to bring the average train.
11 MR. MICHELSON: Yes. But your indicator is the Q, 12 isn't it?
13 MR. JOHNSON: Yes.
() 14 MR. MICHELSON: Yet you are telling us all the 15 time about unavailability, which is not Q. It's related to 16 Q.
17 MR. JOHNSON: No. Q is unavailability.
18 MR. MICHELSON: No. Q is by definition hours of 19 out of service plus this correction factor divided by hours 20 needed.
21 MR. JOHNSON: that is the definition of 22 unavailability.
23 MR. MICHELSON: Unavailability I thought was just 24 hours out per. hours needed.
25 MR. JOHNSON: That's right.
O Heritage Reporting Corporation (202) 628-4888
182 1 MR. MICHELSON: That's not Q. He's got an 2 additional factor on Q. It's not just hours out per hour 3 needed.
4 MR. WARD: No , no. The hours out of service is 5 where it is programmatically out for maintenance.
6 MR. MICHELSON: Or whatever.
7 MR. WARD: And the other is to add in the hours 8 out of service due to failures.
9 MR. MICHELSON: If you tell me this is an 10 unraliability number, that just tells me this is the amount 11 of time it is out when I really needed it. That's all it 12 tells you. Unreliability, I think by definition, is just 13 how many hours do you have it, in this case how many hours l
() 14 don't you have it, when you needed it.
l 15 MR. JOHNSON: Yes.
l 16 MR. MICHELSON: But that's not Q. Q is something i 17 more than that.
18 MR. WARD: Not for his definition, i
19 MR. MICHELSON: I'm looking at his definition.
20 Reliability just takes these two numbers and that is your 21 ratio. That is unreliability. It has nothing to do with 22 this.
23 MR. JOHNSON: This is unavailability.
24 MR. WARD: Unavailability due to failure.
I 25 MR. JOHNSON: Unreliability would be how often did Heritage Reporting Corporation O- (202) 628-4888 l
183
- O7 1. it fail. Unavailability is the probability that it will 2 not be --
3 MR. MICHELSON: Right. Or it's really just the 4 number of hours out of service per hour that you needed it.
5 MR. JOHNSON: Thdt's right. But there are two l
1 6 sources of it being out of service.
7 DR. REMICK: Let me see if I can help.
8 Your Q train, is that what you are calling 9 unavailability?
10 MR. JOHNSON: Yes.
11 MR. MICHELSON: This is something else.
12 DR. REMICK: Carl, he says a Q train, the formula 13 you are looking at there, is what he his calling
() 14 unavailability.
15 MR. MICHELSON: Then it isn't the buual calculated 16 definition of unavailability, then.
17 What you have here is more than out of - you've 18 got numerator, hours out of service; and you've got a 19 denominator, hours critical. That would be unavailability.
20 MR. JOHNSON: This is hours out of service for 21 maintenance.
22 MR. MICHELSON: Plus -
23 MR. JOHNSON: I didn't finish, you're right. I 24- didn't finish that expression. It's hours out of service 25 for maintenance. Plus hours out of service because the
{) Heritage Reporting Corporation (202) 628-4888
184 l' thing is already in the failed state.
2 MR. MICHELSON: And that is not the usual
~
3~ definition.
4 MR. JOHNSON: Yes, it.is.
5 MR. MICHELSON: Of unavailability?
6 MR. JOHNSON: Of unavailability. Absolutely.
7 That's the one that's used in all PRAs.
8 MR. MICHELSON: Is that the one you use then?
9 Then I was thinking of something different.
10 DR. REMICK:- Moving right along.
11 MR. JOHNSON: Thank you. Okay.
12 So you want to talk about data. Where do we get 13 this data from?
() 14 MR. CARROLL: Well, as long as we've looked at two 15 of the plots, why don't we look at the third one?
16 MR. JOHNSON: Oh, the third one?
17 MR. CARROLL: Just because I like to look at 18 plots, I guess.
19 MR. JOHNSON: Oh, the one where --
20 MR. WARD: Plant number 17 21 MR. JOHNSON: No , the one where it went up over a 22 period of time. Okay.
23 Well, I won't bother to show you the other two.
24 We looked at a three-unit plant and we expected to see 25 unavailability in this range here, down here.
Bt.?itage Reporting Corporation (202) 628-4888
! ,, 185 l I) '
1 And for two of the units we did. This is the 2 third unit.
l l 3 MR. CARROLL: Okay.
4 MR. JOHNSON: And so the question is gee, well, 5 how come this?
6 Well, it turned out this is real. This was some 7 particular problems they are having with both the particular 8 turbine train but also with the motor trains. And when they 9 figured out what the problem was, they overcame it, and back 10 down here.
11 MR. CARROLL: And with your last three quarter 12 averaging, if you solved the problem, first quarter of '87, 13 it takes a while for it to drop down, right?
() 14 MR. JOHNSON: Yes. That's right. So the actual 15 peak, if you looked at the unaveraged data, it is much 16 slower.
17 But this smoothing is a tradecff between how fast 18 you see it jump, having it very jumpy, and it is a lot of 19 noise, and it is a tradeoff between the noise versus being 20 able to see the trends. That is why there was a lot of this 21 simulation done to try out the different things.
22 And Dr. Vesley and Dr. Azarm there ran these 23 simulations to come up with some reasonable ways of being Ei able to see results. And this happens to be a result of 25 using that smoothing that was come up with in the
() Heritage Reporting Corporation (202) 628-4888 l
- 186 v
1 simulation.
i
! 2 Okay. Data. NRC does not collect this data. So i
3 where could you get this data? Well, here are three 4 potential sources.
5 One is the operating logs, where the outages 6 against limiting conditions for operations are recorded.
7 And this includes all of the information needed for the 8 indicator, but I need to put on the caveat that we talked 9 about before about recognizing or distinguishing when a 10 piece of equipment failed to meet the system requirement 11 versus when it was simply taken out for maintenance.
12 With that caveat, the logs include that 13 information, but for NRC to get this information would
() 14 require rulemaking to change the reporting requirements.
15 Another possible source is INPO keeps an 16 indicator, the industry keeps a very similar indicator, 17 safety system performance indicator, that includes most of 18 this information.
19 It does not distinguish between trains, but other 20 than that it is very similar. And this is not reported to 21 NRC either.
22 DR. REMICK: Is it available to resident 23 inspectors?
24 MR. JOHNSON: This is available to the resident 25 inspectors. It is proprietary information.
() Heritage Reporting Corporation (202) 628-4888
187 I D.
a 1 MR. CARROLL: Same as the first one, really.
1 2 MR. JOHNSON:' Pardon?
3 HR. CARROLL: So is the first one.
i 4 MR. JOHNSON: Yes. That's not reported-to NRC, 5 either.
6 MR. CARROLL: But it is available to resident 7 inspectors.
6 MR. JOHNSON: A third option, possibility, is the 9 nuclear plant reliability data system. This includes part 10 of the information. Namely, it includes the component 11 failures. It does not include outages of equipment due to 12 human errors or surveillance tests or maintenance, 15 preventive maintenance. And the system is voluntary.
(f 14 Again, this system is available to NRC, but on a 15 proprietary basis.
16 DR. REMICK: Is there anything about the indicator 17 that you've talked about that is better than the INPO safety 18 system performance indicator?
19 MR. JOHNSON: Yes. But they are more similar than 20 different. The similarities are the big thing. And so when 21 we talked about, I don't want to blow the differences up to 22 be big differences.
23 The differences are that the INPO indicator takes 24 the numbe- of failures that occurred on critical components 25 in a system and divides by the number of trains, and so it l~
Heritage Reporting Corporation
(]) (202) 628-4888
i 188 1 is-essentially an average train unavailability.
2 It does not, the data do not come in with 4 3 individual trains so you cannot distinguish problems with an But it is very similar to this one. I 4 individual train.
5 DR. REMICK: Well, on the assumption that the NRC 6 needs an. indicator like this, and if it is very similar, why 7 can't the staff adapt to that so there is an additional 8 requirement of data?
9 Because to make this effective, you are going to 10 have to do something about data.
11- MR. JOHNSON: Yes, sir. The staff does not get 12 this indicator.
13 DR. REMICK: I realize that.
() 14 MR. JOHNSON: So yes, it would be very useful if 15 the NRC did receive this.
16 DR. REMICK: But since you have to get data, on 17 the assumption the NRC neede such an indicator, it has to 18 get data somehow, someway.
19 MR. JOHNSON: Yes.
20 DR. REMICK: Either through rulemaking or arm 21 twisting or coercion or whatever, they have to get it.
22 And if this indicator is very, quite similar to 23 the INFO one, why try to seek additional data than the 24 industry is already --
25 MR. JOHNSON: I'm not arguing with you.
() Heritage Reporting Corporation (202) 628-4888
189 1 DR. REMICK: All right.
2 MR. JOHNSON: We would like to get this indicator.
3 DR. REMICK: Okay.
4 MR. JOHNSON: At least I would like to. I 5 shouldn't say we. I would like to get that indicator, see 6 the NRC get that indicator.
7 DR. REMICK: The safety system?
8 MR. JOHNSON: Yes.
9 DR. REMICK: Okay.
10 MR. JOHNSON: Additional analysis that we are 11 doing to wind up this job is completing the analysis of 12 plant data. We collected some data that we think has some 13 less uncertainties in which components failed and which ones
() 14 were simply taken out for maintenance by going to a couple 15 more sites which keep particularly good records, keep 16 particularly good reliability data sources.
17 So that is five more units that are being analyzed 18 now. And another task is testing this type of indicator, 19 using NPIDS data, where NPIDS reports part of that, part of 20 the information that is in the logbook.
21 The idea is to see, would you see the same trends 22 but at a different level. And this is going on now. This 23 analysis is going on now at Brookhaven and SAI.
24 And the third thing going on is, or about to go 25 on, is looking at would NRC implementation of an indicator
() Heritage Reporting Corporation (202) 628-4888
190 L
- O 1
~
of unavailability of selected safety systems induce a 2 utility to put too much human intervention into systems, 3 these particular safety systems, at. power operation, and 4 would be an unprofitable way of operation.
5 DR. REMICK: This leads me to tb3 question of why, 6 how would this agency use either the indicator you talked 7 about or the INPO indicator if they had it available? How 8 would the agency use that?
9 MR. JOHNSON: Okay. Let me try the next' picture 10 on you.
11 DR. REMICK: All right.
l 12 MR. JOHNSON: Two answers. The first answer is, a 13 lot of our regulatory activities have to do with setting
'( ) 14 requirements that we believe if the utilities follow them 15 the plant will operate with a reasonably low risk.
16 And so by converting the feedback from the 17 operation, interpreting part of the data from the operation 18 back, in terms of risk related numbers, we can enhance our 19 understanding of the effectiveness of our regulatory program 20 at that particular plant.
21 DR. REMICK: If your resident inspectors have 22 access to the indicator which is already being developed, 23 and I realize that might not be ideal, certainly that 24 resident inspector is free to report that back to 25 headquarters, verbally at least. So it seems to me
~
Heritage Reporting Corporation
~( ) (202) 628-4888
191
\q
~J 1 headquarters has an indication of how that is going.
2 MR. JOHNSON: But some of these values seem, the 3 don't look believable to me, anyhow, some of them.
4 And so I think we should have the opportunity to 5 poke away at those particular numbers and verify that they 6 really do, they really are accurate. See what I mean? If 7 it is a report that nobody at the NRC reads or is intended 8 to read, it maybe doesn't get treated with the same accuracy 9 as something that we could have an opportunity to check, and 10 put basically some daylight on it, to see would the light of 11 day and scrutiny make this thing more accurate.
12 I think being used -- what I'm trying to say is 13 get it into use rather than keeping it under covsrs.
() 14 DR. REMICK: But isn't the agency's most able 15 person to perhaps determine how those numbers are being 16 generated in the plant is the resident inspector versus 17 somebody reading the report back here?
18 MR. JOHNSON: The policy for the performance 19 indicators is that they are intended for the top management 20 people rather than the resident.
21 DR. REMICK: But you said the resident inspector 22 has access to that information.
23 MR. JOHNSON: Yes. But the performance indicators 24 are intended for our management.
25 DR. REMICK: But if there is suspicion that those
() Heritage Reporting Corporation (202) 628-4888
_ - _ - _ _ _ _ _-____ _ ___________ ___ _ _______ __ _ ___-____-_ __-_______-__ ____________ _ __-__=_-________ -
192
' 3
/
. 1 aren't correct, isn't the resident inspector capable of 2 going in-and seeing how those numbers were determined at the 3 plant?
4 MR. JOHNSON: Since they are not something 5 reported to NRC, I would presume that he has other fish to 6 fry rather than whether some internal number within the 7 utility is accurate.
8 DR. VESLEY: Our discussions with the residents, 9 they have no guidelines on how to interpret or how to use 10 those indicators or whether to focus on that indicator.
11 So if they would have such guidelines, they would 12 certainly help. .But right now they don't.
13 DR. REMICK: No, I think that is a different
) 14 question. That goes back to my question, how would the l 15 agency use such an indicator.
16 But if your concern is that those indicators are i
17 being falsified or --
18 MR. JOHNSON: No, not falsified. Rather, they are 19 new. They need to be flushed out or worked out. That's l
20 all.
21 DR. REMICK: But who better to do that, for the 22 NRC side, assuming the utility and INPO aren't able to do-23 that, to make sure they are legitimate, than the resident 24 inspector who has access to the data right at the plant, and 25 to the people doing it, presumably?
() Heritage Reporting Corporation (202) 628-4088
193 1 HR. JOHNSON: The resident is one key person.
2 Also the resident may have, from plant to plant, may have 3 differing ideas about how do you go about calculating 4 unavailability. And we need to standardize this sort of 5 thing. INPO is doing this. INPO is taking a lot of effort 6 to do that.
7 DR. REMICK: I've seen the INPO description of how 8 ou do it. You can't go too far wrong, if you follow the 9 guidelines. Even I could do it.
10 MR. JOHNSON: Yes.
11 DR. REMICK: With those guidelines, assuming I had 12 the data.
13 MR. WARD: So you are suggesting that the resident
() 14 go ahead and just use this as a tool for himself?
15 DR. REMICK: No. I'm saying if headquarters feels 16 there is a need for somebody back here to double check those 17 numbers, and you didn't say why, but there must be some 18 suspicion that they are not being generated properly, who 19 better than the resident inspector right at the plant where 20 it is being done could look into how are they being 21 generated?
22 MR. JOHNSON: That's true.
23 DR. REMICK: I'm trying to get at, why does the 24 agency feel that it needs its own indicator on how to use 25 it?
() Heritage Reporting Corporation (202) 628-4888
194 O 1 MR. JOHNSON: The agency hasn't made this choice.
2 This is --
3 DR. REMICK: But you are proposing something, so 4 you must have something in mind. ,
l 5 MR. JOHNSON: I'm proposing, yes. It is ]
6 basically, remember, we started out saying our indicators 7 are intended to reflect risk. They don't. They are 8 qualitatively in that area. And in this area, it is safety 9 system failures. I said okay. Let's come up with an 10 indicator that we can trend rather than just simply system 11 failure here, system failure here and syst9m failure there.
12 This is a way of trending to improve our feedback 13 and understanding of what the plant data are telling us.
h 14 Then the next question is okay, here is a possible 15 source of this data.
16 And I think we are agreeing it says yes, that is a 17 good source, let's use it. The indicators are intended for 18 top management. They presently are not included in the 19 performance indicator book that goes to top management.
20 DR. VESLEY: If the resident had it, that would 21 not necessarily mean the top management would get it.
22 DR. REMICK: Top management of NRC.
23 MR. JOHNSON: Yes.
24 DR. REMICK: But it could mean they ec-did, of 25 course, jll Heritage Reporting Corporation (202) 628-4888
1 1 MR. JOHNSON: INPO would have to provide this 2 information.
3 DR. REMICK: I'm thinking about certainly when the 4 resident inspector sees the information, he is free to share.
5 that with NRC top management if they wish'it. Even over the 6 telephone.
7- MR. JOHNSON: Yes.
8 DR. REMICK: Who is going to stop that? I'm not 9 suggesting it.- But what I am trying to get at is, you are 10 telling me:that this. indicator is very close, very similar
- 11 to the INFO one and yet to implement something like this is 12 going to take a new data source. And what I am trying to 13- get in my mind, what justifies this additional potential
() 14 source of data collecting other than the fact that it is not 15 available to NRC top management?
16 MR. WARD: This is a research program. And I 17 think we've got all these performance indicators that 18 different people are using and none of them have been 19 validated in any scientific sense. They are just opinions 20 of people or consensus opinions of people. Hey, these are 21 probably good indicators, of something or other.
22 It seems to me what they are trying to do is 23 develop an indicator here that may have some sort of 24 scientific validation,' and this is what he has come up with.
25 And he is not really saying how it is going to be
() Heritage Reporting Corporation (202) 628-4888
_ m
196 0a 1 used yet.
2 Is'that a fair description?
.3' MR. JOHNSON: That is exactly my opinion. I agree 4 100 percent with what you just said. !
5 DR. REMICK: And I'll buy that. But let me just 6 modify my question then.
7 MR. JOHNSON: Okay.
8 'DR. REMICK: It seems to me a logical thing to be 9 'doing then is.taking the INPO performance indicator in this 10 area and independently validating it's a good indicator 11 rather than rediscovering one.
12 MR.~ WARD: Start using this one.
13 MR. CARROLL: You say the two are similar. But
( . 14 isn't one major difference that the INPO one is kind of a 15 generic one, it is all safety, all defined safety systems?
16 MR. JOHNSON: No. It's three safety systems.
17 MR. CARROLL: All right.
18 MR.' JOHNSON: And it is average train 19 unavailability for each system.
20 MR. CARROLL: But composited together?
21 MR. JOHNSON: No. Separate. Three individual 22 systems.
(
24 MR. JONNSON: For INFO.
25 MR. CARROLL: Oh. Okay. I'm sorry.
/ Heritage Reporting Corporation
-] (:.02 ) 628-4888 l
L- _ _ _ _ _ _ _ _ _ _ - - -
197 0- 1 DR. VESLEY: It does composite the trains. It t
2 doesn't differentiate them. This indicator differentiates 3- trains if they are different, an I think that is the basic 4 difference.
- 5. INPO never looks to see if the trains are 6 different. This looks first before it aggregates. But it's
-7 a minor difference.
8 MR. WARD: But it is another level of sensitivity.
9 MR. CARROLL: Another difference, and it is a 10 problem I guess with both of them, is that you say your gral 11 is to relat s this to risk.
12 But in fact, only for one system, aux feedwater, 13 have you even established an acceptable unavailability
() 14 criterion.
15 When is somebody going to look at that?
16 MR. JORNSON: We will have to do that this year.
17 MR. CARROLL: But that is in your plans?Y 18 MR. JOHNSON: Yes.
19 MR. CARROLL: And how do you go about that? I 20 mean, the number for aux feedwater was kind of pulled out of 21 the air as I remember after TMI. It isn't really a number 22 based on say current PRA considerations.
23 MR. JOHNSON: This would be backing down from PRA )
24 considerations to say here is a reasonable estimate.
25 Bill, would you like to comment on this?
() Heritage Reporting Corporation (202) 628-4888 ]
l 1~ DR. VESLEY: Those columns were based on PRA 1
2 considerations, on looking at the PRA results to come up 3 with guidelines. These bounds are obtained from PRAs.
4 They happen to be fairly close, on aux feedwater !
5 are fairly close to what the staff has come up with. So 6 they are essentially validated.
7 MR. CARROLL: They did a good job of pulling the 8 number out of the air.
9 MR. JOHNSON: One other thing I would like_to 10 point out in answer to your question about why would the NRC 11 want to look at something like this.
12 This is not a near term thing. This is a longer 13 range thing. But the NRC has various initiatives going in
() 14 area of safety goal implementation, technical specification 15 improvements, maintenance improvements, aging research, 16 license renewal, focusing inspection on areas to try to find i'
l 17 out the causes of problems. And feedback from performance 18- of a plant in terms of the effectiveness of the regulatory L 19 program in holding risk at a reasonable level would be 20 useful, I believe, in conjunction with all the other 21 information we've got, would be-simply another point of view
- 22. on looking at that information to help us sharpen up and in 23 particular possibly help to integrate individual i
24 initiatives.
25 And that is a long range thing.
O Heritage Reporting Corporation (202) 628-4888 I. - _ - - - - - - _ _ _ _ _ -
199 1- Okay. So in summary.
2 One attribute of safety is availability of safety
-3 systems.
4 We have developed an indicator of unavailability 5 of selected risk-significant safety system.
6- And the validation and sensitivity tests of this-7 indicator are promising, but are not conclusive.
8 We plan to conclude this validation analysis this 9 calendar year.
10 And two other follow-up tasks going into next 11 year. And the results I expect will form part of the 12 technical basis for the thing you were just talking about, 13 which is an NRC decision to either use or to not use an
() 14 Indicator of unavailability of selected safety systems..
15 DR. REMICK: Going back to your earlier discussion 16 about maintenance and training and the comment that Mr.
17 Michelson made about failures being due to more than just 18 maintenance, they are due to operator error and so forth.
19 A thought went through my mind at that time.
20 Rather than calling it a maintenance indicator, you might 21 call it a functional readiness or something like this, 22 readiness being including both --
23 MR. JOHNSON: That's really what it is. And 24 that's why we called it an indicator of safety system 25 function trains.
() Heritage Reporting Corporation (202) 628-4888
,3 200 l l ) 4 1 DR. REMICK: Okay.
2 MR. JOHNSON: And it is intended as a risk, as one 3 aspect of risk. i 4 DR. REMICK: Yes.
5 MR. JOHNSON: Certainly not all the risk. Just 6 one little piece of it. But an important piece.
7 DR. REMICK: Yes. And risk even is a little bit 8 of a misnomer, because the way we think of risk now may be 9 safety. I don't know. We kind of think of risx as public 10 risk, at least I do. And it is certainly not public risk 11 related. It is safety related, though.
12 MR. JOHNSON: That's right.
l 13 DR. REMICK: It's a trivial point, perhaps. But
() 14 sometines a word can throw people off.
15 MR. JOHNSON: Yes.
16 MR. WARD: You are in disagreement. He keeps 17 saying it is related to public risk. And I think it is, if 18 you look at that equation. It depends what you plug in over 19 the consequences.
20 DR. REMICK: Yes. But I can't take one plant and 21 compare this with another on risk.
22 MR. JOHNSON: No. This is plant specific looking 23 at how is this plant doing. And just saying that if the 24 unavailability of some risk-significant, risk-important 25 safety systems is degrading, then that is something that
() Heritage Reporting Corporation (202) 628-4888
1 l
1 201 0 1 ought to flag the NRC management attention along with other 2 information.
3 DR. REMICK: Yes. I have no problem with risk-4 significant systems, that type of thing. But if it is just 5 risk, I.think of public risk. To me that'means exposure and G so forth. I start thinking tnat train.
- 7. .MR. JOHNSON: So this is just one aspect in that.
8 chain of events.
9 DR. REMICK: My only point is I guess selection of 10 words is very important.
11 MR. JOHNSON: Yes.
'12 DR. REMICK: So I don't think maintenance is 13 enough to describe what you are doing.
() 14 MR. JOHNSON: So that's why the term safety system 15 function trends.
16 DR. REMICK: Yes. Yes. Okay.
I 17 MR. CARROLL: Have you and your contractors 18 discussed some of these concepts with INPO?
19 MR. JOHNSON: I don't know about the contractors.
20 I have. I have periodically phoned up my counterpart at 21 INPO and we swap notes on who is doing what. That is about 22 the extent of it.
23 And also, I have visited there a couple of times.
24 They have visited here occasionally. So that we, in this 25 area at least, we keep each other apprised of what is Heritage Reporting Corporation (202) 628-4888
1 1
e 202 k- 1 happening.
2 But the reason -- well, the data are not made I
3 available to the NRC. At least when I have asked my )
l 4 counterpart from INPO would INPO provide this information to )
5 the NRC, the answer has been no.
6 DR. REMICK: I understand that. And if the agency i 7 needs an indicator of this type, I'm personally not 8 convinced of that yet one way or the other because I haven't 9 heard enough to know if it is needed, until I know now it 10 would be used. But even if there was a need for an 11 indicator demonstrated and INPO is not providing 12 information, if NRC did something, I still haven't heard a I
13 justification for doing something different than what
() 14 industry is already using and possibly is going to be used 15 worldwide, if these are very similar, what is the advantage 16 of what you are proposing over what is already being done, 17 even if NRC has to duplicate it on its own, because INFO 18 won't make that information available, is there is enough of 19 a benefit in doing something slightly different?
20 MR. JOHNSON: Probably not.
21 DR. REMICK: It is a question I have.
22 MR. JOHNSON: Yes.
23 DR. REMICK: So from your standpoint, you think 24 this is an idea indicator?
25 MR. JOHNSON: No, I don't.
(}
( Heritage Reporting Corporation (202) '628-4888 l
i 203
'I :
1 DR. REMICK: Okay.
2 MR. JOHNSON: This is a simplified indicator.
3 DR. REMICK: Close to ideal.
4 MR. JOHNSON: And it is.an improvement on the 5 existing indicators that NRC is using and it is intended as 6 a step, evolutionary improvement. This is a simple, as you 7 said, a simple indicator.
8 DR. REMICK: Yes.
9 - MR . JOHNSON: It is one that could be supported 10 with data from alternative sources, one of which could be 11 the INFO indicator. To simply use the INPO indicator would 12 give us 90 percent of what is in this indicator.
13 DR. REMICK:- Okay. Any further questions?
14 (No response) 15 DR.'REMICK: Shall we continue?
16 17 18 19 20 21 22 23 24 25 Heritage Reporting Corporation (202) 628-4688
{
204 O= l' DR. REMICK: Thank you. I bet you get many 2 questions about your name. Like if some people ask me what 3 my name is, I will say Remick and they will say what is your 4- last name.
5 DR. LOIS: My name is Erasmia Lois, and I work for 6 the Office of Research. And my topic is programmatic 7 . performance indicators. And I will start out with 8 indicators on maintenance, and then the training indicators 9 driven by the work done on maintenance.
10 (Slides shown.)
11 DR. LOIS: This work was done by SAIC.
12 Mr. Wranthall is with us today. And also contributors were 13 Joe Tragola and Peter Appeguiani among other people. We
( 14 have gone through the process for developing these 15 indicators. The SAIC work started with a literature search 16 on maintenance. First in order to identify important 17 attributes of maintenance that relate to safety and also 18 ddentify performance indicators, performance indicators of 19 maintenance.
20 And we did identify about 78 of those. Nuclear 21 plants use many indicators. IMPO is using maintenance 22 indicators. In addition to the NRC and its contracts, we 23 have identified a number of mechanical maintenance 24 indicators.
25 MR. WARD: Excuse me, they identified how many
(' Heritage Reporting Corporation (202) 628-4888
f-( 205
-V 1 indicators?
2 DR. LOIS: 78.
3 MR. WARD: 78, okay.
4 DR. LOIS: It could be 108 now.
5 MR. WARD: Okay.
6 DR. LOIS: So the first framework that was 7 developed as a process model. This is the picture of the 8 functional units in a nuclear power plant that shows also 9 flow paths amongst the different units. And they are used 10 by the analyst to examine influences among these units, and 11 whether or not they could identify some points where 12 measures can be developed.
13- The process model was developed on the plant level
() 14 as well as on the maintenance level. This process model 15 revealed that a nuclear power plant is very self-correcting.
16 It is a self-correcting entity. It is very difficult to 17 identify contributions from individual programs to plan 18 performance.
19 And therefore a conclusion came out of this model 20 that process indicators for regulatory use since the NRC as 21 an agency regulates this that they are very difficult to be 22 used, primarily because the information from one functioning 23 unit to another has many, many different ways to go. You 24 cannot plan it. And therefore the indicators were 25 differentiated or characterized as process indicators or
(} Heritage Reporting Corporation (202) 628-4888
206 f
1 output indicators.
2 Then frameworks were developed to characterize the 3 indicators versus the attributes of maintenance. And we 4 found that actually most of them tried to either capture the 5 quality, or the frequency, or the duration of maintenance, 6 and also tried to-capture some human factors issues. The 7 data availability was taken into account in evaluating these 8 indicators.
9 MR. CARROLL: Just so that I understand your 10 terminology, give me a simple example of a process indicator 11 versus an output indicator.
12 DR. LOIS: Right here. If you look at the very 13 first indicator, fraction of components under condition
() 14 monitoring, this is a process indicator. It characterizes 15 how the maintenance proce.ss is done within the plant. Its 16 data is very difficult for NRC to get it. This matrix here 17 shows how the different indicators, the 78 indicators, were 18 evaluated.
19 The first column is the data availability. The 20 big Xs show that the data is easily or possibly to be 21 obtained. O is output, and P is process. And some of the 22 output indicators most of the time, it is very difficult to 23 characterize distinctly as output or process. I mean they 24 are not totally exclusive ideas.
25 This column here has to do with the maintenance Heritage Reporting Corporation
(']
/
.\
(202) 628-4888
-r. 207 l
1 inspection tree developed by NRR to do maintenance 2 inspections. And we back to all of the elements of the tree l
I 3 and tried to identify whether or not some of the indicators 4 had a significant impact on safety related to maintenance 5 versus looking at the different boxes of the inspection and 6 associating the indicators with these maintenance boxes.
7 MR. WARD: Wait a minute, if you could slow up a 8 bit. The column labeled output process, you have got big Ps 9 and little ps there.
10 DR. LOIS: Yes.
11 MR. WARD: What does that mean?
12 DR. LOIS: I think that shows the determination, 13 yes. These are process indicators. If you could help me on
() 14 that, John.
15 DR. WRANTHALL: I am John Wranthrall. The O and 16 the P characterized whether it was a output or the process 17 related indicator. A bold or capital was a strong 18 association to process or output. A small one was a weaker 19 connection. It was the strength of connection between as we 20 assessed it the indicator and measures of process or 21 measures of output. So a bold figure or a capitalized 22 letter indicator a stronger connection than a lower case or 23 an unbolded letter.
24 MR. WARD: Okay.
25 DR. LOIS: Frequency, duration, and quality are
() Heritage Reporting Corporation (202) 628-4888
l' the .three important aspects of maintenance related to 2 safety. And the indicators try to capture either one of
'3 these qualities or sometimes associated with all.of them.
4 And also regarding human factors issues, work load, 5' . communication, training and administration, again the 6 analyst tried to express what kinds of human factors issues 7 these_ indicators attempted to measure.
~
8 MR. MICHELSON: What is the difference between a 9 big X.and a little x?
10 DR. LOIS: Again the same difference as P.
11 MR. MICHELSON: How about when there is no X at 12 all? No, no. Further down on quality, there are a number 13 of places. No, no, on quality.
() 14 DR. LOIS: Okay. This indicator is not an 15 expression of quality.
16- MR. MICHELSON: Okay.
17 DR. LOIS: This is primarily a quality indicator.
18 It does not express quality.
19 MR. MICHELSON: There is no relationship between 20 maintenance staff size and the quality of maintenance, is 21 that what it is saying?
22 DR. WRANTHALL: It is saying that that indicator 23 is not considered evidence of staff size. There may 24 ultimately be a connection. For example the amount of 25- rework in maintenance was seen as a strong expression of
() Heritage Reporting Corporation (202) 628-4888
209 0 1 quality, whereas simply the staff size was not seen as an 2 expression of quality.
3 MR. MICHELSON: There is no real discernable 4 relationship I guess is what you are saying.
5 DR. WRANTHALL: There may be an obscure or very 6' indirect relationship, but it was not seen to express a 7 strong one.
8 MR. MICHELSON: The same on mean repair time, that 9 there is no discernable relationship there either, 10 DR. WRANTHALL: Correct.
11 DR. LOIS: Having done that, some criteria were 12 developed to reduce the initial, and it have been called a 13 preference scheme.- Because actually there is no hard rules.
) 14 It is still an expert judgment. So preference was given to 15 indicators that relate to output as opposed to process, 16 because they are less easily manipulated and have a more 17 direct tie to safety. To quality related indicators for the R18 same reasons. And then the next preference was given to 19 indicators expressing frequency. And preference for human 20 factors issues associated with communications and 21 administration and supervision.
22 Data of duration was very important. And these
- 23. criteria here are the ones that the task group established 24 for the NRC's indicators.
25 From the 78, nine indicators were identified as Heritage Reporting Corporation
([ (202) 628-4888
~ 210 1 worth to explore and to validate. Wrong unit or wrong 2 train, scrams due to test and maintenance, and ESF 3 actuations while reactor was at power based on maintenance.
4 This indicator I am going to talk about in detail and define 5 exactly.
6 Briefly here the DPL, daily power loss, is defined 7 as the maximum dependable capacity minus the average daily 8 power produced normalized over a maximum capacity.
9 DR. REMICK: Could you give me that again, 10 maximum?
11 DR. LOIS: Dependable capacity.
12 DR. REMICK: Dependable capacity.
13 DR. LOIS: Minus the --
() 14 DR. REMICK: On a yearly basis or what?
15 DR. LOIS: Daily.
16 DR. REMICK: It is a maximum daily?
17 DR. LOIS: Yes.
18 DR. REMICK: Okay.
19 DR. LOIS: Daily.
20 DR. REMICK: And that would not be a 100? I guess 21 that I do not understand.
22 DR. LOIS: It is megawatts. It is the power used 23- in the grid, the ideal power. j 24 DR. REMICK: Okay, all right.
25 DR. LOIS: It comes from data, NUREG-020.
() Heritage Reporting Corporation (202) 628-4888
211 7 3 x.>
2 DR. LOIS: This is the average daily power. This 3 is megawatt hours per day divided by 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. And 4 normalized over a maximuu dependable capacity.
5 MR. CARROLL: It is one minus capacity.
6 DR. REMICK: One minus capacity.
7 MR. WARD: That is a funny name, daily power.
8 MR. CARROLL: In fact of course INPO is getting 9 away from the capacity factor as a performance indicator, 10 because there are too many people cheating. HDC for a 11 cooling tower plant gives you a big break during the cooler 12 weather months of the year where you can run at 107 percent i 13 capacity factor. So they have gotten into unavailability.
() 14 DR. LOIS: The rationale behind this indicator is 15 that you would like to have your plant in a near ideal 16 condition. So we have tried to develop the departure of 17 power generation from its ideal conditions. That is 18 delta P, and normalize it and train it. I am going to talk 19 about this more.
20 For these indicators, data was collected for eight 21 plants for three years. Some of these data are LAR data, i 22 This for example was a joint effort with AEOD with plant 23 data.
24 DR. REMICK: Let me ask a question, on MDC do you 25 factor in the fact that perhaps the need is not the
() Heritage Reporting Corporation (202) 628-4888
v
-s 212 T
1 _100: percent of full power?
2 MR. WARD: Base loading.
3- DR. REMICK: Right.
4 DR. LOIS: This is net, MDC net.
5 DR. REMICK: That still does not answer my 6 question. If these were all based loaded plants, then I can 7 see it.
8 But suppose that they only request that the plant 9 run at 50 percent that day, is that going to be called a 10 daily pcwer loss?
11- DR. WRANTHALL: If I could answer that. This 12 indicator is still really under development. And right now 13 we are trying it out only on base loaded plants. When
'14 Erasmia gets to the part of the presentation that deals f )z 15 specifically with this indicator, what-turns out is that we 16 are concerned with the patterns of outages really caused by 17 equipment failures.
18 So that for plants that were not base loaded, we 19 would go back and modify the definition accordingly to look 20 for losses of power due to equipment failures usually in the 21 balance of the plant. But right now the indicator is still 22 under development, so we do not have a final answer on that.
23 MR. CARROLL: That is not part of the NRC's 24 regulatory purview.
25 DR. WRANTHALL: Right. What we are looking for is
() Heritage Reporting Corporation (202) 628-4888
f
= 213 evidence of the quality of maintenance.
1 2 DR. REMICK: Please continue.
3 MR. MICHELSON: Are these items on a plant basis 4 or a unit or what?
5 DR. LOIS: A plant basis.
6 MR. MICHELSON: So the mean time between repair, 7 what is that talking about, what repair, any repair in the 8 plant at all scould be included in the mean time between 9 repair?
10 DR. LOIS: I think that it concerns vital 11 components. It is not everything in the plant.
12 MR. MICHELSON: In other words, there will be a 13 set of prescribed components and you will track the mean
() 14 time between failure of that set for that set?
15 DR. LOIS: Exactly.
16 MR. MICHELSON: Okay.
17 MR. WARD: You mean that you can scrt of 18 conglomerate somehow mean times for different kinds of 19 systems, or is there a mean time for a given system?
20 DR. LOIS: The way that this was done is I guess 21 it was on an aggregated basis. It was not per system. It 22 was data collected from plant data demonstrating mean times 23 between failures. I am sorry, mean times between repairs 24 for a number of different systems.
25 MR. MICHELSON: Well, you have got to define
() Heritage Reporting Corporation (202) 628-4888
l
- 214
~j 1 repair there too very carefully. Because equipment is 2 sometimes repaired before it fails, and sometimes it is 3 repaired after it fails and so forth.
4 DR. REMICK: Preventive maintenance repairs for q 5 example.
6 DR. LOIS: We did not take this indicator further.
7 I know that there are a lot of difficulties regarding this 8 indicator. However we do not have enough evidence that 9 relates to safety, so that we could work on that further.
10 The statistical correlation that we did was not conclusive 11 on that.
12 MR. MICHELSON: Now the wrong part events, if one 13 event consists of finding different wrong parts, that is
() 14 etill just one event, is that right?
15 DR. LOIS: The wrong --
16 MR. MICHELSON: The wrong part events you have got 17 in there, a .4 umber of them.
18 DR. LOIS: Yes.
19 MR. MICHELSON: That is not the number of wrong 20 parts that I have found, but that is just the number of 21 times that I found wrorsg parts I guess.
22 Is that what that is?
23 DR. LOIS: I think that it is every time that you 24 find a wrong part that it is an event.
25 MR. MICHELSON: That is an event?
I)
Heritage Reporting Corporation (202) 628-4888
r~s 215
(.) 1 DR. LOIS: Yes.
l 2 MR. MICHELSON: But if I go through and do an 3 audit and I find that there are fifty different wrong parts 4 and it might take me three days to make that determination, 5 that is still reported as one report and one event?
6 DR. LOIS: Yes. So the data consideration there 7 is important.
8 MR. MICHELSON: That is a strange number in fact.
9 DR. LOIS. It is not reported to the NRC. These 10 indicators do have data definition problems. It just 11 demonstrates that we should look into output indicators as 12 opposed to process indicators. And when we did the 13 statistical analysis based on whatever data we had, in
() 14 actuality we did not find a lot.
15 MR. MICHELSON: What is a realignment error?
16 DR. LOIS: It is when after maintenance that you 17 have to realign. Assuming that you tear down a valve and 18 you have to put it back, quite often you put it in the wrong 19 position.
20 MR. MICHELSON: But it is not a calibration type?
21 DR. LOIS: No.
22 MR. MICHELSON: That requires a screw driver 23 sometimes too, and you just wind it up wrong. But this is 24 strictly valve alignments and that sort of thing?
25 MR. MICHELSON: Or similar events.
() Heritage Reporting Corporation (202) 628-4888
l 1
- 216
~
1 DR. WRANTHALL: It is strictly system 2 configuration mistakes.
3 MR. CARROLL: It is not what a maintenance person 4 would think of as an alignment.
5 DR. WRANTHALL: It is a system configuration 6 problem.
7 MR. CARROLL: How dces that get to be a 8 maintenance problem? Realignment after maintenance is done 9- by operations as part of the maintenance process I guess.
10 DR. WRANTHALL: Yes. In the working definition 11 that was adopted I think across all current NRC maintenance 12 programs, then that task be it done by operators or 13 maintenance people is part of maintenance.
() 14 MR. MICHELSON: So realignment errors made during 15 operation is not counted here, but just if it is made in 16 conjunction with maintenance, is that the idea, because 17 there are a lot of realignment errors that are strictly 18 operational where no maintenance was even done. You just 19 lined something up wrong.
20 DR. WRANTHALL: It is appropriate to the 21 maintenance activities.
22 MR. MICHELSON: It seems totally inappropriate to 23 maintenance when the operator screws up and puts the valve 24 in the wrong position.
25 DR. LOIS: So what you suggest is that it could be
()fs Heritage Reporting Corporation (202) 628-4888
1 'in operations as opposed to maintenance.
e 2 MR. MICHELSON: I can understand operations in 3 conjunction with a maintenance activity coald be called 4 maintenance, but not in conjunction with a normal 5 operational activity.
6 DR. WRANTHALL: It was done in association with 7 maintenance.
8 DR. LOIS: One of the difficulties of this program 9 are the contributions from ind!vidual programs. The 10 maintenance and operations is so tightly connected, it is 11- very difficult to differentiate.
12 In order to validate the indicators which means to 13 establish relationships, we used statistical analysis and
() 14 also engineering analysis. The statistical analysis was a 15 cross-correlation to test relationships between the 16 indicator d.ata and measures of safety, and estimate lead or 17 lag time between the indicators and measures of safety.
18 Measures of safety were used as current indicators.
19 One of the biggest problems in this work was data.
20 For most of these indicators we did not have well defined 21 data. And even the ones based on LER data, the density was 22 such that it did not allow to make meaningful analysis.
23 These two came up with relationships, significant 24 relationships to safety criteria with the ESF actuation as a 25 leading indicator.
() Heritage Reporting Corporation (202) 628-4888
7s 218 i
~
1 MR. CARROLL: What is MOS again?
2 DR. LOIS: Measures of safety.
3 MR. CARROLL: Measures of safety.
4 Now if all of this is true, why is the staff 5 continually putting plants on the problem plant list when (
6 they set record performances in terms of generation? That 7 seems to be a trend here these days. I do not expect you to 8 answer the question.
9 MR. MICHELSON: An irresistible comment.
10 DR. LOIS: This picture here demonstrates the 11 statistical analysis results on ESF. These are a couple of 12 plants, and you can see that you have a trend of ESF 13 actuations. You have some other plants with no or very few O)
(_ 14 ESF actuations. These two plants, LaSalle 1 and 2, you can 15 see that although they are the same utility that you can see 16 a difference in trends.
17 MR. WARD: What two plants?
18 DR. LOIS: This is LaSalle 1 and 2, and this is 19 Brunswick i d 2.
20 MR. WARD: ESF for LaSalle 1 and 27 21 DR. LOIS: Yes. So in order to understand and to 22 make sure that this statistical analysis tells us something 23 about what we call the engineering analysis. And the 24 engineering analysis was based on looking at the earlier 25 descriptions to see what do the licensees report as l () Heritage Reporting Corporation (202) 628-4888 l
l l
l I
219 L
)
I corrective actions.
2 And also going back and reading the SALP reports 3 on maintenance. And since this is a retrospective analysis, f
4 comparing the text i the SALP reports with what we find as 5 ESF events. There is a consistency between these patterns 6 that I showed before and SALP reports. As a matter of fact 7 for Brunswick, it seems that that part preceded the SALP 8 findings.
9 Another interesting thing was that plants that 10 have this consistent pattern of ESFs usually report as 11 corrective actions counseling and disciplining. And that is 12 nicely demonstrated in these two events at LaSalle. This 13 unit when you look at corrective actions, you can see that
() 14 it is discipline or counseling this unit with the report on 15 procedure improvement modifications, design modification and 16 sometimes discipline.
17 MR. CARROLL: You are using the word discipline as 18 I would use it, that you give the guy who screwed up a week 19 off without pay?
20 DR. LOIS: That is right.
21 MR. CARROLL: Okay. That is sort of an archaic 22 concept of course. We now talk about positive discipline.
23 DR. LOIS: However it seems that plants that use a 24 lot of do not do it again that they never go back and look 25 at root causes to find out why they have these repeated
() Heritage Reporting Corporation (202) 628-4888
1 patterns of events. It could be that some of it may be 2- designed based or procedure based. And therefore these 3 indicators seom to be able to differentiate plant 4 performance, attitude of plant performance.
5 MR. CARROLL: Tell me again what you see in 6 following Plant E and Plant F, what is that telling you?
7 DR. LOIS: That Plant E has many more events, 8 ESFs, than plant F, a consistency of frequency. And this 9 plant if you look at the earlier correctiva action, they 10 report discipline and counseling consistently.
11 MR. CARROLL: Disciplining and counseling?
12 DR. LOIS: Yes. Where they discipline the guy who 13 made the mistake and created the ESF and counseled him not I) 14 to do it again.
15 MR. CARROLL: Now if I happen to know that in 1987 16 that Plant F was out for refueling.for four or five months, 17 I am not sure that I would draw that conclusion.
18 DR. LOIS: These indicators of ESFs is for a 19 reactor at power.
20 MR. CARROLL: What do I put in your plot there if 21 the unit happens to be out for refueling the whole quarter?
22 DR. LOIS: It will not be there. The ESFs during 23 refueling were not. Is that correct, John?
24 DR. WRANTHALL: That is correct.
25 MR. MICHELSON: Maybe the maintenance was lousy
() Heritage Reporting Corporation (202) 628-4988 l
uit .L r. -
221 RP 1 though. You do not know that from this. Could you repeat 2 again which ones hr the bad SALP maintenance ratinge.
3 DR. LOIS: Brunswick.
4 MR. MICHELSON: No, by E and F. I do not care u
5 Lbout the name of the plant.
6 DR. LOIS: .E and F.
7 MR. MICHELSON: Not A, Plant A, was that a bad 8 SALP? It looks like it was one of the worse as to the 9 number of actuations.
10 DR. WRANTHALL: Plants A and B are sister plants, 11 and both had consistent poor ratings.
12 MR. MICHELSON: But poor you mean what?
13 DR. WRAliTHALL: I mean in the narrative text we
() 14 did not go by the numerical ranking. We went by the 15 description, the assessment of the problems, and d-16 narrative description of deficiencies.
~
17 MR. MICHELSON: Were any of these Category 3 SALP 18 ratings?
19 DR. WRANTHALL: Off the top of my head, I do not 20 remember.
21 MR. MICHELSON: Were you just trying to do a 22 tuning of Category 2 as to what it meant by reading the 23 words, is that the idea?
24 DR. WRANTHALL: Regardless of the numerical 25 rating, we went essentially by the description of
() Heritage Reporting Corporation (202) 628-4888
222 0 1 deficiencies and the problems that had arisen in 2 characterizing the quality rather than the SALP rating 3 per se.
4 DR. WRANTHALL: If I could answer your earlier 5 question, this is one summary chart. In the report that is 6 being produced, we did plot events identified where they 7 were plant duration, refueling outages, and other periods 8 off line. And we did not find any significant different 9 trends by taking account of that.
10 MR. CARROLL: But you might conceivably?
11 Dh. WRANTHALL: It would be possible, but we did 12 try to look for that. The one case I think is Plant D which 13 for other reasons had a very extended outage. It is one of
() 14 the plants that had a major incident. And there were very 15 brief periods when the reactor was at power, and during one 16 of those periods it had two events. So in looking at those 17 additional charts, you would see such patterns.
18 MR. CARROLL: But not by looking at this onel 19 DR. WRANTHALL: This is simply a summary across 20 plants. And particularly the differences that we saw in A 21 and B, and the trends in fact in E and F.
22 MR. CARROLL: Summary charts, because people draw 23 conclusions from them, you would like to take that kind of 24 thing into account. I am not sure how you do it.
25 MR. MICHELSON: Is an ESF actuation one LER, is l
() Heritage Reporting Corporation (202) 628-4888
.223
. \_)
I that the way that you are counting numbers of actuations?
2 Because sometimes a LER might have three or four actuations 3 in a two or three day period. And it is written up as one 4 LER, but actually there were several ESF actuations, and 5 they finally got their problems straightened out. But how 6 do you count?
7 DR. WRANTHALL: In going back through the LERs, 8 not just on a LER count but actually reading the text, if 9 the plant had got back to power and had stabilized ared then 10 an ESF event occurred again but it was counted, if it was 11 while the reactor was shut down or during testing at 12 start-up, then it probably would not be counted.
13 MR. MICHELSON: So you are only at full power?
() 14 DR. WRANTHALL: No. Basically at about five 15 percent.
16 MR. MICHELSON: Five percent and up.
17 DR. WRANTHALL: I do not remember to be honest 18 that when we did this that when we came across events that 19 there may be more than one that was questionable whether it 20 should be counted or not.
21 MR. MICHELSON: I can recall LERs in which they 22 got up around 10 or 15 percent and tripped out. They went 23 back and tried again, and the LER described how many tries 24 they made before they solved whatever their problem was. I 25 just wondered if that was counted as one event here or one
() Heritage Reporting Corporation (202) 628-4888
224
-0. 1 actuation rather.
1 2 DR. WRANTHALL: Again these were due to test and 3 maintenance errors.
4 MR. MICHELSON: I thought that maintenance 5 everywhere is being defined including these operator errors, 6 but maybe not.
7 DR. WRANTHALL: In this particular case, it'was a 8 testing function or a maintenance activity in the sense of 9 repair and periodic testing.
10 MR. MICHELSON: Not as opposed to operating?
11 DR. WRANTHALL: Correct.
12 MR. MICHELSON: Okay. Maintenance has got a 13 different definition here.
.() 14 DR. REMICK: The bullet up there says " appears to 15 precede SALP assessments." I am not sure what the 16 significance of that statement is, but I would just assume 17 that because SALP is a retrospective look at the past twelve 18 months or eighteen months. So I would think that the SALP 19 asressment would come out after any kind of active 20- indicator. I guess that I do not find that surprising.
21 DR. LOIS: No, it is not. However the thrust of 22 this program is to develop some indicators that will give 23 SALP results. In actuality, these indicators are indirect 24 indicators of safety. They are not directly tied with 25 safety. And this is more to present mind sets of plants or Heritage Reporting Corporation O- (202) 628-4888 L _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - _ _
1 l
l
,- 225 5
I plant managers, how do they treat small events as opposed to 2 saying that the risk in the plant increases or decreases.
3 MR. CARROLL: How do you make a judgment about 4 that?
5 DR. LOIS: Going back to this picture, it seems 6 from the analysis that we have done that these plants here 7 are types of plants that are not known for getting into 8 problems and correcting it. This again is a retrospective 9 analysis, and it validates what we know for these specific 10 plants. So what we are saying now is had we done chis 11 23 years ago that it could be that this situation would have 12 attracted the NRC's attention earlier.
13 MR. WARD: I am still puzzling over E and F and
() 14 what you attribute the difference there to. Those are two 15 identical units at the same site, and you attribute the 16 difference there to different personnel practices as far as 17 discipline and counseling people who made errors. And I 18 guess that I found it hard to believe that they would be so 19 different at those two units.
20 DR. LOIS: John, can you elaborate on that?
21 DR. WRANTHALL: I think that the biggest 22 difference is actually more between A and B and the other 23 plants. Plants E and F as I recall were not that different 24 in their practices. They were different through time which 25 was interesting, and that is in the period of 1986 to 1987.
Heritage Reporting Corporation I f~l)
' (202) 628-4888 l
1
f 226 1 There seemed to be a period when agai: discipline was 2 raising its head. The contrast that I have is installing
~
3 barriers or changing the design or changing the procedures.
4 Then there was an evolution away again from that.
5 So as I recall Plants E and_F that it was more of a change 6 through time. I do not recall a big difference in E and F 7 separately.
'8 DR. LOIS: However in the analysis that you have 9 in the report.
10 DR. WRANTHALL: For those events that occurred.
11 DR. LOIS: The corrective action here is reported 12 as consistent.
13 DR. M.*,ANTHALL: There were no challenges in the
() 14 ESF situations in 1987. Therefore at that time when they 15 were reporting discipline that it appeared to be a higher 16 element at E.
17 MR. MICHELSON: Which plants were E and F again 18 DR. WRANTHALL: LaSalle 1 and 2.
19 MR. MICHELSON: It is not different plants. Those 20 are just units, two units at the same plant, not two 21 different plants.
22 DR. WRANTHALL: That is correct.
23 MR. MICHELSON: It says Plant E and Plant F which 24 suggested to me two different plants.
25 DR. LOIS: No , these are two different units.
() Heritage Reporting Corporation (202) 628-4888
L' 227 l :- .
1 MR. MICHELSON: Two different units. They are 2 obviously the same personnel serving the two units.
3 Is Plant A and Plant B the same situation too, two 4 units at one plant?
5 DR. LOIS: Yes.
6 DR. WRANTHALL: Plants C and D.were not connected.
7 There were additionally two other plants that had no events 8 across the whole period.
9 MR. MICHELSON: I think that.your indicator does 10 not do much. The same people serving two.different units !
11 are getting different performances out of the units, right, 12 in the case of E and F for instance, and in the case of A 13 and B also? ,
() 14 MR. WARD: Well, they might not be the same 15 people. We do not know for sure how those two plants are 16 organized. l 17 MR. MICHELSON: No, but.it is pretty unlikely that 3 18 they have got a technician for A and one for B. l 19 MR. CARROLL: They are sister units.
20 MR. MICHELSON: Yes.
21 MR. WARD: But maybe by not looking in nore detail I
22 at how they are organized that you are missing the important 23 point.
24 DR. WRANTHALL: We did look at Plants A and B to 25 see that if the pattern looks different was it statistically Heritage Reporting Corporation l
-( ). (202) 628-4888 l
228 1 significantly different and it was not.
2 MR. WARD: Oh, okay. So E and F?
3 DR. WRANTHALL: That le for A and B. Excuse me, 4 we only looked at A and B as to whether those were 5 different. Because superfic3 ally they looked different.
6 MB. MICHELSON: There were not statistically 7 significant differences in E and F.
8 MR. MARD: I can accept that, but I thought that I 9 heard an explanation for the difference between E and F 10 earlier a few minutes ago.
11 DR. LOIS: However in developing leading 12 indicators of safety,. we were working in the SALP area of 13 plant performance.
() 14 MR. MICHELSON: Now SALP is done on a unit basis 15 or a plant basis?
16 MR. WARD: On a plant basis.
17 MR. MICHELSON: Then SALF ratings are the same for 18 Plant E and Plant F. The ratings were the same, but the 19 performance was different in terms of the indicator.
20 DR. LOIS: The only difference if you look at the 21 qualitative analysis is that you will see a small difference 22 that for some reason for Unit 1 that the maintenance were 23 always disciplined or counseled. They cerrected procedures 24 and they did some mods. So that probably explains the 25 difference.
Heritage Reporting Corporation
(]) (202) 628-4888
i
, 229 t
il MR. WARD: We have gone back around again.
L 2 MR. MICHELSON: Full circle..
1 3 DR. LOIS: I do not know. Let's move on to the 4 daily power level. My viewgraphs do not correspond to what 5 .I have here. We just changed it or improved it. Again it 6 is defined as the maximum dependable capacity minus the 1
7 average daily power level.
8 And the rationale for'that is that it reflects 1
9 actual. To develop this indicator, it was figured out from 10 the statistical analysis that identified the relationship of 11 this indicator with equipment failures. And we feel that it 12 may make sense since it is objective and it relates to BOP 13 components for system failures. It is operational instead
() 14 of stand-by.
15 And it will drive the maintenance activities on E16 BOP to be as important or compatible with the safety system 17 maintenance. If it turns out to be an indicator related to 18 safety, it would be cost effective for the utilities.
19 DR. REMICK: I just do not see how on a practical 20 basis that you are going to weed out those plants not base 21 loaded or because of environmental concerns they have 22 restrictions and those types of things.
23 DR. LOIS: Let me elaborate a little bit on that.
24 DR. REMICK: All right.
25 DR. LOIS: What we tried to do here is to explore
() Heritage Reporting Corporation (202) 628-4888
- 230 1 k_ if it can be used as a signal generator for plant 1
2 performance. So in actuality this variable here is becoming 3 symmetrical with respect to the horizontai axis by minus the 4 average DPL over the period of performance that we are 5 considering. Then we use numerical methods and derive the 6 power spectrum density profiles.
7 This is a diagram for an annual one, for one year, 8 where zero is this line, and it does indicate that this 9 indicator behaves like a signal. And you can see that in 10 this here that one could identify svents occurring in the 11 plant. Some of them are safety related and some of them are 12 in the BOP part of the plant. So by doing the analysis, you 13 do get the power spectrum density.
()
p 14 DR. REMICK: Now what caused the peak here in this 15 case, was it hypothetical, was it actual?
16 DR. WRANTHALL: The low frequency component there 17 is because in the year that we started doing the analysis 18 that there was a refuelitj outage which is consequently like 19 a square wave of a relatively long duration. If you go back 20 to the previous chart, it ic essentially this first period 21 in roughly the first sixty days of that period. There was 22 one trial start-up at about the thirty day mark.
23 But essentially it is that duration of the 24 refueling outage that is causing that peak of the low 25 frequency component. This is an indicator that we are still Heritage Reporting Corporation f' ') (202) 628-4888
f-231 wrestling with and really doing research on.
2 MR. CARROLL: What are the units of that axis?
3 DR. WRANTHALL: Of the vertical axis?
4 MR. CARROLL: Yes.
5 DR. WRANTHALL: It is a normalized scale of power 6 loss. It is essentially percentage power lost except that !
7 the value roughly at the .7 is the equivalent of complete 8 shutdown. In the equation we had an adjustment factor of 9 the minus average daily power level, ADP.
10 That term is in there simply because the numerical 11 processing method requires that the total integral of that 12 ' distribution be zero. The scale is best-described as an ,
i 13 artifact of the analysis. But the plateaus at the top are
() 14 complete shutdown. The plateaus on the bottom are at full 15 power. !
1 16 MR. CARROLL: And again the units on both axes?-
17 DR. WRANTHALL: Are not an effect of the analysis. ;
18 MR. CARROLL: So I do not have to know what they )
19 are?
20 DR. WRANTHALL: Right. If I may mention why we 21 took this approach. We were concerned for three reasons.
22 One is that we were concerned with plants that appeared to 23 have a pattern of very frequent recurring needs for repairs, 24 and a very short duration back to power and then a very 25 short duration of power and then back out again. Other O Heritage Reporting Corporation (202) 628-4888 1
b
,gss 232 V 1 plants in a qualitative sense seemed to take time to make 2 repairs and then come back to power and stay at power.
3 We were interested in trying to differentiate 4 patterns like that in a way that you could apply some 5 technique and get a rating out of it. One approach was to 6 look at a content of signal data for want of a better word.
7 Could you discriminate high frequency versus low frequency 8 behavior.
9 The first two reasons was looking for 10 differentiating those plants. But also we were researching 11 whether there was almost a central tendency in shutdowns in 12 the balance of plant. Because the balance of plant system 13 is to some degree a large single train system. And
() 14 therefore independent failures occur at random, but does the 15 system as a whole exhibit any kind of central tendency, a 16 recurring frequency of shutdown.
17 We were looking to capture whether that second was 18 true and first of all could we differentiate the plants.
19 That is why we went looking at the signal content in those 20 terms in terms of this power signal.
21 DR. LOIS: And the plans for the future is to 22 examine this concept between outages, between the refueling 23 times I am sorry, and identify whether or not there exists a 24 power spectrum density signal. And if it did exist, find 25 out how that could serve as an indicator. If it produced a Heritage Reporting Corporation r(] (202) 628-4888
233 1 signal that the NRC could use as an indicator. Of course 2 this data, we do not have to collect any, we have it. And 3 it is an unobtrusive indicator, an overall plant indicator.
4 That concludes my presentation on maintenance.
5 On training this has been an effort going on for 6 awhilo. It started out with maintenance, but it was driven 7 mostly by conclusions that we derived from maintenance.
8 Again output of training are desirable indicators for the 9 NRC. So nine candidate indicators were identified and these 10 were tested with measures of safety.
11 However the results are really inconclusive. Even 12 the statistical correlations do not make any sense. So what 13 we tried to do is go back and examine carefully training
() 14 programs of fuel plants and look at the process indicators, 15 training process indicators, as well as output indicators.
16 MR. MICHELSON: What are examples of training 17 process indicators, if I understood your term correctly?
18 DR. LOIS: For example we know that plants have 19 their own training and have feedback. One they have the 20 experience of an event, they either report it to the NRC, 21 mostly operator error, and they put it on training. And the 22 next time everybody is trained on the event. That is a 23 characteristic of a good training program.
24 If we can have axes for this data, we can see what 25 plants do it and what plants do not do it. It is one way to
(}
Heritage Reporting Corporation (202) 628-4888
234 D'~' 1 experience or learning from their own events.
2 MR. MICHELSON: So you find out that they do it, 3 what ck) you know then as far as an indicator?
4 DR. LOIS: I hope that that would indicate that' 5 they have an effective training-program.
6 MR. MICHELSON: I thought that the better 7 indicator of whether it is effective or not is how'many of
- 8. these experiences that they have to keep feeding back.- That
'9 would be more of a measure than just the fact that they have 10 a procedural method of doing it. Just having the existence 11 of a procedure that-says when you make a mistake that you L
12 feed it back, that alone does not mean anything.
13 DR. LOIS: We have operator errors, errors of 14 ESFs, or scrams due to the operator. However there are very (f
~
15 few events at least reported to the NRC that we can make any 16 sense out of it.
17 MR. MICHELSON: What you are doing then is you 18 look at how many operator errors caused the ESF actuations 19 and then you-look to see whether they have a feedback 20 program and if they did not?
21 DR. LOIS: No , I am not saying that. What I am 22 saying is that we are going to go back into the plants and 23 look at training programs and see the different ways they 24 use for effective training and come up with some ideas for 25 training effectiveness.
Heritage Reporting Corporation (202) 628-4888 1
l l ._ _ - ._____________- -
@,.i __ 235
"# 1 DR. R.YAN: Erasmia, if I could just try this.
2 This goes back to the overhead that I put up on the 3 left-hand column where we said that we do not have the kinds 4 of data available to us, especially in the amounts to C) any 5 kind of statistical analysis between that left-hand column 6 and the safety measures.
7 And I think that what Erasmia is saying is that 8 what we are hoping to do, and now I take you to the long 9 term ideal that I talked about originally, is to be able to 10 go in and not simply say that somebody believes because the 11- reported a LER or some kind of engineering analysis that we 12 have done that we associate some kind of intermediate output 13 we will say in the area of maintenance.
() 14 What we are interested here is going in and taking 15 a look at maintenance programs and what we know from other 15' areas about what constituted good or bad maintenance 17 programs, the attributes. To get that kind of data and try 18 to do some statistical analysis between those attributes and 19 those criterion measures.
20 MR. MICHELSON: That slide that you presented 21 talked about programmatic performance indicators.
22 MR. ROSS: Right.
23 MR. MICHELSON: And I am trying to search out what 24 is the indicator that you are thinking that you are going to 25 find or search out, how does that relate to this?
(']
N-Heritage Reporting Corporation (202) 628-4888
=
g 236
Okay. Well, Erasmia indicated that
- 1. DR. RYAN:
2 .there is belief that one attribute of an effective training 3 program is this feedback.
4 HR. MICHELSON: You mean the existence of a 5- procedure?
6 -DR. RYAN: What we are talking about is that we-7 start out and we train you on day one, and you go out and 8 you collectively -- the people who train have some 9 experiences. That some kind of an event occurs. In some 10 instances-that never gets fed back into the training for the 11 other people, and you when you go back through retraining.
12 So these events continue to happen. I am not talking here 13 specifically necessarily about a nuclear power plant.
14 We know some things about what constitutes
-( )
15 effective versus non-effective training programs when we 16 look at the outputs. In one case, the events keep occurring 17 over and over again. And one of the things that we find is 18 that experience is never fed back into the training.
19 MR. MICHELSON: I can understand a number of these 20 occurrences as being an indicator, but I do not understand 21 the existence of a program being an indicator.
22 DR. RYAN: Well, if you are interested in 23 prevention, you would very much at some stage of the game 24 would like to be able to statistically characterize a 25 program with certain attributes, each attribute having
Heritage Reporting Corporation
' (A (202) 628-4888
237 77 1 demonstrated some kind of a real relationship with whatever 2 the criteria may be.
3 HR. MICHELSON: By some kind of accounting 4 process.
5 DR. RYAN: That is right. And we would like to be 6 able to go in and focus on the process and not be reacting 7 to the outputs. That is sort of an ideal for the reasons 8 that I mentioned earlier. But what Erasmia is getting at 9 here is the stage in the research where we can go in and 10 start looking at the actual attributes of the program, the 11 process indicators if you will, the feedback here, and then 12 try to find out whether or not that really does relate to 13 the criterion measure that is selected in the case of a r
( 14 nuclear power plant.
15 And that is basically what we are saying. That is 16 the ideal. Right now we are only at the point where we are l
l 17 able to take a look at what we are calling intermediate 18 outputs inferring things about them, maintenance, training 19 and whatever and doing a statistical analysis between the l
20 intermediate output, the daily power loss, or the ESF 21 actuation and these things that we are calling safety 22 criteria.
23 DR. REMICK: Please continue.
24 MR. CARROLL: Before you take that slide down.
25 DR. REMICK: Good try on my part.
Heritage Reporting Corporation
/~)
(202) 628-4888 i .
)
238
[D 1 MR. CARROLL: About a third of the way down, 2 number of scrams at start-up, and then you have some 3 mathematics in the bracket.
4 Can you explain that to me, what is p?
5 DR. LOIS: I think for small power. I am sorry.
6 I am pretty sure that there are a lot of typos.
7 DR. LOIS: Small power, how is that defined? It 8 is less than one percent, or one percent or something like 9 that.
10 DR. WRANTHALL: I think that we took it in the 11 draft report as five percent, between zero and five percent.
12 MR. CARROLL: Between zero and five percent?
13 DR. WRANTHALL: I do not remember exactly now.
() 14 MR. CARROLL: So you were looking for scrams at 15 start-up.
16 DR. WRANTHALL: It was before the plant was really 17 in a truly balanced state.
18 MR. CARROLL: Okay.
19 MR. WARD: This would be a good time for a break.
l 20 DR. REMICK: It probably would. All right. It is l
l 21 so ordered. A ten minute break until 3:15.
22 (Whereupon, a recess was taken.)
23 DR. REMICK: All right. Could we please continue.
24 DR. LOIS: This part of the presentation will 25 describe efforts to develop indicators of organization.
(' Heritage Reporting Corporation (202) 628-4888
D l
239
'O 1 DR. REMICK: We are having a little bit of trouble ,]
1 ;
- 2 hearing you at the moment, i 3 DR. LOIS: Again I say this part of the' 4 presentation will describe. efforts of identifying indicators 5 of organizational effectiveness. This work has been done or 6 it has been led by the University of Minnesota. However 7 Wayne State University and American University as well are 8 participating in the work. Dr. Thurber of American.
9 University is with us today, and he will assist me on the
- 10 details of it.
11 It started out again with the development of a 12 framework of linking organizational factors to safety. It
'13 ' integrated and looked at the academic perspectives, the
() 14 perspectives of the academia regarding why organizational 15 factors are important for safety as well as for the industry 16 perspectives and the NRC perspectives.
17 Dr. Thurber interviewed a number of NRC experts 18 asking them what from their perspective are the important 19 aspects of organization with respect to safety.
20 This is a pictorial representation of a framework.
21 It is built on NUREG/CR 32.15 that was developed in 1985.
22 It depicts organizational factors as setting the 23 preconditions for safety through thebe intermediate 24 outcomes.
25 The important organizational factors are the Heritage Reporting Corporation O (202) 628-4888
E
.240
.O-1 environment, the context, the organizational governance, 2- organizational design, and emergent processes. I am going 3 to define a little bit each one of these.- And the 4 intermediate outcomes could be characterized as efficiency, 5 compliance, quality, and innovation.
6 Under environment'one would read the national 7 culture, or regulation, or self-regulation, cwnership, 8 control, outside dependence. Under context would be staff 9 size, budget, technological sophistication of the plant, or 10 technological variability.
11 The organizational governance deals with the way 12 that authority is distributed into the organization.- And 13 the study is characterizing it as traditional, modern or
() 14 federal, in depicting the different ways that the authority 15 has been distributed in the organization.
16 MR. CARROLL: Give me an example of traditional, 17 modern and federal.
18 DR. LOIS: Traditional is the one where authority-19 has been defined by the position. While in modern usually 20 authority has been defined by competence where the hierarchy 21 has the responsibility to coordinate the activities. While 22 in the federal authority is more defined by professionalism, 23 and again the hierarchy has the responsibility to assure 24 coordination.
25 MR. WARD: I am not sure that I get the difference k
() Heritag'e Reporting Corporation (202) 628-4888 I
I I I
L E <
241 f(
.1 between modern and the federal, the difference was 2 professionalism versus competence in the way that you 3 described it?
4 DR. LOIS: I am saying that modern can be 5 characterized as -- it actually describes the way that 6 management works in a specific organization. And federal 7 organizations usually are governed by.this ideal of 8 professionalism. And the hierarchy usually coordinates the 9 activities as opposed to this traditional type of management 10 where position defines and takes the responsibility for 11 everything.
12 MR..MARD: That is.all.right.
13 DR. LOIS: The emergent process describes the
() 14 dynamics in the organization in the day to day work, and 15- also assists them to react to an unanticipated condition.
16 For example, organizational culture, goal setting, decision 17 making are examples of emergent processes. These are 18 focused on emergent processes because there is a convergence 19 in the perspectives, NRC's perspectives and academia's l 20 perspectives that mostly influence safety.
1 21 MR. CARROLL: And emergent means what emerges from 22 the first four items?
23 DR. LOIS: No, I do not think so.
24 MR. CARROLL: Okay.
25 DR. LOIS: It has to do with the dynamics of the Heritage Reporting Corporation (202) 628-4888
I 242
,_s
)
L>
1 organization.
2 DR. THURBER: Emergent processes come from l
3 traditional organization theory, and it is something that ]
4 evolves from the context of the institution that you are 5 looking at, the institution that you are looking at. And it 6 may indeed be influenced by the context, the organizational j 1
7 governance, and certainly the environment, but there are l
, 8 other factors that would influence emergent processes. !
l l 9 I guess that the importr.nt thing that you will get 10 to later, Erasmia, is that these are in the realm of theory.
11 When we got down to testing hypotheses with actual data, we 1
12 were looking at other variables which you will explain.
13 This came about as a review of the industry literature,
( 14 literature from academia and what has been done by 5%C and 15 is more theoretical than applied or even hypotheses tested 16 from this at this point.
17 MR. CARROLL: But it is not emergent in the sense 18 that I do not have anything and something is emerging from?
19 DR. THURBER: No. It is emergent in the sense i
, 20 that it is dynamic.
l 21 MR. CARROLL: Okay, 22 DR. THURBER: It is jargon. It is academic 23 jargon.
24 MR. CARROLL: I know, but i am trying to 25 understand it.
I
() Heritage Reporting Corporation (202) 628-4888 l
l l
E----___
, 243 I) 1 DR. REMICK: Did you have to say academic?.
=
- 2. MR. WARD: Two strikes against it. ,
3 DR. THURBER: But it is.a beginning to think about L
4 a complex phencmenon.
l 5 DR. LOIS: And as a result, I have to deal.with l
6 the jargon. The intermediate outcomes of safety.in terms of 7 nuclear power plants were operationalized as-follows.
8 Quality was viewed as hardware performance, design, 9 integrity and reliability. And in human performance, 10 policies, performance monitoring, feedback, accountability, 11 recognition, et cetera. And availability of resources.
12 These are some examples. I am just trying to 13 characterize these variables. The report is thick and done
() 14 by different professors. As a matter of fact, there is some 15 divergence of opinion in the report.
16 Compliance has been viewed as maintenance of 17 defense-in-depth. And also as human conformance to 18 standards. The efficiency in general form is defined as 19 the ratio of inputs and outputs. This is the classical 20 definition. Another way that it was viewed here is resource 21 allocation. That would relate to nuclear power plant 22 safety.
23 Innovation is characterized as learning and 24 problem solving capacity. And also the development and 25 application of new technology for normal and abnormal O Heritage Reporting Corporation (202) 628-4888
i I
244 '
1 conditions.
2 MR. CARROLL: I am having trouble relating ;
I 3 efficiency to resource allocation.
4 DR. THURBER: May I comment on it?
5 DR. LOIS: Please.
6 DR. THURBER: Let's just say that you are a plant 7 nanager and you are trying to comply with NRC regulations, 8 but you are also trying to make money and have a profit.
9 And this is sometimes a hard thing to do, to comply with 10 certain regulations and make a profit.
11 Efficiency has to do with making a profit, the 12 amount of resources that you put into something and what you 13 get out of it, the ration of inputs to outputs. And you
() 14 want to run an efficient operation, and you want to make 15 money. But you also want to balance it with compliance and 16 quality of the operation, because you feel that quality is 17 relsted to safety.
18 Innovation. You do not want experimentation.
19 That is not what innovation means. You want people to learn 20 in the organization and have an attitude of always 21 improving, always to do the best that you possibly can, to 22 be better than the best. You want to have that sort of 23 framework within the plant.
24 So innovation, efficiency, compliance and quality 25 need to be balanced if you are a good manager. And Heritage Reporting Corporation
(~}
x (202) 628-4888 I
1
_ _ . . _ . _ __.______________________________________J
245 1 efficiency in this context means availability. But you do L 2- not want to just run it until it breaks all of_the time.
3 You do not want to spend less money now to keep availability 4 up, because you know in the long run that that-is~ going to 5 reduce quality and you may have a problem with compliance.
6 MR. MICHELSON: Why is quality an outcome of 7 safety?
8 DR. LOIS: I think that it has viewed as a 9 precondition to safety.
10 MR. MICHELSON: Oh, that is quite different than 11 an outcome of safety.
12 DR. THURBER: If I might comment on this.
13 Quality, compliance, efficiency and innovation are general
() 14 concepts. There are a series of hypotheses below them.
15 They_are independent variables in the model as' independent 16 _ variables of safety and performance of the plant.
17 DR. LOIS: It is that organization set the 18 preconditions to safety, and prc7onditions to safety can be 19 expressed in terms of these variables. It is a way of again 20 tying this right-hand column.
21 DR. THURBER: Let me run through one possible I 22 scenario.
23 MR. MICHELSON: Being safe does not necessarily ;
24 mean that it is being efficient.
i 25 DR. THURBER: Sometimes it does. A really good Heritage Reporting Corporation O (202) 628-4888
w 246 b
1 manager has an efficient operating plant that complies, that 2 has quality and is innovative.
3 MR. MICHELSON: But a good efficient plant has 4 also been known to blow.up.
5 DR. THURBER: That is right. Because they-have 6 not been focused on the dimension of quality, innovation and 7 compliance. In this model that you see up here, it is not 8 really a model in the strict scientific sense, but this 9 approach. The idea that things like environment, the 10 political environment.- Whether Congressman Markey is 11 focused on a particular problem or not has an influence on 12 compliance for example, the general political environment.
13 And it also has an impact on efficiency.
() 14 MR. MICHELSON: It might not have any impact on 15 safety.
16 DR. THURBER: That is right, exactly. In fact it 17 may have a negative impact on safety. And then below 18 environment it is not only political but it is the economic 19 environment, attitudes of the public about nuclear power.
20 All of those things have an impact on those intermediate 21 outcomes which then affect the manager-in his ability or her 22 ability to run a safe nuclear power reactor. J 23 MR. MICHELSON: I understand everything that you 24 are saying, but I just have difficulty with calling that j 25 intermediate outcomes of safety. I thought that these were
() Heritage Reporting Corporation (202) 62C-4888 l
247
- 1 the-things-that' led to safety.
2 DR. THURBER: The intermediate. outcomes mean that 3 you do not want;a Three Mile' Island, so you want to be able
'4 ,to measure something before that'happens. So these are 5 intervening variables that you look at asla manager before 1
6 you get to a severe incident.. And we have used as the-7 ' safety indicators the safety indicators that have been 8 adopted by NRC as a dependent variable.
9 MR. MICHELSON: So you are saying that safety is 10 'the indicater, is that what you are saying?
11 DR. THURBER: Safety indicators at the end in the 12 model are the indicators that have been accepted by the NRC-13 at this point. '
() 14 MR. MICHELSON: And knowing what those indicators 15 are, then you can tell whether the operation is efficient.
16 DR. THURBER: No. We come up with measures of 17 efficiency, quality, compliance, et cetera. And we have 18 looked statistically at the relationship of those measures I 19 to the indicators. Taking the left-hand side off and 20 looking at it simply, the independent variables in the model 21 are efficiency, compliance, quality, and innovation. And 22 the dependent variables are the six indicators from NRC.
23 MR. MICHELSON: Which six indicators are those?
24 MR. CARROLL: The next slide.
I 25 DR. LOIS: Significant events, outages, Heritage Reporting Corporation O (202) 628-4888
248 1 et' cetera.
2 MR. MICHELSON: Go ahead.
3 DR. LOIS: So based on this theory, a number of 4 suppositions were developed trying to link organizational 5 factors to safety. I will read some of the suppositions.-
l 6 MR. MICHELSON: I do not think-that we'ever said, 7 what I think you are going to call safety indicators, that 8 those are indicators of safety. Those are performance 9 indicators I thought which may or may not have anything to 10 do with safety.
11 DR. THURBER: There is always the problem of 12 access to data, existing data that is collected in the same L 13 way over a long period of time. Our assumption in this
() 14 project at this point was that the performance indicators 15 are the same thing that we are called the safety indicators.
16 Because the assumption of NRC is that they are related to 17 safety. That is why you are looking at them ultimately. I 18 think that is right.
19 MR. MICHELSON: That is not validated.
20 HR. WARD: He is working at a different level. He 21 really needs the safety indicators, because he wants to 22 relate those to organizational factors. Nobody has really 23 given him any good safety indicators. So he is stuck with 24 using these surrogates with the expectation that certr. inly a 25 responsible agency like the NRC would not be using something
() Heritage Reporting Corporation (202) 628-4888 l
\ . , - _ _ _ _ - .
E -
+ : 249
{t '
' l' that is entirely worthless.
I 2 .DR..REMICK: Surely not.
3 DR. LOIS: So a number.of suppositions were l
l~
4 . developed, hypotheses linking organizational factors to l ,
5 safety. For example,'how you~can_ link organizational.
6.- factors by innovation to safety. The greater the flow of-7 information the. greater problem solving capacity and 8 learning and the safer of the plant..
9 The greater the coordination of plant functions 10 'and activities then the greater the problem solving capacity 11 and learning and the safer the plant. The greater-the L 12 motivation and morale then the greater the problem solving' 13 capacity and learning and the safer the plant.
() 14 MR. MICHELSON: You are not reading from a slide, 15 are you?
16 DR. LOIS: I am not reading from a slide. So the l
17 report has hundred of suppositions. And again it is an 18 effort to characterize organizational factors regarding 19 safety. A small empirical study was performed to test some 20 of these hypotheses. A number of variables were tested. j l
21 These showed up as having some significant relationship, 22 statistical correlations now, to the indicators.
23 MR. MICHELSON: That is not the same slide as I am l
24 looking at. Major violations for instance is not on here.
]
25 DR. LOIS: I took it out. In the new version, it O Heritage Reporting Corporation (202) 628-4888 1
250 F .
-1 is not major violation-any more as an indicator of learning.
2 DR. THURBER: :Although if you want me to talk 3 about major violations, I can, j' 4 MR. MICHELSON: Not really. I am trying to 5 understand it. This is a positive correlation. In other 6 words, the older the plant the more likely it is.
7 DR. LOIS: This is wrong. Here I tried to correct l
8 it. It is positive or negative.
9 MR. WARD: Positive or negative.
10 DR. LOIS: That is right.
'll MR. MICHELSON: Okay. That helps immensely.
I 12 DR. LOIS: .As.a matter of fact, the plant age does 13 not differentiate. It is not related either positively or
() 14 negatively. Plant size does. It seems the bigger the plant 15 the less safe. The number of plants per utility does. It 16 seems the bigger the number of plants per utility that they 17 are less safe.
18 MR. MICHELSON: That does not appeal to intuition 19 at all. The number of plants, usually the larger the number 20 of organizations. So the bigger the bureaucracy, the less 21 safe it is. Maybe that is it.
22 DR. THURBER: That was the hypothesis. And the 23 preliminary statistical analysis found that the fewer plants 24 per utility tend to be associated with a significantly 25 better safety record.
()' Heritage Reporting Corporation (202) 628-4888
\
p~
251 1 MR. MICHELSON: So a one plant utility is the 2 best.
3 DR. THURBZR: It goes against intuition, I agree.
4 MR. MICHELSON: It was the best.
5 MR. WARD: That is the whole idea.
6 MR. CARROLL: You do not have much of a data base 7 to go on.
8 MR. MICHELSON: Or much of a parameter of measure 9 of safety. You do not have a measure of safety. Before we 10 leave that though,- could we clarify.
11 You talk about earlier LERs, what does that mean?
12 DR. LOIS: I have it here. I think that this 13 chart here is a better chart, but I took it out of the
() 14 report. It describes better the results.
15 MR. MICHELSON: But I cannot read it.
16 DR. LOIS: You cannot read it. It seems that if 17 you look at region that there is a relationship to safety, 18 and this is efficiency and this is the indicators.
19 MR. MICHELSON: You mean that some regions are 20 safer than others?
21 DR. THURBER: If you do not mind if I jump in.
22 This is consistent. We did a study back in 1981 and 1982 23 that found that there is great variance across region. It 24 is explained by the way that people look at plants in the 25 regions, because they have different views in the way that O Heritage Reporting Corporation (202) 628-4888
,_ . 252 t 1 1 they evaluate plants.
2 MR. MICHELSON: How you look at a plant does not 3 mean how safe it is.
4 MR. CARROLL: But they are inspecting them against 5 exactly the same criteria that comes from headquarters, so I 6 do not understand this.
7 MR. MICHELSON: That is an interesting 8 observation. That would create a thesis all by itself.
9 DR. LOIS: Then the number of plants per utility 10 has a positive relationship to efficiency. Here efficiency 11 is defined as critical hours, and a negative relationship to 12 current indicators.
13 MR. WARD: What did you say?
() 14 DR. LOIS: Number of plants per utility. If you 15 compare it with efficiency measures, electrical output, it 16 is positive. With current indicators, it is negative.
17 DR. LOIS: It could be plants that are not-18 nuclear.
19 MR. MICHELSON: That is not nuclear plants then, 20 that is total number of any kind of plant?
21 DR. LOIS: Yes. They looked into utilities that 22 have non-nuclear plants too.
23 MR. MICHELSON: I do not know what that means 24 then.
25 DR. LOIS: Some utilities have coal plants.
(~ Heritage Reporting Corporation (202) 628-4888
253
,e\
1 MR. MICHELSON: Yes, some have hydro.
2 DR. LOIS: That is right. So the bigger the 3 number of plants per utility.
4 MR. MICHELSON: The less safe the nuclear unit is.
5 DR. LOIS: Yes. But these are preliminary 6 results.
7 MR. MICHELSON: They are nonsensical.
8 DR. LOIS: They are not defendable yet. Plant 9 size has a negative correlation. Here major violations have 10 a positive correlation.
11 DR. THURBER: If I could explain that a little 12 bit.
13 DR. LOIS: They looked at major violations as
,n
( ,) 14 earlier events. They looked at major violations between 15 1983 and 1985 and compared to indicators from 1985 to 1987.
16 Anc it seems that plants that experienced violations in the 17 previous years were safer regarding the safety indicators.
18 DR. REMICK: What are the safety incicators.
19 DR. LOIS: Significant events, forced outages, and 20 scrams mostly.
21 DR. THURBER: Erasmia, would you like me to run 22 through the more specifics?
23 DR. LOIS: Yes.
24 DR. THURBER: In this violation thing, 25 specifically the more violations a plant has in 1985 and
( Heritage Reporting Corporation (202) 628-4888
254
~
1 1984, the fewer scrams it has in 1985 through 1987 2 statistically. The more major violations a plant has in 3 1985, the fewer the forced outages it has in 1985 to 1987.
4 The more major violations that a plant has in 5 1985, 1984, and 1983, and 1982, the fewer critical hours it 6 has in 1985 and 1987. The more major violations that a 7 plant has in 1985, 1984 and 1983, the higher outage rate it 8 has in 1985 and 1987. We are limited in the data base.
9 MR. CARROLL: And a major violation is level 3 and 10 above?
11 DR. THURBER: Yes.
12 MR. CARROLL: You are just counting 3s and above?
13 DR. THURBER: Yes.
() 14 DR. REMICK: But you count security as a violation 15 here. So you have a violation in security which might have 16 nothing to do or have anything at all to do with it.
17 DR. THURBER: I cannot answer that, I am sorry. I 18 did not code these. Another colleague of mine coded. I 19 cannot answer that question.
20 MR. MICHELSON: They do seem to have some 21 semblance of understanding that you could see how they could 22 possibly be associated, but some of these others do not.
23 You know, the number of plants that a utility has, the more 24 pl. ants you have the less safe your nuclear unit will be.
25 Maybe all of the rest of them are hydro. And the more
, Heritage Reporting Corporation (202) 628-4888
j.'.. _ _ _ _ g I
4 .
"f
.c 255 l
.{ l-
" ~
l' - plants 1you have the less safe from that viewpoint.
2 DR. LOIS: Especially if.you have coal plants or 3 hydro plants, you use the.same maintenance crew for example.
'4 MR. CARROLL: No, not necessarily. ..
5 -MR..MICHELSON: 'It is a different crew entirely.
-i 6
- 7. -'
8 9
< 10 11 12 t
-13 l
14-
'15 16' 17 18 19 20 21 22 23 24 25 Heritage Reporting Corporation (202) 628-4888
256 p_
'(-) 1 DR. LOIS: The culture is the same.
2 DR. REMICK: I would say of that chart, 3 interesting but not completely convincing.
4 DR. LOIS: I agree with you. And i do not present 5 them as final conclusions or convincing. We struggle with 6 them. And that was a test. As a matter of fact, it is a 7 demonstration on how you can link some organizational 8 factors to safety.
9 MR. MICHELSON: Did you go back and look at just 10 the number of nuclear units the utility has and not coal and 11 fossil and so forth, just to see as the number of nuclear 12 units increases whether the safety goes up?
13 DR. THURBER: Yes. Respectfully, I wanted to say
() 14 that it was the number of nuclear units.
15 MR. MICHELSON: That wasn't the number, total 16 number?
17 DR. THURBER: Sorry. Correction. Right.
18 MR. MICHELSON: Okay. It is just nuclear units.
19 Okay. That is interesting.
20 DR. LOIS: It is the number of nuclear units only?
21 DR. THURBER: Yes. I'm sorry. Yes, it was. And 22 secondly, we suggest strongly in the report that this is a 23 preliminary thing, that you need to look at this over time, 24 and it may be a statistical anomaly. And the N is very low 25 and all those kinds of problems with the statistics.
Heritage Reporting Corporation O~ (202) 628-4888
257
.,,)'
(" 1 MR. MICHELSON: Because most of the arguments I've 2 ever heard kind of argue in the opposite direction.
3 DR. THURBER: Absolutely. And that is common ,
)
4 sense.
5 MR. MARD: This is all in a report that we have I
6 now, is that right?
7 DR. RYAN: That's correct.
8 MR. MARD: One of those three reports you said we 9 had?
10 DR. RYAN: The report on the risk-based 11 indicators, and the organization and management findings was 12 transmitted on the 7th of August and the report on the 13 maintenance indicators was transmitted on the lith. So you
() 14 have all of them.
15 DR. THURBER: Erasmia, may I add one more very 16 interesting finding?
17 DR. LOIS: Yes.
18 DR. THURBER: And that is that profitability, in 19 particular, earlier profitability in the history of a 20 utility running a plant, tended to be significantly, 21 positively related to safety indicators, which is intuitive, 22 but we confirmed that hypothesis with data.
23 MR. MICHELSON: Based on what kind of 24 profitability? A bottom line and a balance sheet for the 25 annual report? Is that the profitability you are talking Heritage Reporting Corporation CTJ (202) 628-4888
258
'~
1 about?
2 DR. THURBER: Yes.
3 MR. CARROLL: For the entire utility's operations?
4 Not just --
5 DR. THURBER: For the nuclear division broken out 6 in the plant, I mean in the utill.ty.
7 MR. MICHELSON: Part of the bookkeeping.
8 DR. THURBER: Xight. Now, we don't know about how l
9 they keep books; and maybe profits from one side cost-10 subsidize another side. But with what is reported, that is 11 what we found. And it's the nuclear side.
12 MR. MICHELSON: You might want to check back to 13 see if it relates to the bottom line total profitability.
() 14 DR. THURBER: Good idea.
15 DR. LOIS: Yes. I have this thing of area 16 resources that is positively related.
17 This work as I said started out in October. A lot 18 of work, Phase II will start pretty soon. And the thrust of 19 Phase II will be to identify some key propositions to be l 20 tested and then identify data sources to test the 21 propositions and possibly validate some useful indicators 22 for the NRC.
23 MR. MICHELSON: Since we got these big books, I 24 sensed that that was some kind of a final report? Is it 25 really or is it an interim report yet?
Heritage Reporting Corporation O' (202) 628-4888
_ _ _ _ . _______________________a __
R :
259 1 DR. LOIS: It is.a draft report. It.was improved 2 a little bit. I have a better version since we've sent 3- that.
4 MR. MICHELSON: So you are suggesting it is really 5 a. draft final report, then, not an interim report?
6 DR. RYAN: . It is a draft final report of Phase 1.
7 It is not intended as the final report on research.
8 MR. MICHELSON: Is there a Phase II coming up?
9 DR. RYAN: That is what Erasmia was just talking 10 about there.
11 MR. MICHELSON: Okay.
12 DR. LOIS: Phase II will deal mostly with 13 identifying indicators and validating.
.14 DR. RYAN: Further validation of the indicators 15 that have shown-some promise thus far. And also looking at
.16 the issue of interpretations of the results. I think this 17 gets to some of the questions you have been, asking, you 18 know what does it really mean.
19 MR. MICHELSON: What time frame does that one get-20 wrapped up in as a report, Phase II?
21 DR. LOIS: Starting in October, 22 MR. MICHELSON: When is it scheduled to be 23 completod?
24 DR. LOIS: One year.
1 \
25 MR. MICHELSON: One year? I O Heritage Reporting Corporation (202) 628-4888
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - . - _ - - - - . - - - _ - - - - - - - - - 1
l
_ 260
"' 1 DR. LOIS: One year.
2 That concludes my presentation on organization 3 indicators.
4 We have two additional activities on the 5 development of leading indicators of safety. I will walk 6 you through quic' '.y.
7 One idea is to utilize non-nuclear experience to 8 identify indicators. And that came in as a proposal 9 responding to competitive procurement.
10 DR. REMICK: Is this the George Mason University 11 work?
12 DR. LOIS: This is not the George Mason 13 University.
( )- 14 DR. REMICK: It's not.
15 DR. LOIS: This is done by a company called CTA 16 Communications Technology Applications, something like that.
17 It is a local firm.
18 And the objective is to learn to utilize the high 19 frequency and less severe accidents that occurred in non-20 nuclear plants, learn about unsafe plant behavior, and 21 identify indirect indicators of safety that can be 22 translated into nuclear environment.
23 This work and the George Mason work will be 24 collaborating, will be tied together.
25 MR. CARROLL: Non-nuclear plants meaning utility
(~) Heritage Reporting Corporation
\> (202) 628-4888
l l 261 L h. 1 power plants or chemical plants or --
,% 2 DR. LOIS: Chemical plants. They started this in 1
3- June,-and they looked at different utilities and they feel l 4 that chemical plants are the best approximations to nuclear 5 plants.
I 6 In addition, too, they found a lot of interest 7 from the. chemical industry to provide data, because they are 8 very excited about the development of leading indicators for 9 their purposes.
10 DR. RYAN: Could I just inject something? Through 11 these presentations, one of the underlying themes is we do l
l 12 have a data problem.
13 And so we have been a little bit attracted to the
() 14 idea of attempting to go out in a non-nuclear setting for 15 which there is a. lot of data, do some of the developmental I
16 work, and then once we have validated some of these 17 indicators in an environment like the chemical industry, 18 then work on translating them or integrating them into the 19 nuclear setting.
20 So this project is to some extent a response to .
I 21 the data problem we have, much like the George Mason 22 University one is, where we have a real paucity of their 23 probability to be used at some of the risk-based analyses we 24 are trying to do.
25 DR. REMICK: Is it correct to infer, then, that O Heritage Reporting Corporation (202) 628-4888 i
262 7.s
') 1 the chemical industry has not developed similar performance 2 indicators on safety?
3 DR. RYAN: They have done a certain amount of work 4 and in fact the firm CTA, Communications Technology 5 Associates, is working with those>, the chemical industry, in 6 taking a look at come of the indicators they are pursuing.
7 They have developed a list of criteria to do down through 8 those so that we can pick out a few that we think at least 9 look like they might fit in the nuclear industry, and then 10 focus our resources on the continued development and 11 validation of those before we attempt to transfer.
12 DR. LOIS: This second slide addresses what Dr.
13 Ryan just covered, the high frequency of accidents, and the 14 data availability in non-nuclear environments. And this 15 schematic here outlines the tasks fer this work.
16 Again, it starts out with the current indicators 17 of safety, and direct performance indicators in non-nuclear 18 environments will be identified in a way that matched to our 19 performance indicators.
20 That is why we felt that chemical plants probably 21 are the most appropriate analogous plants.
22 Once direct performance indicator are identified 23 in non-nuclear environments, then indirect indicators will 24 be, or a number of hypotheses will be developed to link 25 programmatic performance in non-nuclear environments with Heritage Reporting Corporation (202) 528-4888 I
263
(
l' their direct performance indicators.
2 And once this.is done then it will be a mapping 3 process, hopefully identifying over here indirect 4 performance indicators, leading indicators of safety, that 5 come out of this experience, from that mapping process.
6 Again, this is leaning from non-nuclear 7 experience.
8 DR. REMICK: Now, you indicate that the non -
9 nuclear plants have higher accident rate. What do you mean 10 by accident rate?
11 DR. LOIS: Less severe. I mean ideally what we 12 would like to have over here,. direct performance indicators, 13 safety indicators would be if we had many accidents in
() 14 nuclear plants'then we would find what triggers these 15 accidents, and where we would define mechanisms, 16 organizational mechanisms that are causal factors.
17 DR. REMICK: Right. I understand. What is 18 defined as an accident in a non-nuclear plant? Is it an 19 injury to an individual? Is it loss of property?
l 20 DR. LOIS: A lot of yes, injuries, contaminations, 21 a lot of safety system failures that do not evolve to 22- radiation release and massive --
i 23 DR. REMICK: So you are using accidents in this 24 case like we would failures previously?
25 I'm trying to understand the basis for that Heritage Reporting Corporation O- (202) 628-4888 i
264
,r m
\' )
I statement in the chemical industry, what we are defining as 2 accidents.
3 Is it failures that lead to loss of life?
4 DR. LOIS: I cannot answer the question, because 5 we haven't done this work.
6 DR. REMICK: Okay.
7 DR. LOIS: So what I am trying to say here is that 8 the safety will be, in the chemical plants, will be 9 characterized in terms of performance indicators.
10 DR. REMICK: Yes.
11 DR. LOIS: And hopefully will be events as opposed 12 to equipment unavailability.
13 DR. REMICK: The reason the question came to mind I
() 14 is you know, DuPont company is credited with having an 15 excellent industrial safety record. They are a chemical 16 industry. And therefore I thought well, why is this 1
17 statement that they have more accidents, if they have such a 18 good industrial safety record? There seemed to be a 19 disconnect there. It is not important. I was just trying 20 to understand that statement. l 21 DR. RYAN: Well, I think the one thing we might 22 want to keep in mind is we have, we characterize the 23 performance indicators or the criteria measures that we are l I
24 us 2. ? in this industry as penultimate. I mean, they are not l 25 really the passive accident type thing.
f^)
k/
Heritage Reporting Corporation i (202) 628-4888 1 1
265
(~#l-'
1 DR. REMICK:
Yes.
2 DR. RYAN: We are going to be looking for similar 3 things in the chemical industry, things that might happen 4 far more frequently than do in this industry, which would 5 give us a better pool of data to work with statistically and 6 conceptually.
7 And certainly the vertical lines going, well, this 8 way between the nuclear industry and the chemical industry, 9 certainly we would like to see at least conceptually some 10 similarities between those, 11 DR. REMICK: V$s, But apparently you have 12 concluded they have a higher frequency rate..
13 DR. RYAN: That is correct. The contractor went
() 14 out and did some preliminary work, looked at a variety of 15 industries. They came in and gave us a rather extensive 16- briefing. And considering the resources we have and the 17 match and the availability of data, and I guess the types of 18 events that they have, it was decided that where to sort of 19 kick this off is in the chemical industry.
20 DR. REMICK: Please proceed.
21 DR. LOIS: Finally, the work on performance 22 indicators that I started out with, and I focused on 23 maintenance, has been continued, trying to cover the range 24 of programs within nuclear power plants, and come out with a 25 set of indicators that adequately covers the performance, Heritage Reporting Corporation O (202) 628-4888
'266 0- 1 and in an optimal way.
2 This work has been done by SAIC again and the 3 University of Maryland is collaborating in this work, and 4 Dr. Modarres is here with John Wranthall again, who is the-5- principal investigator.
6 DR. REMICK: I'm sorry. I missed the university.
7 DR. LOIS: 'Of Maryland.
8 DR. REMICK: Maryland. Okay.
9 DR. LOIS: You cannot hear me well, still? I have 10 a feeling that I speak very loudly.
11 DR. REMICK: No, I heard you as Melbourne, and 12 that didn't make sense.
13 DR. LOIS: Maryland University.
() 14 DR. REMICK: And I saw U.M. and I thought it must 15 be Michigan on the next slide.
16 DR. LOIS: So the goal of this work is to define 17 an integrated set of performance indicators. It is a very 18 high goal.
19 During '89, a set of indicators that adequately 20 cover the nuclear power plant programs will be identified.
21 And in ' 90 they will be checked for completeness, 22 remove duplications, to the extent possible.
23 DR. RYAN: Erasmia, could I say one thing about 24 the fiscal years? That says FY '89 but in some of these 25 cases we have started this research fairly late in the year.
O Heritage Reporting Corporation (202) 628-4888 I
I i E_ _ - - _ _ _ _ _ _ _ _ _ _ - - - - _ - - - - _ _ _ !
267 O 1 And what this is oriented to is the funding and so on. So 2 this is very preliminary. this project has not been going 3 on very long.
4- DR. LOIS: Regarding the process, how Task 1 will 5 be done, the University of Maryland has developed a model 6 relating nuclear power plant operations a few years ago that 7 was a big DOE project. And what they did is develop a top-8 down logic, hierarchical logic, based on goal trees and l
L 9 success trees.-
10 This methodology, called integrated approach, can 11 be extended beyond the hardware performance, to address the 12 programs and program performance, and therefore try to tie 13 directly programmatic aspects of plant performance to
() 14 hardware performance.
15 Let me explain. Let me show this. Thank you.
16 It-is a classical top down logic.
17 It starts with a top goal, component availability.
18 It splits down to component timely restored is 19 achieved. Goes down, how this top goal is achieved, and 20 goes down to the individual activities that have to be 21 performed in order to accomplish this top goal.
22 This work has been, the biggest portion of this 23 work has been done at the University of Maryland as followup 24 work for a number of utilities. And therefore, during 25 Calendar Year ' 90, this diamond that starts out with safety O Heritage Reporting Corporation (202) 628-4888
268
\~') 1 and goes down to the nuclear power plant organization, will 2 be complete.
3 It involves a number of experts. It is hardware 4 people, experts in PRA, experts in organization and 5 management, experts in plant operations are involved in this 6 work, in order to complete this diamond here and hopefully, 7 through this top down process, identifies some linkages 8 between programmatic performance to hardware performance, 9 and defines some indicators that are adequately covered in 10 the nuclear power plant program.
11 MR. WARD: Those last two figures are from some 12 report?
13 DR. LOIS: Yes.
() 14 MR. WARD: Do we have that report? Is that a 15 published report?
16 DR. LOIS: Dr. Modarres?
17 DR. MODARRES: I'm Mohammed Modarres from 18 University of Maryland. This, as was said, the project just 19 was started in June, and these figures are taken from 20 proposal that we were giving to the NRC. But this structure 21 essentially is partially developed currently but is not in 22 formal report yet.
23 MR. WARD: In this column, in this figure, what is 24 the column over on the right? What do those words mean?
25 ARe those examples?
[~ Heritage Reporting Corporation (202) 628-4888
269 I)
' 1 DR. LOIS: Safety functions, major -- I guess you I '
2 go frora inaugural safety to safety functions to subfunctions 3 that support the overall functions, and then component i 4 reliability down to the programmatic performance.
5 DR. MODARRES: Can I add something?
6 Essentially it shows that this is a hierarchical 7 breakdown of the top objective, such as if it is safety, you 8 can break it down to critical safety functions. Examples 9 would be maintain reactivity, control reactivity and things 10 like that.
11 Now, objective there is essentially a, it is a 12 major function within, major safety function. And when you 13 break it down you eventually get to lower level functions.
n
() 14 You get to things such as maintain a subcooled margin in the 15 reactor. And eventually when you follow that down you get 16 to equipments that are there to perform them and in 17 combination those generate success paths, but you have 18 several success paths. Objective of that success path is to 19 achieve reliability. So that right column shows that what 20 is the objective of having each of these things exist.
21 DR. REMICK: I take it this diamond tree is a 22 traditional diamond tree used in other purposes and that 23 right column is to translate it to this specific proposed 24 use? Is that it?
25 DR. MODARRES: Yes.
() Heritage Reporting Corporation (202) 628-4888
t j-. 270
- f] l' DR. REMICK: Okay.
2 DR. LOIS: So through this process, hopefully we 3 will be able to select a set of indicators by overlaying 4 predefined indicators on this hierarchical structure in 5 evaluating how they monitor critical functions of goals in 6 plants.
7 And in that respect, the work we've done for-8 maintenance with the 78 indicators or elsewhere on training 9 will be~ utilized for the purposes of. identifying and 10 selecting indicators.
1
'll Having done that, the indicators will be tested 12 for adequacy, completeness, and duplication will be removed.
13 And how that will be done is I guess by ranking
() 14 plant activities and programs based on their impact on
-15 higher level of safety functions.
16 Select and test aggregations of indicators using 17 statistical methods and determine a final reduced set to 18 validate and interpret.
19 DR. REMICK: Okay. Any questions?
20 (No response) 21 DR. REMICK: It has been an interesting 22 presentation. I'm sure you can see that the subcommittee is 23 ' steeped in the academic jargon of organization and 24 management science. But it has been interesting.
25 I guess one of my comments, I realize this is O Heritage Reporting Corporation (202) 628-4888
- I l
271 ;
,e ss- ,
V 1 research, and I certainly defend research. Eventually the i a
2 agency has to ask the question, though, how would it use 3 these? The ACRS frequently asks people in the Office of
'4 Research where'is "there," and if you don't know where j
5 "there" is how do you know when you are going to be there?
6 It seems as if eventually the agency has to ask 7 the question of how would it use such indicators assuming 8 that out of this research you develop some that look like 9 they are valid indicators.
10 And I'm not sure where that fits in the 11 organization. But it seems'to me it has to be addressed at 12 some point.
- 13. DR. RYAN: Can I make two answers?
() 14 DR. REMICK: Please, Tom.
15 DR. RYAN: First of all, basically, the intended 16 use of indicators as we understand it within the NRC now is 17 to guide senior management meetings.
18 So I would assume that these would be used in the 19 same way.
20 More specifically, the work that we are doing has 21 emerged out of a list that the Commission directed to the 22 staff that ranged all the way from management indicators to 23 various technical support program indicators and so on.
24 And basically this research was to some extent 25 triggered by that and we are responding to that.
Heritage Reporting Corporation O- (202) 628-4888 w___________
1 l'
'I 1
272 1 A little bit later, the. issue came up of so we L
2 have 15 indicators, how do I step back from this array of 3 indicators that are all going in different directions and 4 make some kind of a conclusion or see a pattern or 5 something?
6 Part of this work, this bottom up, is an attempt 7 to start taking a look at collections of these indicators 8 that we're developing and attempt to develop some 9 composites.
10 So basically the answers are we would assume that 11 the indicators would continue to be used by senior level s
12 staff management in their functions. How the Commissioners 13 themselves individually would like to use them I'm not quite
() 14 sure.
15 The other thing is basically the staff has been 16 directed to develop indicators in these various areas.
17 MR. CARROLL: When you speak of senior staff 18 meeting, you are talking about the semi-annual --
19 DR. RYAN: Quarterly. Semi-annual meeting. Yes.
20 Senior management meetings.
21 MR. CARROLL: With the Regional Administrators?
22 DR. RYAN: Right.
23 MR. CARROLL: That look at how various plants are 24 doing?
25 DR. RYAN: Yes.
O Heritage Reporting Corporation (202) 628-4888
., 273
(#)
1 MR. CARROLL: You don't mean anything more than j 2 that?
3 DR. RYAN: Well, that is basically what the agency f 4 is using them for now. The degree to which I guess they l 5 weuld expand their use I guess would be determined by a l
l 6 number of things, including the veracity of the indicators E
7 that are ultimately developed and the degree to which they l 8 can be defended scientifically.
9 MR. CARROLL: Have you ever been to one of the 10 senior staff meetings to get calibrated as to whether they 11 really use them or how they would use them?
12 DR. RYAN: Those meetings as I understand it are 13 limited to a certain group of people, and unfortunately we
() 14 are not included in that circle.
15 DR. LOIS: It could be that as a byproduct of 16 this work will be disseminated as a technical, technological 17 transfer to the plants.
18 We learn about organizational behavior. If we do 19 not use it as indicators in this agency, I think it is very 20 useful information for the industry. The same for 21 programmatic indicators, training indicators. We will see 22 if we can identify indicators appropriate for the industry.
23 However, if we cannot, with training effectiveness 24 indicators, I guess, that itself is a very valuable 25 information.
(']
\s Heritage Reporting Corporation (202) 628-4888
274 7
\') 1 DR. REMICK: That reminds me, I'm not suro, but 2 the Office of Research apparently has an RFP that is being 3 evaluated to hold a workshop of people on training l
4 effectiveness. Is that Research?
5 DR. LOIS: Yes.
6 MR. COFFMAN: Yes. That is scheduled, or it is 7 not yet scheduled. The contract is in the process of being 8 negotiated. So I would expect at least six to nine months.
9 DR. REMICK: Is that something that you are going 10 to invite training people to attend, training managers and 11 so forth, or is it going to be a smaller group?
12 MR. COFFMAN: It would be a selected group, but I 13 think one of the primary groups of attendees would be n
() 14 utility maintenance management.
15 DR. REMICK: There is certainly a lot of interest 16 in knowing how do you evaluate the effectiveness of 17 training. And so I would think there would be a lot of 18 interest in that workshop.
19 I realize sometimes you want a workshop to be just 20 a small group of individuals. Perhaps you can accomplish 21 more. But I predict there will be a lot of interest in 22 that.
23 DR. RYAN: Could I just add a footnote to what 24 Erasmia said about the use of the indicators beyond just 25 senior level management? And I think this applies
(')
b-Heritage Reporting Corporation (202) 628-4888
- _ _ = _ _ _ - - --
f.4 275
\',)~
l primarily to what we are calling the programmatic 2' performance indicators, the process indicators.
3 One would hope that if we are going to establish 4 any kind of criteria for what constitutes a good versus a 5 not so good training program or procedures or anything else, 6 much like the military and other industries, would very much 7 like to be performance based.
8 You tell me to have feedback in your training 9 program. It costs a lot of money. Show me that this has 10 some relationship to the ultimate goal of safety.
11 And so I think one of the, what do I want to call 12~ them, the offshoots of this, in its stricter sense, is 13 certainly a much firmer technical basis for a criteria for
() 14 opening in a lot of these areas as to what constitutes good, 15 bad or indifferent, rather than simply accepting some of the 16 approaches that have been followed in other industries, for 17 example, just trying to piggyback on them without any 18 demonstration that in fact in this particular setting it 19 really makes a difference.
20 MR. WARD: Do you think, for example, a thoughtful 21 nuclear utility might make use of some of these research 22 results to modify perhaps the design of its organizational 23- arrangements?
24 DR. RYAN: I would certainly think so.
25 e As a matter of fact, in the case of the process Heritage Reporting Corporation (202) 628-4888
1
,s
- 276
()-
1 indicators, you know, there is this tremendous concern, can 2 they fake the indicators.
3 MR. WARD: Yes.
4 DR. RYAN: Well, if we have demonstrated 5 statistically that there is a relationship between certain 6 attributes of a training program, with some measures of 7 safety, I would say let everybody have at it and be more 8 than happy to see everybody go to the effort to include 9 those attributes in a good training program.
10 So I don't see these just as something to hold 11 over people's head in the sense that we are going to count 12 them --
13 MR. MARD: That's not quite the same as faking the
() 14 indicators, though.
15 DR. RYAN: Well, we're just as concerned about 16 faking the output indicators. Some we could manipulate.
17 Well, I think on the process side if we can demonstrate that 18 certain attributes of any of your technical support 19 programs, for example, can be related to some index of 20 safety that we all agree upon, I think that is information 21 that should be shared and that should be a goal that 22 everybody would have in putting together a training program 23 or procedures or anything else.
24 MR. WARD: Yes.
25 DR. RYAN: So I see that as an another offshoot of Heritage Reporting Corporation (d~)
(202) 628-4888
'277
/
.1 the attempt to develop these programmatic, more process 2 oriented indicators.
l 3 DR. REMICK: Any further questions or comments?
4 (No response) 5 DR. REMICK: Well, the subcommittee will follow
[
1 6 with interest Phase II. And I'm sure that it would like to 7 hear the completion of Phase II of the further development.
t-
- 8 MR. WARD
- Yes. Some of this, I think you have l
9 pointed out these three documents, which we obviously 10 haven't read.
11 Is there any other written material that isn't too L 12 advanced that you would suggest? I'mean, the last two 13 subjects aren't really reported on yet I don't believe, are
() 14 they?
15 DR. RYAN: I think that basically the three 16 reports you have are ones that we're hoping to move to press 17 within the next very short period of time.
18 I guess we're probably talking another six, eight 19 months before we are going to be at a point where any of 20 this follow-on work will be at a point where we would be 21 generating another document.
22 But certainly we are available for in-process 23 presentations any time the committee would like to have 24 them.
25 DR. REMICK: Anything further? Dave?
Heritage Reporting Corporation
/]" '
(202) 628-4888
1-t j- 278 l a)i
'~
1- MR. WARD: Well, I think this is potentially 2 important work. But it is, I think we need to hear enough 3 about it so that we kind of get used to the --
4 DR. REMICK: So we understand it fully. Sure.
5 MR. WARD: Understand it a bit. And getting hit 6 with a whole block of it all at once --
7 DR. RYAN: One of the things that I would just 8 make a suggestion, we have both staff and contractor:
9 personnel available and if you have a particular issue or 10 one of these projects you would like to go in in more 11 detail, certainly we would be happy to make a more indepth 12 presentation on any one of the projects that you would like 13- to hear about.
() 14 I know Erasmia tried to cover a lot of material 15 this afternoon. And it's kind of difficult to do.
16 DR. LOIS: One of the objectives of my 17 presentation was to give you a flavor --
18 MR. WARD: I sort of contrast the scope of the 19 extent of I guess the intellectual content of what we 20 listened to this afternoon, spent four hours on with what we 21 spent four hours on this morning.
22 DR. REMICK: What conclusion am I supposed to draw 23 from that?
24 (Laughter) 25 DR. REMICK: Anything else?
O Heritage Reporting Corporation (202) 628-4888
(l) 1 MR. MICHELSON: Yes. What is the monetary scope 2 of this kind of program? What does it cost? j 3 DR. RYAN: Let me think.
4 MR. WARD: About the same as a quarter of a 5- semiscale test.
6 MR. MICHELSON: Whatever that is.
7 DR. RYAN: I believe that for FY ' 90, the entire 8 organization and management' research program, which this 9 represents about 50 percent of it -- we have other projects 10 in that particular area, but this'comes in under that 11 umbrella -- I think we are talking 'about somewhere around 12 500K for all of these projects collectively. That's for FY 13 '90.
() 14 MR. MICHEI SON: Yes. Has it pretty much been done 15 in FY '907 16 DR. RYAN: Well, it depends.
17 MR. MICHELSON: FY ' 90 hasn' t started yet. '89.
18 Has it been done in '89? Is this about a year's work we are 19- looking at?
20 DR. RYAN: Well, it depends now. The Maryland 21 work started in June. The SAIC work on maintenance 22 indicators is about a year and a half ago, isn't it?
23 The University of Minnesota, at alt work on the 24 top down, the organizational effectiveness, I guess is about 25 a nine-month. But there was some time we spent even after O Heritage Reporting Corporation (202) 628-4888 )
_s 280
/ )
1 the contract was signed really getting things moving.
2 And the CTA work, that is going to the chemical 3 industry and trying to develop indicators --
4 MR. MICHELSON: What you heard today I guess is 5 about 250K worth of work.
6 DR. RYAN: Well, I would stay maybe 350K.
7 In the SAIC maintenance indicator work, a lot of 8 that goes into bringing the data together and sitting down 9 and poring through these sundry sources of information to 10 develop some of these things.
11 So on the front end it doesn't tend to be quite as 12 expensive when it is more conceptual, but once we get out 13 trying to get the data, that is when the cost sort of
() 14 expands exponentially almost.
15 MR. CARROLL: Do I remember right, Tom, that the 16 Commissioners specifically asked you to come back in a year l 17 to brief them on the progresc of the organization and 18 management?
i 19 DR. RYAN: Yes, they did. However, my guess is 1
20 that I'm not sure that the indicators were the motivator.
21 It is some work that we are doing now in one of the 22 utilities, actually doing some organizational modeling, 23 collecting data on organization and managers and attempting 24 to integrate it into the PRA process.
25 What I'm trying to say is we have an Heritage Reporting Corporation O- (202) 628-4888 l
l l
281
~
1 organizational and management research element.in our plan 2 and if you read it along with some of that modeling work and 3' so on we also have the indicators work under it, because 4 this involves more than individuals.
5 But the emphasis of that briefing.I think is going 6 to be'more on that aspect of organization than this.
7 Although that may all change between now and August when 8 they have: asked us to come back and give them a status 9 report.
10 DR. REMICK: Any further questions or comments?
11 (No response) 12 DR. REMICK: If note we thank you very much, Tom,
- 13. and your associates, for coming down and giving us this
() 14 presentation. I are happy to see it developing, and I agree 15 with Dave, I think there is a lot of intellectual input 16 being put into it from a diverse group of people, and it is 17 good.
18 DR. RYAN: Thank you very much.
19 DR. REMICK: With that, I adjourn the meeting.
20 (Whereupon, at 4:25 p.m., the meeting was 21 adjourned.)
22 23 24 25
() Heritage Reporting Corporation (202) 628-4888
g . _ _ .
i L-i
~
1 CERTIFICATE L 2 3 This is to certify that the attached proceedings before the 4 United States Nuclear Regulatory Commission in the matter
.5 of: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 6 Name: MEETING:.
SUBCOMMITTEE ON HUMAN FACTORS 8 Docket Number:
9 Place: Bethesda, Maryland 10 Date: September 27, 1989 11 were held as herein appears, and that this is the original 12 transcript thereof for the file of the United States Nuclear 13 Regulatory Commission taken stenographically by me and,
(] 14 thereafter reduced to typewriting by me or under the 15 direction of the court reporting company, and that the 16 transcript is a true and accurate record of the foregoing 17 proceedings.
18 /s/ D% N 19 (Signature typed) : JOAN ROSE 20 Official Reporter 21 Heritage Reporting Corporation 22 23 24 25 O Heritage Reporting Corporation (202) 628-4888
g.s p
l.
INTRODUCTORY STATEMENT BY THE CHAIRMAN OF THE ACRS SUBCOMMITTEE ON HUMAN FACTORS, SEPTEMBER 27, 1989 ROOM 110, PHILLIPS BUILDING, BETHESDA, MD.
THE MEETING WILT. NOW COME TO ORDER. THIS IS A MEETING 0F THE ACRS SUBCOMMITTEE ON HUMAN FACTORS.
I AM F. REMICK, CHAIRMAN. THE OTHER ACRS MEMBER 3 IN ATTENDANCE !RE J. CARROLL, W. KERR, AND D. WARD.
TODAY'S. MEETING WILL BE TO DISCUSS: 1) THE ACCESS AUTHORIZATION RULE AND 2) PERFORMANCE INDICATORS.
E. IGNE IS THE COGNIZANT ACRS STAFF MEMBER FOR TCDAY'S MEETING.
THE RULES FOR PARTICIPATION If4 TODAY'S PEETING HAVE BEEN ANN 0UNCED AS PART OF THE NOTICE OF THIS MEETING THAT WAS PUBLISHED IN THE FEDERAL REGISTER ON SEPTEMBEP 13, 1989.
THIS MEETING IS BEING CONDUCTED IN ACCORDANCE WITH THE PROVISIONS OF THE FEDERAL ADVISORY COVMITTEE ACT AND THE GOVERNMENT IN THE SUNSHINE ACT.
WE HAVE RECEIVED t:0 WRITTEN OR ORAL STATEMENTS FROM MEMBERS OF THE PUBLIC.
IT 15 REOUESTED THAT EACH SPEAKER FIRST IDENTIFY HIMSELF OR HERSELF AND SPEAK WITH SUFFICIENT CLARITY AND VOLUME S0 THAT HE OR SHE CAN BE READILY FEARD.
DO ANY SUBCOMMITTEE MEMBERS HAVE INITIAL COMMENTS AT THIS TIME?
(CHAIRMAN'S COMMENTS, IF ANY) WE WILL NOW PROCEED WITH THE MEETING, AND I CALL UPON SHER BAHADUR FOR THE FIRST PRESENTATION.
gy 1(h . .
BRIEFING FOR SlECOMITTEE ON HlFAN FACIDRS ,
ADVISORY ComITTEE ON REACTOR SAFETY (ACRS) 1 l
ACCESS AUTHORIZATION PROGRAM FOR NUCLEAR P0HER PLANTS BY <
SHER BAHADUR, CHIEF O~
REGULATIONS DEVELOPENT BRANCH ;
DIVISION OF REGULATORY APPLICATIONS OFFICE OF NUCLEAR REGULATORY REASEARCH .
SEPTEMBER 27,1989 O
\
- m. -
.g, ,
a.'
c.,
E 1
,lQ). .
- e
'ICRCDUCTION CRGR RAISED BASIC ISSUES RLLE STILL UNDER DEVELOPENT THIS PRESENTATION IS A PROGRESS REPORT a
O'
.i O
2
a l
l-
~ -
l' l
l 1
L ,
BACKGROUND l
PhRCH 198tl PUBLISHED PROPOSED RULE PhRCH 1988 PUBLISHED PROPOSED POLICY STATEMENT PhRCH1989 PRESENTED OPTIONS TO THE O'
- coHsiSSION l APRIL l!E9 RECEIVED COWIISSION DIRECTION TO:
(I) DEVELOP THE FINAL RULE (II) DEVELOP THE REG GUIDE ENDORSING INDUSTRY-6UIDELINES WIm APPROPRIATE EXCEPTIONS 1
l' l
l l'
O 3
l
!N l v 1
PERFORMNCE OBJECTIVE THE UNESCORTED ACCESS AUTHORIZATION PROGRAM IS TO PROVIDE HIGH ASSURANCE THAT INDIVIDlMLS GRANTED UNESCORTED ACCESS TO PROTECTED AND VITAL AREAS ARE:
TRUSTWORTHY AND RELIABLE
('y- NOT AN UNREASONABLE RISK TO THE PUBLIC HEALTH AND SAFETY (INCLUDING THE RADIOLOGICAL SABOTAGE) i e
O 4
T fJ ElBEES OF M RJLE-1%JOR ATTRIBLTTES OTHER UNESCORTED ACCESS ALITHORIZATIONS
- TEMPORARY
- EXISTING
- REINSTATED
- TRANSFERRED OI -
ca D w screen 1=
REVIEW PROCESS PROTECTION OF INFORMATION RECORDS ALIDITS i
? (
O 5
~
'n..
l N JOR ATTRIBLTTES BACKGROUTO ITNESTIGATION I
l PSYO10 LOGICAL ASSESSENT BEHAVIORAL OBSERVATION PROGRAM O.
W O
6
., . . (
n yi) s, s ,$
g' f 3
.V 1.n l
l ISSUES RAISED BY CRGR 1er A RULE IF UTILITIES ARE ALREADY DOING WELL?
{* MW NOT ENDORSE. ItRISTRY GUIDELINES IN TOTAL RATER THAN TAKING FOR EXCEPTIONS?
Mu c0LD SHUTIXM4 PROVISIONS IN THE RULE ARE SO LIBERAL?
- O' l
l l ..
O V 7 l
l 1
l l
d Per Resefor G c ans, bT blo. oF ug:Tip ,, p,,.
' LEVELS 9Edo.d.ENGTS (30,e . s$oyyct) UTsuTlES *sde Can Erob k . . . .
- P8MN IM E. .. ..10 0 .
EdeAks *
.. . Eve-Jed thanry lidtny 150 k (2.05o Incruental Erpe>ue.to woai., . iso x y.
').
. . .- . . . . . . .. ~.- ..
._
- N .fd"ab M .._..200% .
Sui M M6_ .. . . - . . . . - . . ..
- . . . ate. .. . . . . . . 7. 5. k .. .
k
.M..M -
. _ IB5 .3_- U go
. _. . .. . w __ ..._th..; w q _ .-noe- =
. . - .......e . . - - . . . . . .a Incre .k.J , ~ %- .
g s omu.eee e.a - o s e es. e . . og . o n. e. e
...ee e
=
m Aas s.s Gn 2) .
&Pe ^~ _ .
g E*pp i
. .. j
_. d 1o cM ads d . Eec.evds . =
-- tr e<g .
q **
.. . . . . .. .. . Psy ErmJ_s@w 5.g\
=
. . . - . . . . . . ... . .sekuAero M. ..
U
= . soq
. g
% mum E
M61 Itll . fhekg. Giu 5 C .
Pg aabdb.
. w.m1 d E r
2-meme
v V. , ,
li .
g' u/ .
m.
I.
_NgDFORARULE ENSURE THAT LICENSEES NOT C0fflITTED TO BASIC STANMRDS IMPROVE TEIR PROGRAM.
- PROVIDE A UNIFORM STANMRD FOR THE NRC TO EVALUATE
. INDUSTRY PROGRAMS.
PROVIDE A WELL-DEFINED ECFMNISM FOR EFFECTIVE ENFORCEENT
?
O' PROVIDE ASSURANCE THAT VOLUNTARY AND IMPROVED PROGRAMS ARE NOT DROPPED.
PROVIDE PROPER GUIIMNCE FOR FUTURE PLANTS.
ESTABLISH AN INDUSTRY-WIDE STANDARD ON ACCESS AUTHORIZATION PROGRAMS.
D 9
- j ';,
,v>~. :
EXCEPTIONS TO THE INDUSTRY GUIDB.INE MILITARY HISTORY TOTAL, NOT JUST FOR 5 YEARS.
RELIABLE AND VALID PSYCHOLOGICAL TESTS, NOT JUST RELIABLE.
- ~
REVIEW PROCESS EXTENDED TO ALL EMPLOYEES, NOT JUST FOR PERMANENT EMPLOYEES OF THE LICENSEE.-
O' -
AUDIT eveRv vEAR IF A ticeNSEe c0" Tracts OUT EVEN A PORTION OF ITS PROGRAM.
l l
O l
10
? .
30 VISIONS FOR COLD SHIJTDOWN ACCESS GRANTED TO AREAS WHICH HAVE BEEN " DEVITALIZED" UNDER COLD SHUTDOWN REQUIRES FINGERPRINTING - NOT A TDTAL BACKGROLND INVESTIGATION .
I REQUIRES BEHAVIORAL OBSERVATION PROGRAM REQUIRES A SERIES OF COMPENSATORY ftASURES VISUAL INSPECTION OF PROTECTED AND VITAL AREAS APPROPRIATE PROCEDURES FOR START-UP AND SAFE OPERATION i -
NO ACCESS TO AN OPERATING UNIT IN A MULTI-UNIT SITE DURING l
COLD SHUTDOWN 1-O I
1 11 I - - _ _ - _ _ _ _ _ _ _ _ _ _ _ -
..a N- .
i FUTURE WORK RESOLVE ISSUES WITH THE lNDUSTRY AND NRC 0FFICES-PRESENT FINAL PACKAGE TO CRGR FOR CONCURRENE RECEIVE ACRS REVIEW AND CO MENTS
<O+
stb 11T RULE TO THE C0 MISSION FOR FINAL APP D'AL O
12
ges@'
$7
.O.
ACCESS AUTHORIZATION l
HISTORICAL PERSPECTIVE l
o AUGUST 1985 NUMARC EXECUTIVE COMMITTEE ENDORSES GUIDELINES (REV. 7) o OCTOBER 1985 SECY 85-381 PACKAGE WITH GUIDELINES FROM EDO TO COMMISSION o JANUARY 1986 C) -
NUMARC EXECUTIVE COMMITTEE ENDORSES GUIDELINES (REV. 8) AND AGREES i TO IMPLEMENT SECURITY PLAN CHANGE O FEBRUARY 1986 ACRS ENDORSES NUMARC GUIDELINES o JUNE 1986 COMMISSION VOTE (3-2 AGAINST) ON SECY PACKAGE (SRM DIRECTING STAFF TO DEVELOP POLICY STATEMENT; OBTAIN COMMITMENT FROM NUMARC TO INCORPORATE GUIDELINES INTO PLANS;
()
.C) !
HISTORICAL PERSPECTIVE (CONT.)
O OCTOBER 1986 REVISION 8A ISSUED WITH MINOR CHANGES MADE O MARCH 1988 NRC PUBLISHED PROPOSED POLICY STATEMENT FOR PUBLIC COMMENT ENDORSING REVISION 8 OF THE GUIDELINES O MARCH 1989 COMMISSIONERS PROVIDED SECY 89-098 WITH OPTIONS FOR RULE OR POLICY STATEMENT O APRIL 1989 COMMISSION VOTED FOR GENERAL RULE WITH A REG GUIDE THAT WOULD ENDORSE THE INDUSTRY GUIDELINES l WT'M EXCEPTIONS AS NECESSARY O
4 D l HISTORICAL PERSPECTIVE (CONT.) j O MAY 1989 NUMARC/NRC MET ON TWO OCCASIONS
)
j TO DISCUSS POTENTIAL MODIFICATIONS I TO THE GUIDELINES O JUNE 29, 1989 SUBMITTED REVISED GUIDELINES, REVISION 8B', TO THE NRC O O AUsuST 10, 1989 PUBLISHED REVISION 8B AS NUMARC I 89-01, " INDUSTRY GUID.ELINES FOR l NUCLEAR POWER PLANT ACCESS AUTHORIZATION PROGRAMS" 1
0 SEPTEMBER 27, 1989 )
NUMARC ADDRESS ACRS ON PROPSED RULE I I
O !
- - - - - ---- --- - --_-- _ _ _ _ a
i
.O ACCESS AUTHORIZATION EREVIOUS INITIATIVES 0 AUGUST 28, 1985 "USE OF THESE GUIDELINES IS A FAVORABLE ALTERNATIVE TO PROPOSED RULEMAKING ON ACCESS AUTHORIZATION AND I ENDORSE THEIR USE."
O JANUARY 30, 1986 C) -
"THE ' GUIDELINES FOR INDUSTRY NUCLEAR POWER PLANT ACCESS AUTHORIZATION PROGRAM' ARE ENDORSED AND WILL BE INCORPORATED INTO THE SECURITY PLAN AT OUR PLANT (S),
AS NECESSARY, TO MEET OR EXCEED THE GUIDELINE ELEMENTS."
l 0 BOTH INITIATIVES WERE CONTINGENT UPON THE NRC ENDORSING THE GUIDELINES IN A POLICY STATEMENT O
1
l NUMARC 89-01 ki l' .' p+ I - ,, i'
' i '.
. rl t ' . ,
Industry Guidelines for O Nuclear Power Plant Access Authorization Programs August 1989 i
Nuclear Management and O Resources Council, Inc.
1776 Eye Street, N.W.
Washington, DC 20006-2496
'?
W
" ).
5 4 t
TABLE OF CONTENTS Paae 1.0 PURPOSE ................................................... 1 2.0 SCOPE ..................................................... 1 3.0 RESPONSIBILITY ............................................ 1 4.0. APPLICABILITY ............................................. 2 5.0 GENERAL REQUIREMENTS FOR UNESCORTED ACCESS ................ 2
'6.0 UNESCORTED ACCESS AUTHORIZATION PR0 GRAM.................... 3 6.1 Initiation of an Unescorted Access Authorization
. Program ............................................. 3 6.2 Background Investigation Elements . . . . . . . . . . . . . . . . . . . 4
.6.2.1 Employment History ........................... 4 6.2.2 Education History ............................ 5 6.2.3 C r i m i n al H i s t o ry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
( 6.2.4 Military Service ............................. 5 6.2.5 Character and Reputation ..................... 6 6.2.6 Verification of Identity ..................... 7 6.2.7 Credit Check ................................. 7 6.3 Psychol ogical Eval uation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
- 6.4 Temporary Unescorted Access Authorization ........... 8 7.0 EVALUATION CRITERIA FOR UNESCORTED ACCESS AUTHORIZATION ............................................. 8 7.1 Criteria ............................................ 8 7.2 Review Process ...................................... 10 8.0 TRANSFER AND REINSTATEMENT OF UNESCORTED ACCESS AUTHORIZATION ............................................. 10 8.1 Transfer ............................................ 10 8.2 Reinstatement ....................................... 11 8.3 Update Requirements ................................. 11 p 9.0 CONTINUAL BEHAVIORAL OBSERVATION PROGRAM (CB0P) ........... 11 10.0 SCREENING DURING COLD SHUTDOWN ............................ 12 l() 10.1 Devi taliz ation of Vital Areas . . . . . . . . . . . . . . . . . . . . .. . 12 11.0 GRANDFATHERING ............................................ 13
s_ . , . . _
3 ( ' h '. "
- (
t 5 4. -
12.0' CONTRACTOR AND VENDOR REQUIREMENTS .......... . . . . .. . . . . . . . . 13 i.
.13.0 EVALUATIONS AND AUDITS ..................................... 13
-13.1: Utility Programs .................................... .13 13.2 Contractor.and Vendor Programs ............. ......... 14 '
14.0 RECORDS ........................ ,,,,,,,,,,,,,,,,,,, ,,,,,,, 3 4..
L Attachment A, Minimum' Audit Criteria ............................ - A-1
-O
'l l
I
__-___.--____._._____________________-___._____.-___.____s_ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ - ______________.___-._-___.____m.___- --___-- _ a
9 Q ACCESS AUTHORIZATION
)
INFORMATION GATHERING PROCESS OF GUIDELINES O EMPLOYMENT HISTORY o EDUCATION HISTORY O CRIMINAL HISTORY o MILITARY SERVICE O o CHARACTER AND REPUTATION o VERIFICATION OF IDENTIFY o CREDIT CHECK O
_ - - - - - - - - - - - - - _ _ - - - - - - _ _ - - - _ _ _ - - _ _ _ _ , - - - - - . - - A
'.O ACCESS AUTHORIZATION INFORMATION TO BE EVALUATED THAT INDICATES UNTRUSTWORTHINE11
. o WILLFUL f?M.",SSION OR FALSIFICATION o DRus OR ALCOHOL PROBLEMS o CRIMINAL RECORD O O EMOTIONAL INSTABILITY l
0 POTENTIAL TO BE COERCED l
0 UNRELIABLE IN COMMUNITY /WORKPLACE/
FAMILY / FINANCES /ETC.
o PSYCHOLOGICAL EVALUATION UNFAVORABLE l
0 OTHER DEVELOPED ADVERSE INFORMATION O
~
ACCESS AUTHORIZATION O LICENSEE IS THE RESPONSIBLE AGENT THAT GRANTS UNESCORTED ACCESS BASED ON A COMPREHENSIVE PROCESS. )
O JUDGMENT TO GRANT ACCESS BASED ON EVALUATION OF ALL COLLECTED INFORMATION.
O PROCESS TO GRANT UNESCORTED ACCESS PROVIDES REASONABLE ASSURANCE OF O TRUSTWORTHINESS.
O BASIC UNIFORMITY PROVIDED THROUGH GUIDELINES WILL ENHANCE TRANSFERABILITY OF ACCESS AUTHORIZATION.
O CONTINUED BEHAVIORAL OBSERVATION PROGRAM IS DESIGNED TO IDENTIFY EMERGENT PROBLEMS.
O AUDIT PROGRAM EVALUATES EFFECTIVENESS WITHIN FIRST YEAR OF IMPLEMENTATION.
)
s' /
SIw.
o BRIEFING TO ACRS ON !
PERFORMANCE INDICATOR '
RESEARCH HUMAN FACTORS BRANCH DIVISION OF SYSTEMS RESEARCH OFFICE OF NUCLEAR REGULATORY O RESEARCH PRESENTERS:
THOMAS G. RYAN CARL E. JOHNSON ERASMIA LOIS O
1 l
- l. :
BRIEFING TO ACRS ON . PERFORMANCE INDICATOR RESEARCH l PURPOSE OF BRIEFING:
- FAMILIARIZE THE COMMITTEE WITH RISK-BASED AND PROGRAMMATIC PERFORMANCE INDICATORS RESEARCH COMPLETED l
AND ONGOING WITHIN HFB-RES.
- SEEK COMMITTEE COMMENTS ON RESEARCH COMPLETED,
. AND YOUR SUGGESTIONS REGARDING ONGOING AND FUTURE RESEARCH IN THESE AREAS.
PURPOSE OF RESEARCH
- SUPPORT THE AGENCY'S EFFORTS TO ACHIEVE A SET OF CREDIBLE DIRECT AND INDIRECT INDICATORS OF SAFETY --
WHOSE MEASURES CAN BE USED BY THE COMMISSIONERS AND SENIOR STAFF MANAGEMENT.
O
l h '
i ll BRIEFING TO. ACRS ON PERFORMANCE INDICATOR
! RESEARCH l: . FOCUS OF RESEARCH:
AT THE REQUEST OF THE COMMISSIONERS AND AEOD, RES HAS BEEN FOCUSING PRIMARlLY ON DEVELOPMENT AND VALIDATION OF INDIRECT, OR LEADING, INDICATORS OF SAFETY.
RISK-BASED INDICATORS:
- SAFETY SYSTEM FUNCTION TRENDS (UNAVAILABILITY OF SELECTED O . SAFETY SYSTEMS)
PROGRAMMATIC INDICATORS:
- MAINTENANCE (ESF ACTUATIONS, DAILY POWER LOSS)
- TRAINING TOP-DOWN INDICATORS:
- ORGANIZATION (ORGANIZATIONAL LEARNING AND EFFICIENCY)
- MANAGEMENT O soTTou-uP INDICATORS
- SUMMARIES / COMPOSITES (COMBINATIONS OF INDIRECT INDICATORS) l
q E I
( ,.
l O, BRIEFING TO ACRS ON i PERFORMANCE INDICATOR RESEARCH ISSUES 'OF INTEREST:
1
- 1. C_ONCEPTUAL FRAMEWORK (S) FOR IDENTIFYING CANDIDATE INDICATORS
- 2. CRITERIA FOR SCREENING AND REFINING CANDIDATE INDICATORS O
- 3. CRITERlON MEASURE (S) OF SAFETY WITH WHICH CANDIDATE INDICATORS HAVE AN HYPOTHESIZED RELATIONSHIP
- 4. VALIDATION TE'9NIQUES FOR TESTING HYPOTHESIZED RELATIONSHIPS BETWEEN CANDIDATE INDICATORS AND CRITERION MEASURES OF SAFETY
- 5. RESULTS OF INDICATORS DEVELOPMENT (ISSUES 1. THRU 3.)
AND VALIDATION (ISSUE 4.) STUDIES i
O
HE S CL E AP C
, EI N T E S U M RL Q E OU E T FM S S N
) Y DN O RS S: NO OM -
O U TE YI OD X AT TT BE CS EA S Y IY FR RA T DS A EB I N SE P L IY H PF I T TT UF B EE N ) A A CF AF E NT L NA CO C ) AS I AS I I L A M FS V A HC V RF IE R C TR OOO F
NU GL E
S I
T IN W N U
A RY IA I Y ET SV F R OB X PI - O C I L KY TT G j
DI SL T T AE NS EB IR U N RS IT SA : RE O A TN AL T L S ,
AE S R M P IM BI E IV
- A DA E HE TE KV V EU T S TT ,
L I SA O TQ S R S .
N IN V C Y U GYE I RU EG S O NS .
( N LN H I I I EI S ( RY SD R AT; T N U PES P RE O MFA E OR H OAR C FT ( CSP N )
=
O )
C 1 2
( ( K E S O H I R
T l l
e r
n e l u
u) m v i
a sd ae te s E F e av se t t Mle s yr n n Su a c
e nD M ya l
i R m o A p eF t i i e r R n i u
em t i C f
a i g q i s
i- T S S S E s y
C, ( l a
n A
S )s l R e a c
O T m i t
o i s
A c t C t u t a
I y S D O i t
N I e s l
i s b E
t a e o l a
C i d m nL i
a N e o o v A c ir A
m r
t u
t aw e M
R t e O u t o e
c n
AcPy r
O I e u O F R
E n
o t
i d
a Fil Sa o s sR eE n P d e ED RL i o
t e m a C
I s
u r
e cs T c t Ri s A o I n l t y nl M F ea dn M l y I A
A t e
R n e cg r n G s ui r O e r oe R P :
s Se n P (
m ai g a
r t
an g e DE o c s r
P r n o a g m t n n t
c a
r r e a a o p
t n
i i
n F o
r p i a a r
l a
P u M T n S i o
l t a a i
c i z
n n a
o h T
c e
d n
a O gr
- 1 1 1 I
m t
e s
- y
' s y
S t e
D f a
s N
E t n
R a c
T i f
i G N i n
g N
I O
I k s
9 T T i s
r 8 E C , 9 N d e
1 O
E U t c r F e e M l e
s b
M f m S E o t
e R T y p S i t
e C
l i
Y b a S A S l i
a v
Y a n
T u E f F o r
A. o ,
S t a
i c
d i
n O A n
(
S E
C N
E U
iQ E
S N
O K C R
O W
E M Y A T I
R L F I R K O T
L S I S: 80a OAU I R
M E
C I
T T A S N P Y S
E C
N O
C G N
I T
r O
E I M E
T R S O Y T S A -
CS Y IE T DR E NU F IL A I S GA NF F I O H TM SE R S C IT O T A XS EY T
A E
G D S C R R R EY I
D A
T P VT N P OE I S A F V TA D NS E E E S D C ME A S U R I ET B D T : H C VE N I S R R OL T E N M / /
RP NY R G E I I PM ET T A T C C MO NI M S P P IC OL Y Y L H PI F E S LF MB I R R R AO OA T A D O O I CL N P E s TR /I E M T W G I NE NA D O C F D P EB IV I C E A E H TM AA L OU RN E PN TU - - S - - -
O 0 o o
I L E C I
I T C B S l P r R R A Y H o/ l C
L O
I S P H
A Y T V
- A T A E C
N F s
- I U A G SI D
- D _
E F S E R U
N I
O D L I
A FO S E N E R T I
A T C O E R H
R T
T L W A A E F A
r R ol /
F O
D I CS I
C P
L R
E B
M D D F N O U
N N l o A :
S A C M E
S N
D T
A T I S A T Y R U S T P N
I
A O
T D
N O
P S
E R
T O T N S Y N E
T T A /
I C E D L C E T N R I N A E B E N V M E O A P T C O L I U
N I
S I
I Q A D A E M T
V T R E A A O Y N H T
F T
U E O E C N M I I V T T R U E B F S O ,
F D N O E O L ID T I TN U A CA O F AM RE FD - -
O o
, I
E T
N I
T S
E T
E C
N A
L L
S I E
D V R
U YN S TE 2
/
ER 1
(
FT .
- L A
A ) C S N S I E T R I O U L
R C
FI T I A S OO C F R U
O R N
+ H -
O F U E C
T R I
V R
A M O E S
C E I
T A F O
DT C I T NS D N
U O
I Y I S
R S Y U O
T I H L
I B
A L
I A
V A
O N U a l
l ,I
j T
I S N E F A C O C N I E
U E F Q C I E
S N N N A G O M . I I
C
( R S S O D
- F R E K R A B S .s / E D Y P N N I T A A R I L G T C I N S S B I D A GSC N T L YM S
GIR E C I A TE A N R E V ET LS T F A N
FS
,Y FMW D F U S EO
[ ( FTL N O A OSE A YB E Y NS D T ) j O Y G IYT U I Y N TT L C I TTI IN N T AEL I E AS CFI NI B U IT IAB GO Q TN DSA AP A E IE MD L R NV M I F E ITL N I [ ( NE RA A EAR OT V NC/ TS A OIY A FT CK N ,I I IS U ENL DI RGI NR OIB I FSA A E L ,
o o
O E .
S I
HS TT E S N C T TEE NR .
C NRN AOS E EOO N M VTP EEE E L ESM TTT B F REO NAS E PRC IRY OS R A S SS OY& ME LE S TL RY AN S S TS EUT E M E SPE VLE NV E N NMR IIF AI TE OOU TAA CT IRT CFS C S V TPC E EE Y I C U FNT DF S T AOR FIN UF C TT E A TE F S NDC I Y E OD INI NE T F N , EF GC E F EAS FRI AN F E T M OTN MA OAS E A ,E GFT NGI G
DE N
EOS ONS NT C R Y II - AN F N GES TDK I O A GR AAS DA N AUF CRI NM LO IGR E Y E EI DE RA T T HAN NDF T I N TFO I O M L I I R RO I A SRT EOY OR IOC N T TF B M N OLI A A ENU EL CT .
L COF ,V I INT I NI EEB DAN A ATD RLA NCI NAE O L IIO V EDD FHI FP A TAN EGA ,I D N NRE RIV SNN U IGT EHA UGA AEN H N HIT MDI TAU TSS O o o 0 _
l 1 lllllllb
T A !
i H Y I _ '
T L 7
, i 3 D
SI P
_ l 3
3 WA OR _
l 9
2 H S ,
S S _ l 5 R D _
2 E
T N N _
1 2 U R
A O O _ .
+
Q OI T P i i 71 F
0 A ES i
?
O N
L 4 U R 4 31 i
i M R i i 9 e SOT I
RA M@ 5 i
EC TI 6 4 2
1 8
6 4 2
A O
. 1 U D 0 0 1
1 1
. 0 0 0 0 0 0 0 0 0 0 PN I SYS[M hf B Y 9
M
- CO C
Y F
I T
E N E
U D Q
I I
N Y N A C T H I L
I C Y R
O Y Y B
A E R T T T L T
I O
I A L L I
A O T C I
B I
B S I V N I D A A A H N L L N O
I T I
I A
I A U N R V V T A E A A G N
A L P
H T
N U
N U
I G
D E E
H N
I H W A L Z W G O H A Y I H L C L T V A N
S - - -
E A T
- O *
- N I V N-I O A T TO I
I SI S C N T U E A-I D
L C N SI C ?
I
,D N
O D I
DN I
I C L L A A V V YE S BC D A N S ^I T Y M U
E W M R
T L I
B A H F 3 OO L
L F R L L I
A G
I H
O O E F P 2
OU V ,
^A N DG EN S 'U N I W
DI V AO E ' O R R GP E M R DI E
C E N
^ C A S
)
N )5 ND M 2 A EN E
T
- M& TA N
^t R A EG I I S n O 4 a F CO l
P R ,
M NR
( E #3 D AP) 1 P E L L Y#
T s N Dt n VIEI O Tt n E O a R VL a
^V Ol P RAl E G (P UUP M SQ(
I
' ' ! {l O
)
S N
O S I T T N A E R M E E P R O I R U R R O Q O O T E T C S F A R A I E C C C L S I G I B R N D N D U O U N I N P O I I T I S T R N N O I R O E O O T A D O P C I I T N F E N T T T A O R A A O D C N M M A
M O
N L
O G R R R G I N O O O T A N T I F F F U I I A G R N C N B T T M N E I R I N I R A P N ;
E M O H F F C T I F C M O O O R O L N E T N P ( I E T T T V S S D R O S L L Y O E A T G L O S M T P O A V R Y E L N Y S O S R L E I T E P E A B G D E D E D T A N U D F U R U N L I L L A L L U I T C U S C T C L A A N O N O S N O V R I W 0 I N D I V A E P R P N P O - - I - - N - - -
O o o o
E V
I S
S E
C X
E
) ) E S S G T D A I R R N P U U N O S ( C I 3 N S 1
(
A E T
Y
- A D L A D L T U A A E O N D L W R A T B
A R S O
T L N L O E A A I T R C A L P
A A U I D
N V C L A I I N O F D A I I O H N F T S T
I I
T E
H I
D I W R N T S E E D Y R H D N A L O T N I A T E E N A H P S A C W E A I D I E D E B T N TG E E I AN D S L UI U S P T LT L E M S AS C S O E VE N S C T ET I A o 0 o o o
E FO AT S
K TNO NIT AL C T IKN FCE IAM NBE GDG IEA S SEN T L FA N A K M E M
E FIR P
T OV OF N M O YRO O I P TP I O I LDO N T A
N O
E ILI T I L G BUT N T A N AOAS E A W LCCE M C E A I IV E I N R AYLI L F E VLPT P I R ALPA M C G NAAI I E E N UI T P S O TEI L S E N L FNTN A C E N OEAI 0 L N C O TR G A A I I SOGY C N L T RPER Y I E / C O TO T N T G E TSNT E H N N P AMIA F C I I S CE L A E A G N ITPU S T M A I DSLG NYEE ISHR - - - - -
O o i
S 9 O F 8 Y OM E G 9 Y T YT N I
1 T
I I
L TS S N -
L I
B I
L Y I
M I
OI B I
S S T A A B O I L AY R S NI L OAV I
A L T P Y I
L A E I
V S S A V F A I A
A T N CN S A S S A EU Y I N UT E
T N D
F R Y N O CO T F Y A E O C A T I
T R N R O M F I I
V A OS A RF I D RTM M S OI N T E I T G I
SI V L OAE U F AI N S A
V FCT OS I
S CS - E U SD M S L SI N I I E E DK E T T C T A NS &N U S I I O
E B ED B D R
N C L SE I
Y P L UT RS ED O M A T T PE T T I
O C TY OT A N O C I OL AT L C D N NE E E E I L H/S EF VL L T L CE NA EE AU I ESF OS DS VB W TUO O * * * *
- S R
O T
A C
I D
N I
E C
N A S W M I
O E
I S F MB SF R
C R AH R
A E E P
C I
T A
M M
A R
G O
O R P
GSR N
I N TO I
AA RC 9
8 TI D 9 1
DN I c
e N D O AC I E T 8
8 C MA 9 1
N h c
A M r a
NA M ER TG NO I
O A RP M
f o
s n
i o
t c
R a r ;
O e : n F
t e o n m i
t a
i S: f e r s KS o h u e RR c D u l
a S s OO ys n
/y s WT a e o c
n i
y EA ri t t i i e
t u s t MIC r i rv oi t
a u p o l
i b
AD pt c i c q e
t u
t c a RNI
- a f i
s r O F a l i
a F l el s F /
s v L E da a /y s n A C o no l C
t e a AN i
U Mit l a c m a T A r u o u t PN s c o r a EE se n u t
a Q P H D CNT c f c
NI ot rn i
d o o o o OA Pa I n
CM l
p -
- ! &I ! .
jk 96 .*} J f. '
s s e e d v e
i v
o e
s.
i A ru R o As L e c l l
h t
a c
a c
r u
o s
e w AR C R x x xX X x x x xX s
r o m x x x X x x x x J t F
c a
X X n
a m
T r
x X xk x x m
l u n_
i i .
. x x m
c d
a a
wr xXX x x x x xX X x x x x x x xx o
w y
ms u XXX x x X x x xX x x X x xXX e a c
s n n t a o P n s a x x x x x XX X x x x te r u
e n o c ia n
a M y x X X x x x n r x x x x t
e F n ? 61 45 i
a 4 0 3 M
? 1 6,5 0 00 003 M 24 0 4.M ? 3 6, 1 1 63 0.6,5 435,2, 1 1 1 1 6 o 4. 1 8R 3 05 835 1 2 0 R.
5 234 63 1
f 3 S 1 4 5 5 1 o
n P,p p p P, p P, i
o g, P P, p p p 0
P, p p p p 0 0 00
. pp o, p p0, p,ppp o 0 t
a l
u ,
a a xX xX x x x x x x x x x xX x x x x x x x x x x v t a
E D
, )
2 s s e em r
l lue i t b a I a F de T - t ni r
e g e ap c me i
n v r
i r ipr e it o s n uy ql S
n o E n t
f o s e
o i e t r dni n g
M e is v
e mu oq O u l u
a s t
n e ia i
o )e c R ta e R f r e r t
n us d e F me eT i
t r u e sF e e d s Dc d d nt r r d e na ne s o d u gt s a o m
p ev ra d e la o i
l P Cromd i p x
e c
t uMc u ei Og p s ra g r
utuir oc c q t l ce o O (s vE sU i
E r
der f
eA n P r
d i r
r eP my er T s e eD iaAP F y y s nP e o e e c paS i s s co t t obb r
o ulam i
t c r O i t t r ia oeoB eurod r oeu r
s i n cd edd u ee E mT t
t nr e d ez a FR nf nn r ees gm -
kr ai n nr o aI E end d c c t t e
un t
c nh oTL ol p( k rc e ya v oi t o RSef f wf aiam et a f
t ne met wwRt eeo i a sr t
nfonP e
nid o se m
s ee etee mDD t n a
n n ota r ei o msteBO RSSMT r t i t ee ort t
iai t it e os BBT u o t
r r n nt s slt st n Cofa Molpb e w wluuOi a Scee eeefoia D p i f
l c c c c n nnep eee e ed ae o oaaC nn DFFLM immm& R f
oel t aa a aa ge i i r i vACDD naiar nn nnnAR eee TTT e1 t
nfoooo f f f f f f f ooooo f
ort ee nn nnnb 2 e eeeee it c auii nn t st t t nnn t t aa aa a mt v eeee t t t t t t t a ed n4aa iaa iaee i
eeeua r etaaa r
t a aa aaa Fl r l I h M MMMMM MMMNP PRRRR RRRRR
, [ll,l l l ,
s n
i o
t a
i c
n u
e m m
c o n c a h e n t l
b e
i e w l a
t i m a i
n e d e v a
a s h c
t a y Ml P i
S c l s o i d
e r s a g c o s e n te n e
t a
a r 0i ed n a f r
e s
i d
c n
s e
u s
t a
a e s d i
r i s y e ed r P r
t o d nr cn i
h c c n a n a
c y teee r o i t
w n
a S o i
t a c u n q t
c m Ci d d e l e o e a s s r r t n tef a r c fr f r l
a o o
i c -
du og s f o e t a s y y n n t r ai o pe l
l ul t s a e "e f e p u o i l
t s c o S t pt a fei e mi i d
h yt v i
ne u v i ui u q sa c i f a r n sh t l i t ot o ai h e rt c i
r mt n r p i
y o u e e e
r o - t r oe ot r e ou s r
o he wl ff gd l
t f yc f f f a e n e t a
ef e/n
.l i t e
i r e si r ei e nn ot r p c ad ct r c oets s C n e e e nt p ci nm n e
i t e y cotc d r s r e r ct ed r r i
d r l p" ei e s o e er x e e a e d p l f
f el em f i ed n f
ed f
e a e p e e r ab r b r r r n r n
P - - P - - P a P I - - - -
O L
S r e
R w O o P
T t A a C
I
't r
o e D c a
c n
N I
e r
a m
r e o E l i
h f r
T w e p
A s D
I i
n o
t n
e D t a m p
N t u
c i
u A a q e
C )
F o S t E E d O C (
t e
N e m
e l a
e A c n
t s r N t s a n S y s e
E n e e y c T
i t t v n t o
e e N i a f a
C D
n I
n m s e A i a
r d S M
/ m g g e n M t n a n
)
P t i
s r
a g i r D i h F n t e ) A e a c o s e L s c p e g O r w t i n
g P - s r
o n
a e
r n o n D t r n i
r T r ( C n r e e n e o t E sD e t e e S e v n e n I
t i u t
n sM o( e t
n i a w i
g L n d e L = e t n
u t L
t r m m e e s r b g e r P a n t D n m a
v eD w
p i
g a e e f
o E o r r d
a o g l
a p m g c n e e C
i w s i n P o r r t
o U f f f
) y r f f n l k
o o o>0 l W o o a c ODE i
P a e a
- # # ( D # # # M B R * * * * * * * *
- a o s r n he t t a a c ot S d i d
i t o E r o f n l a r U t i
e o a o n r r
Q c d e i e e
o n c
s n e I
o N i r
g t w l at a i
H n e e c ed b i u C e r e t
s l
a E :
s t wo f m i
t a e v
e i t
OT S i
S sb y t
y s m I
y I S t gt S e a a p e a fe l
N Y L nif a l
/ a Yh t g L
r O A a hs s d s A n o r n n a f Ni a p I
N T A oiot of l e o Al pt A
i t a s e s x n L al e e e G el a D A l
e r r n
t a r N p I
L C
I r
r t u
s o o s ai t t a i mu s I
Rd n g T c e e a s e E a n A S I
s T miE E
N y w i
V T A s -
c o -
m I Gi f e o d s d r i v
T r n s N e e o S C a a n
a EV r
, l
S P
I H
S N
O I
T A
L E
R L M A &
C T I
- T O S S T T I O L U
T A
E S T U E D S
R S S D F S S E S O I W E L S O Y H T R L S N E A E W N S T O A R R P O E L T V Y A A D L C C: A I I IS N A T DO I D S NM I I T H - -
A oT O T wI S TW
n e
m e
g h a t n N i a
w m O
I s f t
o T n a s A s l
p n e
D p u t r
o k o
I e n f L t a t s t e
i c
A )d r w
i f
n e n e r
a V i i e M
& e n g
mlp s t u
i I
T c i s s c u n e s O P ni a
n s s i
d b
t m s t s ,
o e n n a n e E n r t n o e o s C o f
i i
t v P r s a c e L d e N (c S n
o m a f o
A S
e s
p A I S
i t
a P
L e
v s e b a t n
a N Y L t u A i
t c
n r
d e s c E A c a
S e r t e c n o
i f
i T N A
h r o
t a
e r
i t
n g
N F t i
c p p cd i I S w o ane s G E n A
y r y N d i c
t et v t I t n e s s e M R E e r a n r
S n
e r a
i t
c n t f at s
E t
r p e T u e e a s e e L q p r c s N v mt t U e p r i f t o a oi I
G S r od O N d
a c p E F A C n g N n i E I n - - R - - -
i s - m l 1
e O
d ;
O {L
.s I i
d .:
- m. ! e
$ p g
g hh C
] e c o
O ! >
S i @
d $ e
}
x . Y. {
i o e m Od 'a 7 %
, fu o <
1a p 6
L l1 e
- 0) 0 '7 0 S s
$)k fs bbgQ 8 o .
agBTA g Z
- 4b4
/
7
\
u--
e n r i
o l i
u t
a a m r f e e t s
d m y i
s t e s n s y o y t c s e r / f e s a S d
)
e t
n s S n v e h u i n e t O d t
c e
o p v i
w L
i o j m s i
r b s e e o o a l b
R e y p ( c p i t
E v t l e e )
P r a p
a c e o i
c S P D
a eit pl v m n a B o
y m o
O A a r e c eh m(
r t b
d c P -
e wt f o n a S n a
s e
l ore r l M t s
i Y C b e P A t
i D aP v p f R s e r f i v
L M d n y o t o C o w n o t c
I l n a A / ei p a ge a e Si t o d a
)
eD ar l
p c od aP e D P D d f nt a e u t d t e
s c
n o e l
a n c e a te n l
A m v u a c n a u e a t B u a u i
n
- c b l
l e
mg L a oi pFd a t C a r nS e n n x r P i
t t nUER o i
D a eD s s o a m
i t t M v = c e c - - - a
(
m ae e t r
=
v l f
a h e P
= = a e l
e c p Ot O L DPl R Ri O Bi P CDP h D MAD o ow o om
.y 3
, r :n
-- M
. .n
-T "
f '7 (#14:Dy)M1 ,
5 i
F m e an -
f
. i 7 :*
n 15_
1 l
- . pl s a v . - o u m , .o F g ~
E M i 3
- I*
i (dul a w)dul u l
. i ' c, 5 l *N I
- i e
~
e k *e e (P" d. :DTWoR)#1*b ~
N O
C j .6 m p andeX d:DT 3 , m g j E
~
A : l 5 e
= . -
w O c-
^
O 5 _.
m #, % -- -
G E y~
I o e 'o N
.CL, ,e
- n 3
.re E eG 8 -
2 4tr o E2 ~
1 J l-
< ,e
.S m i duld uq g3 m ;
Q '
e o
_J C 1 - ,
\
- i
'O-
,O
- a o
a Y,
- O g o o o O C C M SONY
4 'l' Ie 1
}
- ..g.
.i (m t
l.
I.
l' AN01-1985 T. of POWER LOST after fft process (JAN - DEC) 3000
~2500 -
O g
2000 -
i 1500 -
1000 -
500 .
. E h bA t_~.. -
0.35 0.4 0.45 0.5 0.2 0.25 0.3 0 0.05 0.1 0.15 W
O Figure 4
.[.
r s f o
p ) g t
e S D i l
n P e S (
f u
t m e r
x u r o e t c t N e p
g n s .
S i s l i e s y oo s t y
f u l a
i s g e
r n L n n a ei Dl e m g r o n e r fu r i r
e e f e
w w r- t s e o o n P e i s n i
x g P e e e n n w e y i t S l
i me a b D m r
)
a x e P o f
E f r I m it I
e p
i t
w n n l a
)
t ni o p t at e n sl pi a i
t c n a n o o M a e s
)
r
. (
l p ms a a r
o e ) f p V o s S d O) R E wn f o
d e
- R e
(
r S t i h e r
e s e t
) ni e O
i h
p T f
i t
O i
wrg t a
T n e F (O (M R OSod e g A d i
S E )0 c e F e C e
( <i d Mt n b l
e r P v r n y a l a n c ts l
e < e ba c s i n F 0 S i
wl o min n w af n s )
S( po F a a l
e r . E g pl t
t o r
ed ut E h s i a
G i n
a t
a p et r n
r h t
i t
wdet w o d N mi d c O s a
( ut t u i w e w t
n I
a s e a N
I h
t i
n t yC t R e t
a h o n a l
p r a i t l A
i o a e h .
l r s t "t
i g
w b o sT l u e r r e n R e p : l at a me r r
o o r i n
mu i
T d a re a o ct a o t i e v p r m a a o t i cV c m r s s A OS cT sR d i
t g e r s n t n ff a o o a n OE n o :
s d
e k
r i
n e e e F Si f p l
o i r S T o i a
wt a
r e M# # M :
s E MN So t e t e
e w - - - - t - - - M S D m "e l
r
- u t a n u s
- x
- o e e O F r i N F R N
s s
e n
S e R v O
i t
Tf c A offe C sE I 0 r 9 D ot 9
1 Nt a n .
c e I
c e
Ti d m D C n e g O EI 8 R c n I
a 8 9
1 Dit a t N
I aM m
O c
N mdn d i
o r
Wrag a P e
O o n D r o
- Pit P a O i z
T n a
g O r O
. l
l(/ 11!
~ .
O s
r yt o t
a Y fec i T
E ad F Sn A
_ S S _ _' H E T N I O W C _
S T R U e O O c n o
I y
c n i C E a t a A n i l y v F T
A i e p t i
o -
c ml au n L ID n A o o i
f E f N
c oi M E -
- O R I T
E A T Z N N A
" " = " G R
S O R G O
T N K
C s j
N A $_
_ l n L F t n _ a 6 s L e ne oc j
t t e R A
N m t _ i t n a
z n e ss O F
O n o
x e a z a i
n ge r K I
T r t n~
in nr a g m e co O R
A a gve mr i
Z v o r s e
n O EP W N
A E C ro D E N
G " O G_' = .i A
R R o O "
- _~ F l l1l 1l
K R Y
@ T s E '
e i t
W F y d
u E A S i t
g n i t
M l i
b i
k t
t A O a a sasseR R T i r
a Mn r r si nt ok et S re y i F E t V T n e i
l i
l i o aro/u l
e s T A o c t
a s e u a c RWgeR L i U c i N E t
a e n r e
i g e r ,iM/
t E R l ud P o D et s a g n l
o ,r nF e T e e s n g ai o M
E )d A r p t n h n ut aga e eH -
e cE i ql n bn a t f
t d eimv uT e D TeCN rn l
l eP e s aRi e S S e z f af
/nid eS o n A H r r Nit R i
oNl l
a - oT e AnO ot s c nt r R ar ot ba r P G n i
oT i
u n e t e
McCA t
aO de ab aE e aL k l
u l cVd ,P l
r
,nP m e s O FeSS i
Q( F gl e op e
eN ,
u t t
i G / E r
u s ,mWr s,oi Rt r e eh n t n e L g pL l N A nd g ,
d r C u ol s,e e ,or Ar e oA eOCibonL gi BuS Nd t
@ N r Ct e t ar ol n OoR i
l O u l l n n p aI M P nR yl aanl a epi v I l t
T I t a p A
T A T N
l u&I Cp&P l
S T u &e gal i
cT /aNT il t
o aort r
o ei Z Z E Xi z oZ onE a z n s r i
h y I
Ml a st ,r l I
i I
N N n r i n oE Sf oN iGiov N A O R o neit l
ot cT hnAiR t ni tr a c ekr a ei A G R
I V
i t
ai g a a wloe e Ot a Nf cGd eR aM r E g n p og r u u
@ O E NOVRR CSTOT E OFSWOPP N
R D
, l
K k c s a n
)
R b o s
@ di t ed c r e a n a W e c a w E
Fi n u
me r
o R M gm n f ,
r A m e s i
y r t oo P n
R i l
i iC t i o
F b a
n o n ,
,t s c i e n T
l e Mo r a duS i
R t N ,
ei c n e ,
E) y n g c g s a o on e Mde Y i
t r
g m c e P r i r
o t
a E u T E e r oR ( t i R
diA t
n F n f r s n e dr o I
e ty h t g t S Pi p a M a Nn n g ,b l
i ed d t
u Ao F O
i s
e s
et a d n n a a O n nt M (c i
D c u i S s s S i E :
l o os c e e o dr r u
G M e c
P c c s t n a : o O n :
A ru e ed t sh N e f n n C a c o e c n ta ul p a o
@ T m n s D ci I U r o a e r
aS m
t ui t t
a T O f m f or o oi r c r r f
/s C o A E P
e o o E ef t i
l c n f
Z T r y C u y,A A e o N I e t N n aC Y pt P Cinlb ecr O I i i N D r a
l i
b A n N f a u i
I T
A E Y w n a I
L te a n MT d a l P n E oiol A
Mim V i
a a I
I R L r m C s E A Ha u v a u I ov e O T H A OMH Fi F a AR t
N
@ N I
U Q - - -
C
- - ER - - N I
, l '
S S TL S U E
N S E E V R I
TY CR e v
EA i t
s FN FI i
s oy pt EM I d f e a e
L w s
M E of h o Y T
& R s I s L s e P
I r r T e us U S t
S E O S e a me R E
N C R
I a m O P
FS U I
r T O ah S A s OYL pt gw i
E E Z
T N
L O
R E
S E
R n G A I L TA V I
is A S wn oo L
P R R
E R
E NN l l
oa i
t T
N T
N F O I
L I
L EA f l e A L
A L
O J A
R A
R A
M e r h o r
P P # M E E P L T c - - - - - -
O A L C I
E T V S EI D TA P T I
S l l l
i i 4! ii al v
3 7a S F S O E N N O E
I T -
A V
I N _
T S A _
L CN E
P X
FI O Y E
L FS N L A A EU S C L A I
R MC Y I
P N R M E
& O O o O C E
H T
D E
T S
F Y G N
E O R I T
N A R E
B A Y
- TN A E L T T NI M L L A E EI I
D F
/ W A MLE G N K F
O S
P R I V R Y
F O O O P L O W T I
S L S E L I
E E M B A
R V M I I
A U EY R
- L S E L T E F S E
I A A
- D BE S U V E OF A E A 9 I RA PS H
P H
T S
S M
P I - -
cl l! : l l :
e r
u c
c t
o r
a c t a e l
e re h b
cd i t n
S u n s a NI t c G - f n
e t a
O n o 0 d 0 o i c
h t 9 L Nt n 9 c 9 y
A n et y
c
. a r t e
1 N i mef e e r i o f t a n A s p a D e v a s e ; e g o s v 9 D o e f l
9
- e s
h e f on m8 O R 9
av l
o 8 - b o 9 n e 1
s :
A AD s e s ot n et s
r o n i
r v f 9 8
E y e r
o n l r l a t a e o1 L t u f e p c r e e t h a J od r n C f et c : y of e
i d a n e u n u
U a rod a J J i
d c i l S f n o n n s c n N I i
r e i n
u t
c n u i e
- f s e u s e d n c ot P qt r o e a N n e n e r
n a
t u
o i
d t t r
n n a rm O om f l pb i i
- N a n i
t z ro h
i g r a
e a
n f
i t
yd t
e t
k s fo r
e i i h l r
a n a r o P e c e s l l iv :
e t
U En e
v h t
u n L d n i
a Wfo i
- r e n o t
c ot sd e z o u o O j b
i l
i t
n t t
a e i
r O Ui n SP
V S T AI E HR S F I
E L F SA V AO AS O B T A R C E I
GN P ME NA H T M E R O PL I
HG I
C D E R W L L
- T D S I
O E SN R W R H T T AA O S E E T S R T S S A T O PO A E E DU CN T D T R I L I
D E A NG T VA NM C AN N OV I N DY I
R TO NIT E
I LR PR EE D O CR I I
L EV I I UF C S T PB RN FF C E A I PA T
AIC DE NP
_ I A
L D N R A o
RDN I
S s R Y I
O C SP E B Y E A T OP H TI F NG
- FN G D NL I L I N
_ O H DD C NU RR S T W N I CR I
EN OO A A OA D - T
_ S L CE I N TI HU E ,AM AN OO L
_ CO V OS U TN CO A F OE ND N DM I
_ UG H
- SO RN N E S L S EIT CG E I
I E C T A T T A C DI UN I S Y N NN N R A E T A A T S EA A A A E FE HM L L T E RIC P S P NI VF O SA S
TR O
M I
R ETO S GF RE DC S T O AL A I EE I NR C RM A E T I
E B E R E
- LO L C RITR R E RP E CR C AO OI T E L UP U O F' S A N Y N H CAF F SC N - - G oo E R G NR g NT N E N I
O H AN OE O H O I
T F R A N A N A T TR T HE R - S - - - S - T
O S
E R
U S G A R N RE A I O
EM H E S
U T
Y L N T T OI A C 3 I
E R U S F NET N K R A I S O S SR I
A T S N T A F GC O
T O C O N N N I L F E I 2 D S AO M N N E N TN I D
E K I
S R AEO T S
O I
U R F A C E SR I
S I O TT E A R A N V T N R E U A EG I D
M SHE C D N N AS I
R I
I I T EIL R D SU C
E MB A E N
I RD G N
A L O I R Y T T D T T SC S s C
E F E N E C O O P P R I P A , - I DL R N
.N DS NE P P S S E. O l
N T I V N Y R ,N I N EE T S K OS YE R S F TD EU A
I T FE N I FM A VO I I TN DI A KR NG OTY T N E T NO :
I L R OO 9 D O S ER 0 A E C AL 8 U E DI V 9 O I L
DA F I Y
VIT
.R N Y . .N F 1 2 .N I S 3 E F 4C 5.A
(,
/
)
\ w/
m y "3 as n
W Q L Z>
- 4>
, E N 7N.mg$,P c . .t h .y.*4 @,.u.Y.;g .v U.Y N.7 5 N.'f 7;&3
..ysj c . g .
4 p 2 W U ? St N k Q '^**ihY*% n ed W M*g M gg ~ "
__, !kk$UN Of ' UK mY w W Q. w W4$W3cybe%h ik 4 a. z N;% N., &weg;W;g g. $3 QO o w
EhgO I
(W g g o- -
E
.' m s m.t::ep: %. ;t .p: w:$.s.:. p.J. .
g -
e dig.
y H .T, m g w .E . x.4.J'Q,W :s ~ w;~.f 3.t.h 3* ,&D s.ym c plgl w.
CQp " T ., :hW. y <WhM:lWk E hb A$eh
~
s
>< ';%W%W +46.L W
-Q h NE)c p'"Y M ".
M- Q C 9.
3 $M ^ n@$g,MI4:n..
m "C m.&:&q' dMw'.4 p# o Q
2z O W-O.
- , Q.
.: [% 4 [.'MR%$ d, :>
?.NN d b
- W k
H -
.c' x; ;h;bt. O Q ~
w .
_o $ S:s '<: m ?(Bi k
~
Q w hk y$ b k O H
o 4d75,4 g n> m WM, 3.%eg s
t y p-
.y a
W E
w o
'g6 * ,(
W m p 3 t 1* ' iM -
P 1 M g W d,: ' $$ .
M . .? z
,t 1 k% '
'Cv) o w g >z d w~ M W' [3 g~'~n b
+:pJ I
H Q
y y :g _
~ ,.
p U >0 00R%. 'h W z z- Q&r w e Sh ~.
- n. e W
Oww M,av.,dbw --# N+
u.! 2".%y;my"h jgA'i b g
..g *- sm ag
- ( '.
s
=
Q
- QW ' 63t*MW}F#@":
- !t 1M%Ms& 43.
+
pag **d.a ""5 %J W N
Q.
M'd.li
". Q uW sh y-
.s h z w s
M yW a s.i.%m%ecqu, a -( -E
- s W H
W mQ .j,;tANVaddM, eeE.4
. y; ;.g;.- su. - 5 p.m- O E
p ( -. A g. Ma54 p +ed. s*$g F sng. q'.' E (9
9 .< M4 dw sa .. .: 8 M M
ri $$ Ibh $ *f b 4
M iif:fWp 'A;W; W^Uf W %tti tt;i rgg ts 9" k 4 g [ iD h.h[h'k[4I SklQ(fh[?q M
& M4M. hib!dhQQ'G; ( O 0h45 y eig(hfe;ps$h
- abe wA vp
' -nh y
3 we b w, g
,g N g- dyh, j@
d
,.+
,4
<b:.p E QfdMA$dile ty e maqa U GEE ygo, M:W p3
- g$"
g g-
@m% M&&$$ p$np((ffi[
reMbp' 2":tg?
D tpMdp 59: %
nit %
D
- E. .y ' ;0%*: $$hn. g. A: -
I. '. -
+4s%6F@fa%$%yit
/2:;M:P-f #,
M b
M N 4 WI;./e g a .MyC,T4.OT%.&
.. eg'F, ww ?.
v $_&&lj
- o .]*'Ii;% f. a;lqgfky
..ww V -
ff ket ,M y Iths;p) t
! +- !dh w bhMM cap m . a&'+ '% t R %9a 6
- !
- e .
,.x i::% , , + + v u:#2::
s g n n i o
i t
r e a v i c
c o l p
s i t c u r
o a yt y d t m l e
s a m ef e v
a a t
r c 0 a u s o t
od i n
9 r g q o m a I c
o r et e r
d c c e p a te d i i t D f s a n d a o r l a
n m 9 t ore s i
e t o p m
a 8 s s r a y c e s
e e e d to nl b i
r d h ei U g o
n u t e a i nt t e s s
- ai c s p o o a r J r d f l
mPe :
g n ei s m p
t o t d
o t
n e :
e t
e s
m:e a
ot n c e i i i c
t s r e n an v a
r g v r t o o i t
i t o e x B
p m
P a e r mce j f y r o
p j c
e f
k e o l n fo b i
t b ch af i r O nP O et C oe e eP h GD p 9 d iN 0 Ct o 8
- 9
- Y Y F F o * *
- , 1
! l : ll{ li b c eia t l
s i t b o r t
a a e ;g e o m
ht t h l l
c/s e wna
_ i h sn m es n t o
o
_ a r
r i t
mi ni ut I t t g e t c
I
.of r c o c o n n n r i u y. E sr fu p mn t r t Or efu o r a e s feD p y t l P m ar aa P
N t
dh e u
)
e e
s o er e f f t
ii c c dit r P a a n rc r
e od t PMws i v t n a
sNU dlr a d r o s e o c M ed yhi a tc nt i U ec cnb i i
f n y yn c a gr c e o l
u d n t
e b a s eei de m a
u d r m /a nc rmto ne l
r r p y eo a op e e q o gmf gh t rpw pf e o re n o pe sutbe d ( r yw l i Q t a e p v
e y l
a c
rf et e l
l a o a r s r h h
d te i h po vk e t y f c f e n v g t
l e a r ef r e l i l on s fe d s a r a mt yat i
o e e wr a n s l
P a b u c
s mra i h dl r l
ar gt a or n s a l
i t
at o i cw gd a ha dor po poi t c
r v oe ms or a s n/ ma t
ad me t l
h iP l
e Pit oyi n i c n a
r a R Ni mi l
l vo f a m i d a n d n '
ge or l
We P e s oy - tc n u i e
r ot u s E r mlst aoh i f
e twm pi t t V c n i
ul R or u ea dn t
I T f s h e ar mh c t
e u a Wo i,
st rp t
C oa b g t i a E t e e s ro P ga f i s e
J e c t
e p p i
r ot t a e f t n n l at h B s n l ph / i o e a t a a h s ci c msni se l
O Te dd p f m t ei eh o 9 t r s o ot : b cd s e i
l e c r c o Ci v i
E r u n S al s 8 k T cd wht s ote o t
ef r s ai t r a Oe l
Y e e a .
l e c r xl f l
.io e F S p T 1 r a Nd p e a o 2h g
' ! l L
9
_ saat eve 11an111ry la maatatsed r
I I Companong Coopeeest tame-failure rate is *** restore to staissaed stasaised I I = 1 Delays prior.se Burettee of Delays eve se start of assat. repair to pest maintamance enante is missaised ses:ing are elelaised ato saised I _
I ,
I fame to prepara Delsys due to 14sistssal Magnitude of the Fredettivity et hardware for adotaistrative delays ste repair assh Se saletateers sa constreiste are stataised statassed ar.elatsed q
l seisteneste to
( sistelaed esaletsed Pre-qualified -Spare Ports t y spoest Protective teviroceens prosedures available System limite seadutive to prod-sossegveattal damage estivity j Fre-tretaed .Stacial equipeemt of fe11ere e light barsommel es este o hast
- Ease of disassembly is e messe ,
part of equipseet desist a radiattee I i 1 *Mistas1 protective faae to gata
'I'8DI'8 '8981880 Taas required to fios requires to de-energis e drata aeopement estese to desaged 1 llotivettes and seaposent is is stasassed toeposient is eerste of errk sistaised einssised I force high Tagstes systes praias sonnected Equiposet imetailed to plant dratae for metatesmer assese effistency these typical requirements
'1 *Prepleased goalattes easily 1sterferentes with are satisfied by pleet progress other systees missaal obish beteos messes paths for propleased he ethievement of assessary atselettes devises toelatise S* ale.
5petsal lif ting equipeant ,
I asesseihte fagging protes pre-smets11ed l effitte6ty I
I, t Figure 7. Human-Safety Interactions i 16 n - -n
i O Figure 8.
ORGANIZATIONAL HIERARCHY AND ITS RELATIONSHIP TO RELIABILITY AND SAFETY, OR THE " DIAMOND" TREE l j
REUABETTY AND 1 SAFETY
= OBJECTIVES
/ i
/
SAFETY SAFETY (MAJOR l FUNCTIONS FUNCriONS FUNCTIONS)
= RTB FUNCTION
\SUB FUNCTION o
/ N SUB FUNCTIONS f NAL
CESS PATHS SUCCESS PATHS SUCCESS PATHS (RELIABILITY; o
ACTIVITIES ACTIVITIES ACTIVITIES (QUALITY)
/
I (COMMON i ; *ROGRAMS PROGRAMS CAUSE OR CURE)
, CORPORATE POLICIES (DIRECTION) i FIs l : NRC (IMPETUS) 1 / -C.E.O.
Stockholders /
\ \
Nation's PUC Directors Economy lO I
17
!l!)llltl i! Ii g d e n l
o a t
c i c
t ed s l
e s e i t
s a t a
g eb s t
h s s g g s t n n n
o f mo i s i n
s t
e s
o ai r
t u u gn c i
t a
l s o s r u s s a c e p r r n i
l f o o i f
p n y t t u e d t a a e t c s d
l enaf e i d
l e d i
c h t
d p a nd n r n
a ms oief s i o i f
o o ng mf o s cit i s s vl p c s t
e d e di t e p g i n s o n n c v a o e a a e l i
o d h
t t l r l t e e yt e a e l
p c n e r vdn g c m a a h o a e u n
m ul r d o q p g e g e o v ds i
g r i ct e g h o o t
S e d n e a l a
a E
y s ai k h mh t n t
e V cl nt et e s i r I
aa e a r e f p T u ni h
t r n o d mt a r
e C qf e e y nl d t n
E a t a a n s e i-J d nb c c ad n a e i a o msr p i
B yit f a tcth mfy O r n r i r i oi e t t omt n e m e e e t n
0 f m :
s e a i
e l t e e 9 r d ht e m t
st e k s Di c re i I S D d a
I t
Y e e a .dnh e m3 F Td T 1 i t 2 s 4 S l\ u