ML20248F409
| ML20248F409 | |
| Person / Time | |
|---|---|
| Issue date: | 03/30/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1726, NUDOCS 8904130055 | |
| Download: ML20248F409 (238) | |
Text
- - - - - - - - ---- - - - --
RC237-I'lild, OlG1\ A ~
O l UNITED STATES I NUCLEAR REGULATORY CCMMISSION
........................................................_q ADVISORY COMMITTEE ON REACTOR SAFEGUARDS In the Matter of: )
)
SUBCOMMITTEE ON MAINTENANCE )
PRACTICES AND PROCEDURES )
O Pages: 1 through 189 Place: Bethesda, Maryland _ . __ _
F Date: March 30, 1989 -
0d3,_F0OfII33011i;..
________________________________________________________=
O' HERITAGE REPORTING CORPORATION omw-1220 L Street, N.'W., Suite 600 j
e904130055 s9033o WasMagton, D.C. 20005 PD p,CRS PDC (202) 6M
( .. ... .n : - - - - - - - - -
'l (
4 l TRO4 0\
DELEI'E B. WHITE REIURN ORIGINAL TO B. WHITE, ACBs'+ew-y-3/r ,
Thanks!
Barbara Jo White M ;1407 .
.::t7 A N-
! {
1 l
l
)
I 1
ei 4 4
l l
l I
i I
O l l
1 1
_ i
7- .
nC PUBLIC. NOTICE BY.THE
~
1 2 UNITED STATES NUCLEAR REGULATORY COMMISSION'S 3 ADVISORY COMMITTEE'ON REACTOR SAFEGUARDS 4 MARCH 30,.1989 5
6-7 The contents of this stenographic transcript of 8 the proceedings of the United States Nuclear. Regulatory ;
9, Commission's Advisory Committee on Reactor Safeguards j 10 (ACRS), as reported herein, is an uncorrected record of the 11 discussions recorded at the meeting held on'the.above date. ,
1 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 e
O aerie e- a gorei e correr eie-(202) 628-4888 l
1 1
1
[') '
UNITED STATES NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS In the Matter of: )
) ,
SUBCOMMITTEE ON MAINTENANCE ) l PRACTICES AND PROCEDURES ) l j
l 1
Thursday, March 30, 1989 j 1
Room 110 I' U.S. Nuclear Regulatory Commission 7920 Norfolk Avenue Bethesda, Maryland The above-entitled matter came on for hearing, 1
pursuant to notice, at 10:24 a.m.
BEFORE: MR. CARYLE MICdELSON
(')
K/
Subcommittee Chairman Retired Principal Nuclear Engine 3r Tennessee Valley Authority Knoxville, Tennessee and Retired Director, Office for Analysis and Evaluation of Operational Data l U.S. Nuclear Regulatory Commission
! Washington, D.C.
ACRS MEMBERS _P_R,_EE NT:
MR. JAMES C. CARROLL Retired Manager, Nuclear Operations Support Pacific Gas & Electric Company San Francisco, California i
( MR. CHARLES WYLIE Retired Chief Engineer l Electrical Division
, Duke Power Company ,
) Charlotte, North Carolina l
l
(') Heritage Reporting Corporation
! (202) 628-4888 i
I E _ _ _
A{
2 ACRS MEMBERS PRESENT: (Continued)
'l DR. WILLIAM KERR '
.. Professor of Nuclear. Engineering Director, Office of Energy.Research University of Michigan Ann Arbor, Michigan AQAS COGNIZANT STAFF MEMBER:
HERMAN ALDERMAN NRC RESEARCH STAFF; MONI'DEY TOM KING l MARK WILLIAMS RON FRAHM ,
CARL' JOHNSON KEN HART GARY MIZUONO JOHN ZWOLISKI WARREN PUESKO TOM FOLEY
= BRIAN RICHTER MARK MULLIN-O.
O rit e- earei 9 ce ver ei -
(202) 628-4888
__n___________ -
3 1 E B Q Q E E Q I H E 3,
(] ,
~
2 MR. MICHELSON: Leh's be seated so that we can 3 catch up on the transcript and proceed. This is a meeting 4 of the ACRS Subcommittee on Maintenance Practices and 5 Procedures. Due to a schedule problem the first part of the i
6 meeting was not on transcript which was about an hour and a 7 half period of time covering introductory remarks by myself, 8 Caryle Michelson-. The other attendees at the subcommittee 9' meeting are Charles Wylie, Bill Kerr and James Carroll.
10 From 9:00 to 9:45 the staff made a presentation on 11 their approach outline and final rule making package and 12 from 9:45 until a break time of 10:15 we were hearing from 13 Bob Frahm cf research on a summary and analysis of public
(} 14 comments. The complete outline is in the transcript so I 15 will not go through it. At this time, though, Moni Dey will 16 give a brief summary and catch up on where we are at so we 17 can start back on our regular schedule. Moni?
18 MR. DEY: Thank you Mr. Michelson. The purpose of 19 the >riefing this morning is to present the Subcommittee 20 wich a rule making package and obtain ACRS comments prior to 21 CRGR and Commission review. We requested an ACRS lettei' at 22 the April meeting.
23 Scheduled for our review is the full committee on 24 April 7th, CRGR review and then the rule making package to 25 the Commission on April 21. Subsequently we'll have a
() Heritage Reporting (202) 628-4888 Corporation L
1
4
.; .1 briefing-at about mid-May.
2 Briefly the-regulatory background for this rule 1 3 ~ making initiated back in 1980, but the. review ofl status'of 4 maintenance in U.S. nuclear industry followed by a policy 5 statement of public work shop and a notice of proposed rule d
- l. l 6 making in November of 1988. The notice of proposed rule-7 ' making included the definition of maintenance.in 17 8 activities. .The requirements to establish, implement and j 9 maintain a program'and assess program effectiveness and 10 execute correct' actions. .)
11 The' implementation was three months for the 12 licensees to develop a plan.and schedule and two years 13 before implementation from the effective date of the rule.
14 Following is the schematic for the key milestones for
(}
15 implementation of.the maintenance rule.
16 The notice of proposed rule making specifically 17 identified 12 specific areas for comments.. And Ron Frahm 18 will summarize the comments received in these 12 areas.
19 Following.our analysis of the public comments 20 received the staff recommends the publication of a finni l; 21 rule, with some changes as'a result of the public comments 22 received. The definition of me.intenance which we discussed 23 earlier this morning and the requirements are unchanged.
24 Th6y are general, they are non-prescriptive and as I i
25 mentioned before they provide the floor and the focus for a 4
O a rie e- a orti 9 career tio-(202) 628-4888 l ,
I l
5 l
) [1 process approach something which we've expanded d o the 2 regulatory guide. So we think that is appropriate.
3 The scope -- we received a fair number of comments 4 in the scope of the rule and we have modified the scope to 1
5 . read that a program shall include all SSC's whose failure ]
I 6 could significantly impact the safety and security of the 1 7 facility as opposed to all structure systems and components-8' which was in the notice of proposed rule making. And in the i 9 regulatory guide we further expand what is -- and give 10 guidance as to what is meant by this statement.
11 MR. WYLIE: Where is that --
12 MR. DEY: I can't find that, I think that's an 13 extra. slide.
{} 14 MR. WYLIE: No I know what I'm looking at. 7. mean 15 in the rule, where is it stated.
16 MR. DEY: It's in the package with roman numeral 17 II on it. The package I used earlier this morning.
18 MR. KING: Page 55 in your package.
19 MR. MICHELSON: I don't know that that was 20 Chsrlie's question, though, was it? You wanted to know 21 wht e in the package that slide is?
22 MR. WYLIE: No, I want to know where that 23 definition appears in the rule.
24 MR. KING: In the rule, it's in page 55 of the 25 package we sent you last Wednesday which is one of the pages O erita9e aegerei e career eio-(202) 628-4888
.i 6
1' that has the rule on it.
I]
2 MR. DEY: It's in the quantified text. j l
3' MR. CARROLL: F.ight after 17, Charlie.
4 MR. MICHELSON: It's in the package that was in !
5 the thing that looks like this, the very end. Page 55.
6 That's where it's pu norted to be. Yes, there it is. l 7 MR. KING: Yes.
8 MR. CARROLL: Am I right, that the addition of the 9 words insecurity is -- that wasn't in previous versions of 10 the rule?
11 MR. DEY: The previous version did not have safety 1
12 or security., It just said scope of the rule covers all [
13 SSC's. So we've further defined the relationship, i
14 MR. WYLIE: And you say the reg guide is i 15 consistent with that definition? j 16 MR. DEY: We will present --
17 MR. KING: Reg guide puts in a lot more detail 18 that tries to explain what the definition means.
I 19 MR. WYLIE: Well I'm a little hung up on the words 20 significantly impacts.
21 MR. DEY: Okay, we can come back to that when we 22 present the scope in detail.
23 MR. WYLIE: Okay, all right.
24 MR. DEY: A major change in the notice, in the 25 codified text rule is in the implementation section.
l
{
O erie ee aegerei 9 cergereeie-(202) 628-4888 l
N \
E/' l 7
,( ) 1 Initially what was proposed as we talked about earlier, each 2 licensee would in three months provide a. planning schedule 4
3 and in two-years provide NRC with.the certification that 4 they have met this rule.
5 MR. MICHELSON: @xcuse me, did you say you're 6 going to come back later to discussing the. details of the 7 rule itself which starts on page 53?
8 MR. KING: No, to discuss the details of the reg ,
9 guide. This is the discussion of the rule.
10 MR. MICHELSON: This is -- okay., I understand. ')
11 MR. DEY: We will come back to the scope- .
1 12 MR. MICHELSON: I understand. ]
l
-13 MR. KING: The issue of scope is going to come 1 14 back again when we get to the reg guide.
(}
15 MR. MICHELSON: I have one comment if you've 16 passed the scope already. My comment was I think the scope 17 gets lost when it's put under definitions. The scope of this 18 thing is probably one of the most important line itema in 19 here and I think it'e worthy.of a separate section just as 20 well as definitions might be worthy of a section. I didn't 21 consider scope to be definition. I think that it ought to 22 be item C scope.
23 MR. KING: Some of our earlier drafts had it as'a 24 separate item. We'll note that as a comment.
.25 MR. MICHELSON: How did it get lost as a Heritage Reporting Corporation (202) 628-4888
l 1 8 !
f) 1 de finition,. then?
2 MR. DEY: Initially it was in the definition, and
)
3 you're perhaps right it has to;be in a separate section.
4 MR. MICHELSON: I couldn't envision it as a 5 definition at all. -
I 6 MR. KING: We'll look at that, d 7 MR. CARROLL: It could very well . fit under 8 applicability.
I 9 MR. MICHELSON: It might. I just didn't like it 10 hanging there under definitions because it gets lost that 11 way. I don't know what your protocol is but it.might be 12 applicability. But if you passed scope already and let me s 13 stop and I thought you were just telling us and was going tc ,
)
14 go back and go through this in detail. But if not now is
(]) 3 15 the time.
16 There are a lot of things that can significantly 17 impact safety. You know, it doesn't have to be 4 part of a 18 safety system. For instance the cooling tower basin 19 certainly can potentially impact safety. Is this thing 20 intended then to cover such things as the cooling tower
\
21 basin and its cooling system if it could rupture and water ;
22 could run across the ground and get into safety related 23 equipment? Was that the kind of thing you had in mind-here?
24 MR. KING: I think when you look at the definition 25 of scope in the reg guide which has a lot more words O eritae. eertine cerger eien (202) 628-4888
- _ _ _ - - - - _ _ _ - . _ _ - - - _ _ _ _ . - _ - _ _ - - - )
i 9
()' 1- associated with it, it's pretty all inclusive. It covers a 2 lot-of things in the plant.' My guess is the cooling towers !
3 would be included.
4 MR. MICHELSON: I would think there's a lot of ;
5 things impact if you use that word which is used in the rule 6 here. Gee, that'means all the on site storage of 7 combustible gasses.
t-8 MR. KING: Yes.
9 MR. MICHELSON: You might as well say it involves 10 everything that happens inside the fence and in fact outside 11 the fence for a certain range. I can't necessarily. store 12 highly combustible gasses even'outside which you might 13 define as your site boundary. A lot of things impact 14 safety,
[} i 15 MR. CARROLL: The other thing, Caryle, is on page 16 3 of the reg guide, exemplification takes you into things 17 that might inadvertently distract the attention of plant 18 operat ors . Certainly all.the things you're talking about 19 literally are going to do that.
20 MR. KING: I said we believe the intent with the SUL scope words we had would be to cover a lot of what's in tha 22 plant. Maybe something like a wr.rehouse or an office
- 23. building wouldn't be covered but pretty much everything else 24 would be.
25 MR. CARROLL: The xerox machine in the control I
O ritae- R rert1 9 cerror esen (202) 628-4888
10
() .1' room. And if I can't get: good xerox copies and I'm sitting 2 there trying to' operate the~ plant that's a distraction..
3 MR. MICHELSON: . Also depends on what I put~inside- l 4 the warehouse as to whether or not it could impact safety.
5 .I can start combustible gasses in.that warehouse area,and j 6 significantly; impact safety.
7 MR..WYLIE: The staff agrees with you, it covers 1 8 everything.
9 MR. CARROLL: I didn't read it -- I' thought that 10 might be what you had in mind, but I didn't believe it.
11- MR. WYLIE: Yes, that's what it says.
12 MR. KING: Really what it says is the licensee can 13' make a case for excluding some things'and. lays out the
- principals which --
( } '14 15 MR. MICHELSON: But I think they're kind of guilty
- 16. until proven innocent. That everything impacts safety 17 unless you can show it doesn't.
18 MR. CARROLL: I even thought of a better example, 19 a coffee pot for goodness sakes. How can you run a control l 20 room without a coffee pot that's working? If they don't do 21 good maintenar.ce on it that's really going to distract the 22 operators.
23 MR. MICHELSON: On the other hand that could be a 24 fire hazard in the control room. That could impact safety- j 25 so they better get it out. That could impact a lot of I
I' O rie e- Reverei-e cereeraeion (202) 628-4888 1
l
.a 11
() l' things.
2 MR. KING: Well, as you mentioned earlier there 3 has to be some element of reasonableness in all of this.
4 I'm sure we-could always find an extreme example. ,
5 MR. MICHELSON: But you really did intend, and 6 that's the way I read it, you intended this to be a very 7 broad scope definition?
8 MR. KING: Yes.
9 MR. WYLIE: but in that regard I mentioned while 10 ago that the switch yard and circuit break explodes. That 11 distracts the operator.
12 MR. KING: Yes.
13 MR. WYLIE: So that would be included.
14 MR. KING: That would be included.
[}
15 MR. MICHELSON: In other words I think there's j 16 very few things that would be excluded and you would have to 17 make a case for each exclusion is what it amounts to.
18 MR. KING: I think that's a good way to look at i 19 it.
20 DR. KERR3 The Commission apparently is using a 21 maintenance rule to revise the Commission's responsibility i
~
22 for the regulation. I'm not sure that -- in fact I think a 23 revision is probably in order because I think certain PRA's 24 are demonstrated, the so called non-safety systems had a 25 significant impact on safety. But it puzzles me that the !
O serie ee aerereie9 coreov eie-(202) 628-4888
S i 12 1 Commiasion,would take such a momentous step in terms of 2 maintenance; rule. It seems to me.it ought.to look at the 3 total problem of changing regulatory responsibility on a 4 broader basis, just in terms of maintenance. That' puzzles 5 'me.
6 MR. DEY: Well I think the intent of the
.7 Commission that the rule would extend beyond what is covered.
8 in other regulations into the DOP, since a lot of the --
9 DR. KERR: But you see, if indeed the balance of 10 plant things are'important to safety'then it's not just 11 maintenance that ought to be involved, it's selection, it's 12 ' installation it's all sorts of things that bear on safety' 13 'and not just maintenance. And it seems to me it's rather
() 14 peculiar to approach what I think is really a significant 15 change in Commission policy by sort of backing into it, and
~
16 not backing into it all the way but just on the' basis of the 17 maintenance of these systems.
18 MR. KING: I don't disagree it's a policy 19 question, I'll just leave it at that.
20 MR. DE'J : Well, we'll come back to that when we 21 discuss the regulatory guide. As I was mentioning, one 22 major change that we have made is in the implementation 23 section. Initially there was one option for the licensees 24 to provide certification to the NRC directly. In addition f 25 an option has been added where the certification would be O =eritae- eerei e (202) 628-4888 correr eie-
13 c,
, 1 provided by a third party of the licensee's plant 2 maintenance programs. And t'his would be reviewed, the !
3 certification would be reviewed against a maintenance 4 standard. And this has been included in the rule in the 5 codified text.
6 MR. CARROLL: What does that translate to, who do 7 you really have in mind as this third party?
8 MR. KING: Well, this is an important point in ,
9 that we still think the right way to go about regulating 10 maintenance is for the industry to take a lead to write a 11 standard and to have some third party certification process 12 that would assess the plants against that standard. And we j 13 still want to keep that option alive and that's why we've
() 14 put it in the rule. Now granted we don't have any 15 volunteers at this point to step forwar' 'o write a standard 16 or do third party certification but we're still hoping in 17 the future that that path will at some time in the future be 18 taken. And that's why we put that in the rule.
19 MR. CARROLL: And your model, Tom, is ASME and end 20 stamps or code stamps or --
21 MR. DEY: It would be similar to that process.
22 MR. KING: Well, I'm not saying it has to be --
23 we're open to a proposal as to what it would exactly -- who i
24 would do it and how they would do it. We don't have any 4 25 preconceived ideas of how it has to be. But the concept of
(,)
/
Heritage Reporting Corporation (202) 628-4888 l
d 14 1 having an industry standard and a third party certification 2 is what we want to' keep open and keep that option alive.
4 '
- 3. MR..MICHELSON: I sort of envisioned this more 4 like the TUV in Germany. .Is that the kind of model that we p . .
5 might have in mind? Or are you acquainted with it?
6 MR. KING: I'm not acquainted with the TUV. I 7 MR. DEY: I'm somewhat familiar with that. I )
8 think'it would be -- :
I 9 MR. MICHELSON: Is that the sort of thing?
10 MR. DEY: Similar to that.-
11 MR. MICHELSON: Don't they function sort of like a 12 third party inspectorate and so forth?
13 MR. DEY: Similar, not exactly. But I'm not sure 14 of what.
1; . 15 MR. MICHELSON: Yes, I mean it doesn't. function 16 that way quito as much. You're right, codes and standards j 17 but they don't run inspection teams and such.
18 MR. CARROLL: Oh yes.
19 MR. MICHELSON: Well, to get.the stamp, yes. I'm 20 thinking of in the maintenance area. ,
4 j
21 'MR. CARROLL: No, I was saying it was analogous.
22 MR..MICHELSON: It could be extended, then, you're 23 saying.
'24 MR. DEY: And the reason for this recommendation 25 is based on a public work shop and the comments received
-O rie e- eerei e corger tie-(202) 628-4888
(
n 15 1 from the industry on the initiatives that this would f) 2 continue, give an opportunity to continue the initiatives 3 and provide leadership for developing an industry wide 4 maintenance standard and the establishment of the 5 certification program. And the staff is recommending this 6 as preferred option for implementation.
7 MR. MICHELSON: Now where is the staff 8 recommending that?
9 MR. DEY: In the Commission paper there is 10 langauge on this and also in the notice of final rule 11 making.
12 MR. MICHELSON: Could you tell me where so I -- I 13 must have missed it.
14 MR. CARROLL: Rule beginning on page 55, actually
(}
15 on page 56, ii.
16 MR. MICHELSON: Yes, but I didn't -- that's just a 17 provision that this is a viable option. Where is the staff 18 actually coming out and endorsing?
19 MR. KING: In the draft SECI paper on sort of the 20 last half of page 2, we talk about it.
21 MR. MICHELSON: Page 2, okay. I probably read so 22 long that I forgot what was up front.
1 23 MR. CARROLL: At one time, maybe I'm wrong, but I 24 had the a6nse that you thought this third party could be 25 INPO. Given the arguments that they've mado as to why O erie ee Regerti e Cereeraeio-(202) 628-4888
l 16 l 1
I) 1 that's inappropriate, given INPO's charter, do you agree j
,/
2 INPO isn't the third party you have in mind?
3 MR. KING: No, I don't think we've discounted 4 INPO. I think the comments from the industry have 5 discounted that INPO is the right organization to do that, 6 but I don't think we've discounted that.
7 MR. CARROLL: You do not agree it would compromise 8 their whole mission to become a quasi-regulatory body?
9 MR. KING: I personally don't think it would 10 compromise their whole mission. That's my personal view.
11 MR. DEY: It depends on how you look at it, there 12 are already a lot of regulations existing and there are 13 initiatives related to that. There are two ways of looking
(')
\s 14 at it.
15 MR. MICHELSON: Tom, could you point out again 16 where that position is stated and what document now? I've 17 got so many Is it in the draft regulatory guide?
18 MR. KING: It is not in the regulatory guide. The 19 rule has the option in it.
20 MR. MICHELSON: The rule has the option in it, but 21 where is the discussion of your report?
t 22 MR. KING: In the draft SECI paper the thing 23 that's right up front of the package we have.
)'
24 MR. M1CHELSON: What page, is it in thers?
25 MR. KING: Turn to page 2.
i
/^\
i) Heritage Reporting Corporation (202) 628-4888 i
17 1 MR. MICHELSON: Okay. Background. I
[v) 2 MR. KING: We're under discussion section now.
3 You're not under SECI -- you're not in the right thing.
4 MR. MICHELSON: Okay, you're in the paper that was 5 even up front of that.
6 MR. KING: Right.
7 MR. MICHELSON: Okay, that's why I didn't find it.
8 And then page 2.
9 MR. KING: Turn to page 2, fourth paragraph down 10 where it starts compliance with the rule could be certified.
11 And that paragraph and the following paragraph.
12 MR. DEY: Okay, the last two bullets here, the 2.3 staff believes the rule is intended to provide regulatory 14 emphasis, and we have developed a reg guide in the absence
(~>)
- x. ,
15 of an industry commitment by the close of the public comment 16 period. And that concludes my presentation for the rule 17 making package. And we are ready to move into a discuscion 18 of the regulatory guide.
19 DR. KERR: At what point is it appropriate for you 20 to maybe comment on my question as to whether the Commission l I
21 feels that there is significant risk associated with the l l
22 maintenance, now as contrasted with there might be j l
23 significant risk associated later on? i 24 MR. DEY: We can cover that topic, Dr. Kerr, in I
l 25 session VII. I
/
l (_,) Heritage Reporting Corporation (202) 628-4888 l
w - _-_
t 18
.1 DR. KERR: Okay, later. 'And at the'.same point'I 2 woul'd welcome a comment on whether new. reg 1150 analysis of
~3 plants.show any risk associated with maintenance.
4 MR.:DEY: I think we have some data'on that.
5 MR. MICHELSON: As a small comment on the rule 6 itself, item A on applicability, why did you in this 7 particular. case say it applies to all nuclear power reactors 8 when normally you talk about all production or utilization-9 facilities.. Is it indeed a narrowing somehow of where it 10 'doesn't include all production and utilization facilities?
11 MR. DEY: Yes, that's the intent.
12 MR. MICHELSON: This is the first time, gee I 13 thought most of the things I ever read always talk about 14 production and utilization facilities and now we say applies
'( )
15 to all nuclear power reactors.
16 MR. DEY: There are some regulations that use the 17 same statement. I forget which they are.
18 MR. MICHELSON: So'you're certain that those are 19 the correct words in this case?
20 HR. DEY: Yes, I believe so.
21 MR. MICHELSON: I'll take your word for it now but 22 -when I read those other sectiors before I was led to believe 2? you're really talking about production and utilization 24 facilities. I don't know.
25 MR. DEY: Let me start with just giving a format, O Rerieae. ereime cozeeraeien (202) 628-4888
{
19 the outline of the reg guide as it is. It follows the usual
()
m 1
2 pattern of format of reg guides, introduction, discussion 3 and a section on the regulatory position. And we have six 4 subsections within the section on regulatory position. And 5 essentially it follows a logical flow of the process which 6 I'll just come to shortly. And it concludes with a section 7 on implementation which provides guidance on compliance with 8 regulations.
9 MR. MICHELSON: In terms of --
10 MR. DEY: Now I am in the handout which is 11 numbered 4 and 6, draft maintenance regulatory guide. $
12 DR. KERR: The discussion of the regulation on 13 page 38 has the statement that the maintenance ruled defines
'S 14 the minimum level of maintenance necessary to assure (V
15 adequate protection. Where do I find that? Is that in the 16 rule or the reg guide?
17 MR. KING: This is one of those sections that was 18 written when we were heading down t..e adequate protection 19 route that's now going to change. I don't know if you 20 recall in the beginning of this session this morning I said 21 there are going -- I 22 DR. KERR: Will it define the minimum level of 23 maintenance necessary for anything? l 24 MR. KING: It will define what we feel is an 25 acceptable level of maintenance based upon an enhanced l
() Heritage Reporting Corporation (202) 628-4888 l
i
t 20 l 1 safety justification.
2 DR. KERR: That will be in the rule or the reg 3 guide? j 4 MR. KING: The reg guide. That's what we'11 talk 5' about later as we get through this. l 3
6 DR. KERR: Okay. ]
J 7 MR. DEY: Let me start off with providing you with 8 our approach to developing this reg guide, and the way we 9 see it being implemented. And we already heard a lot of 10 questions related to the rule and its purpose.
11 We feel the reg guide provides the basic guidance 12 and criteria for an acceptable program. And it's adopted a 13 process approach which allows flexibility for the specific 14 activities. Essentially what we are putting in the reg 15 guide is guidance on a process, we believe, which will lead 16 to effective maintenance. And allowing flexibility to the 17 plants and licensees or the conduct of the specific 18 activities and not being prescriptive in that sense.
19 We believe it should have a minimal impact on 20 already existing good maintenance programs. Because those i 21 programs would be executing such a process.
22 DR. KdRE: How will you decide which ones are good 23 and which ones are bad?
.24 1m. DEY: I have a slide on crite,ria for 25 acceptability and if you want to wait we can discuss it Heritage Reporting Corporation (202) 628-4888 l
\
21
'. '1 then.
2 DR. KERR: Okay, I'll wait and see what it says.
3 MR. DEY: And this approach allows use of industry 4 guides and programs for details. And I've tried to put this 5 process concept.that we've adopted in the reg guide into a 6 schematic. -And this outlines the process for an effective 7 maintenance program and starting from the top we feel the 8 overall maintenance policy and its relationship to safety 9 has to be established. What do you want maintenance program 10 to accomplish in relationship to safety?
11 Starting with that once you develop objectives for 12 the maintenance program, general objectives.and the more 13 specific objectives depending on the functions and safety (j 14 significance of the equipment. And then derive the 15 maintenance activities for the equipment. So it's taking a 16 big picture look of what you want the maintenance program to 17 do, setting objectives and goals and deriving your 18 maintenance activities based on those objectives.
19 And then after conducting maintenance on the 20 equipment you would monitor the equipment to determine if 21 you have an effective maintenance process. And use the 22 results of that monitoring and assessment and feedback r.nd 23 determine corrective actions depending on if your 24 maintenance program is accomplishing what you want it to do.
25 So this is essentially what the reg guide is O rie ee gereine ce geraesen (202) 628-4888
9 22 l1- requiring that thi's process, staff feels, is essential for
( h-2 an effective maintenbace program.'
3 MR. CARROLL: Little box ther6, Moni, required 4 structure systemsLand components maintenance ~ activities.
5 The intent is that that's' graded, that required activities 6 are graded. Something that's very highly safety related.
7 You'may.have one set of objectives and approaches to doing 8 the' maintenance activities. The coffee pot in the control 9 room you may have.
10 MR.,DEY: -Yes,~ absolutely. And the reg guide 11 allows.and hasLlanguage in it that allows that determination.
12 depending on its relationship commensurate.with safety y 13 significance and impact.' And the criteria for an acceptable j maintenance program is that this process is adopted at the
-().14 15 p. ants. As we have said before in the proposed rulefand in l 16 th,e rule making ' package, the compliance will be verified by-17 NkC audit and inspection.- Moaning that NRC would not review 18 td te maintenance programs after they have developed,- but- q 19 wsuld verify compliance by audit and inspection.
20 MR. CARROLL: Now when you say that are you 21 talking about, at least at the moment, the module that you 22 have for special maintenance inspections, or whatever it's 23 called? i 24 MR. KING: Maybe, why don't we even back up 25 further than that. Moni said we wouldn't review the O erita gertine cergeraelem (202) 628-4888
23
[) 1 program. Wouldn't that require the program to be submitted v
2 to us for approval when the inspector went to the plant to 3 do his audit and inspection he would certainly look at the j b
4 program, the documented program that the utility has on j 5 site. Whether he would use the current NR inspection guide 6 for maintenance, we would expect that when this rule and reg 7 guide are issues that that inspection guide would evolve so 8 that the two are consistent with each other and that that 9 would be a guide that you would use in doing an inspection 10 on it.
11 MR. DEY: I'm sorry --
12 DR. KERR: Just a minute, I wanted to -- what does 13 the last bullet mean?
('] 14 MR. DEY: Okay, I haven't gotten to it. I was on V
15 the second bullet.
16 DR. KERR: Oh okay, you were still on that bullet?
17 MR. DEY: Yes. What we're saying is that the 18 criteria for acceptability is that these steps are used in 19 the plant maintenance program. That one, the program is 20 established, documented and implemented based on SSC 21 performance objectives. These objectives can be graded 22 depending on safety significance and impact.
23 DR. KERR: Is that requirements expected to be 24 described quantitatively?
25 MR. DEY: One means of determining if you are d Heritage Reporting Corporation (202) 628-4888 l
l l
24-
'1 meeting' objectives you could -- the reg guide does encourage
- 2. the use of quantitative parameters, to judge whether one is l 3 achieving the objectives. But the objectives don't need to 4- be quantitative,-'it would be deterministic for safety 5 related systems which function' solely for safety, for 6 systems in the. BOP that are initiators, whose; failure would.
7 initiate a transient or challenge a. safety system, differ 8 relationship for safety.. You would want to minimize those 9 challenges.
10 DR. KERR: I don't~know what minimize means other 11 than zero. And I don't think you intend to make them zero.
12 Is.it your anticipation, I mean would you consider it- ,
.I 13 unsatisfactory-if -- or do you think it should be considered !
14 unsatisfactory if a licensee does not establish numerical 15 goals for reliability?
16 MR. KING: I-think we need to say it a little 17 stronger, Moni. The reg guide does require that the 18 licensee establish a quantitative numerical goals for his ,
19 system structures and components. In addition to that there !
20 would be some qualitative assessments that would be done.
21 DR. KERR: But you don't require that each 22 licensee use the same one?
23 MR. KING: No, there's flexibility there that he 24 can choose.what makes sense for his plan.
25 MR. CARROLL: So when I turn the spigot on the O erit ee ee ;;i e correlation (202, 628-4888
l- !
25 f
\
()
,~
l 1 coffee pot, nine times out of 10 coffee ought to come out of 1
2 it. j 3 MR. KING: He can. set that as a goal. Okay now i 4 there are some general guidance on the level of performance 5 that those goals ought to be shooting for but the reg guide 6 doesn't give numerical values for that.
7 DR. KERR: You're still not to the last bullet, or 8 are you?
9 MR. DEY: I was on the sub-bullet here.
10 DR. KERR: Okay.
11 MR. DEY: So one would monitor if the program is 12 meeting these objectives. But the merit of this approach is 13 a process that the licensee will establish some objectives, (a~)14 plan and establish objectives that would guide his program, 15 And he would look at if his program was accomplishing it and 16 be required to take some corrective action based on that.
17 And I think this is what would put the plants in a position f 18 to look at the program and seeing if that's what it takes to 19 meet the objectives for safety.
20 Now the last ballet here is, as Tom mentioned, we 21 have recommended in the reg guide that quantitative goals be 22 set so that one can monitor if one is achieving these 23 abjectives that the licensee would set. The criteria for 1 24 acceptability for that program, the licensee program is not 23 whether -- if he set a goal, a target or a quantitative goal Heritage Reporting Corporation (202) 628-4888
r.
)
1 26 I~
\_/
) 1 it's not whether he's met that threshold or not.
That's 2 what 4 hat last bullet is intended to say. )
I 3 MR. KING: I see you're still confused. ]
i 4 MR. CARROLL: What is it then?
5 MR. KING: He establishes some goals. His program 6 is directed and being revised to meet those goals. At any l 1
7 given time if you come in, an inspector came in and looked !
8 at the plant against this reg guide, the fact that he hasn't j
9 achieved those goals yet la not grounds for saying he's not i 10 in compliance with the rule and the reg guide. I think what i
11 we were looking for is that he has this process in place, 12 that his program is improving with the intent of meeting 13 those goals. The fact that he isn't there yet is not -- we
(~i 14 don't view as grounds of saying he's not in compliance.
%J 15 DR. KERR: Oh, it's the process that's important 1 16 and not the results?
17 MR. KING: We would expect for him to eventually 18 get there and maintain some level of maintenance consistent 19 with those goals that he establishes.
20 DR. KERR: Presumably the object 3ve is to decrease 21 risk. How does one tie, how is one expected to tie these 22 performance objectives to plant risk?
23 MR. KING: The performance objectives in the reg 24 guide are based upon component failure history, trying to 25 get lower component failure rates. We believe that that's 3
/~~)N
(_ Heritage Reporting Corporation ,
(202) 628-4888 -
l 1
I l
i
-- i
27 1
.! 1 tied -- you can tie that to reduce risk. And that's part of
(
. I 2 what we're trying to quantify in regulatory analysis. !
A 3 DR. KERR: I would hope that one could tie that to g
4 reduced risk also but I don't know how to do it, do you?
5 And do you expect the licensee will know how to do'it?
6 Because the objective is not just decreasing component i
7 failures, it's decreasing risk. I think. And that means, I 4
{
8 would hope, that there should be at this point at least an j 9 inkling of some way of knowing what to pick out that's i
10 important. You know, I heard if the light bulb in the men's 1
11 room burns out every week, as far as safety is concerned, 12 but there are some components that may be heavily related to 13 safety. Do you anticipate that there will be some ,
1 14 systematic way developed to relate these performances to 15 risk.
16 MR. PZY: Yes, in the regulatory guide we're l 17 asking the licensees to --
l 18 DR. KERR: The licensees -- most of them do not 19 have the capability of doing this, Moni. Certainly not in 20 two years.
21 MR. DEY: Well, on the other hand if we did it, it 22 may become too prescriptive. We want to give the licensee
-23 the flexibility.
24 DR. KERR: Well, this in one of the reasons I .1 3
i 25 asked about 1150. Since this -- I mean 1150 might have some Heritage Reporting Corporation (202) f28-4888 {
1
_ _ _ _ - - - _ - _ - _ - - - - - - - - - - - - _ _ _ - - - . - - . . _ _. - }
L i I l l
1 28 l l
[D x/
1 indicators. Has anybody looked to see whether one could l
2 find embedded within all that work some guidance that would 3 say these are the really important components you ought to ,
4 look at and here's the levels or at least some indicators of 5 levels if one's going to decrease risk?
6 MR. KING: Well, we're in the regulatory impact 7 analysis trying to estimate the reduction and risk. And 8 looking at 1150 and looking at a lot of things. Today we ;
9 don't have any numbers to give you. Next week the intent is 10 to have that package 'n your handa.
11 But what w say in the reg guide is we believe the <
12 licensee has to have the responsibility and the flexibility l
13 to look at his own plant and decide what are his goals and -
(~) 14 what we say in the reg guide is that he ought to be looking ..
ss 15 at those things that have caused failures in the past, 16 failures that have caused challenges to plants, actuation 17 and safety systems and so forth, and be setting goals to [
18 improve the failure rates of those systems up to what is 19 achieved by the best performers in the country.
20 DR. KERR: I would assume that if he uets a 21 ridiculous goal you will probably object and I don't think 22 most licensees would try to set a ridiculously --
23 MR. KING: Set a ridiculously low goal we would 24 object, yes.
25 DR. KERR: No, I mean even a ridiculously high (m,) Heritage Reporting Corporation (202) 628-4888
{
y .
is $ i
- . .i2 -
29 1.' goal for' failures that are acceptable. 'You wouldn't find 2 thr.t?. :That msans that somebody on the staff must also have <j
.3 some way of' deciding what is ridiculously high. Do you have 4 in place some sort of way.of developing these numbers? When 5 'you begin to start inspecting, or do you have in mind that.
6 you will need to develop those numbers?
7 MR. KING: I think the inspection guide'is going' 8 to have to give some cuidance on how those numbers will be 9 developed and so forth. I think,what wo h&ve in mind now 10 is, we've got data from the 110 plants as to what's causing 1
11 -- what components-are failing, how often they're failing '
12 and what --
1 13- DR. KERR: But that tells you almost nothing about 14 their contribution to risk. Fortunately or unfortunately it 15 doesn't. And what you want to concentrate on I think if 16 you're serious about this is thesse components and systems 17 that might have a significant contribution to risk.
16 Otherwise yo.u' re not going to accomplish anything.
19 IG. KING: And the reg guide has some general ,
20 words in there that ask the licensees to go look at it on 21 that basie, what's is1portant to risk.
22 MR. DEY: If you Icok at this slide, Dr. Kerr, 23 the process we are asking the licensee to go through is j i
2.4 general objective of maintenance is to prevent the failure f 25 of degradation. We wanted to nake a deterministic i
O serita9e neeertins coreoraelen (202) 628-4888
= . - _ _ __ - _ __-_ _ -_-______ - _-__________ __ _ _ _ _ b
l l
30 i x I
,)
1 evaluation of function and safety significance to risk of 2 these components. 1 3 DR. KERR: The nearest thing that a licensee, 4 unless he's rather unusual and may be part of 1150 will have i 5 to this is his IPE or PRA. And the PRA' a may very well not (
1 6 permit one to do this in their present form. I would be j i
7 surprised if they do particularly in terms of telling them -
l 5
- see it strikes me that if you go through this that it !
9 might well be possible not to look at everything in the 10 balance of plan. You can just throw some things out, maybe,
}
11 not because they have zero likelihood of contributing to !
12 risk but Lecause it's very low.
I 13 And maybe the way to do that is to have each l
]- 14 licensee look, perhaps the most effective way, I'm not sure 15 ft's the most efficient way. But at some point there has to 16 be, it seems to me, thia interface between risk and 17 component failure and I don't think it. exists in a very well 18 defined form at this point.
19 MR. KING: You don't need a PM to know what's 20 been causing unplanned scrama, what's been causing safety 21 system unavailability or unplanned actuation.
l 22 DR. KERR: But you need a PRA to know what 1
f 23 contribution to risk has occurred as a basis of thia. For 24 example, you can have one channel unavailable for three l 25 whatever one calls it, system. That may not contribute a
- l l
( Heritage Reporting Corporation ]
I (202) 628-4888 '
31 >
l
() 1 great deal,to risk or it may. I mean the mere fact that 2 there has been unavalAability to failures does not per say 3 indicate that this has been a high risk. May be, may not.
4 But you've got to look at the total system, I think, to I 5 undarstand this. And you certainly'want to concentrate your i
6 sources on those things that are likely to decrease risk.
7 MR. KING: Let me try -- I don't disagree with you 8 and I think the philosophy of the reg guide is trying to do 9 that. And maybe the words don't come across that way but I i
10 think that's the intent. But nhe burden would be on the i
11 licensee to go make those determinations. Staff isn't going 12 to point out three or four or 15 different components. ]
j 13 DR. KERR: If the staff is going to inspect it 14 needs to have some indication of whether the job being done
(' ) i 15 by a licensee make sense, j 1
16 MR. KING: I agree with you. j 17 MR. CARROLL: I think the other implication of 18 what you're talking about is if you're really going to do I 19 yhat you say it's going to make a hell of a big impact on i
20 people that run the maintenance program f.n the plant to put )
i 21 this kind of analysis together. They're going to have to 22 work with other disciplines. They are a key part of -- you .
1 23 just can't go off and tell what PRA lire guys -- tell me how i
24 I should run my maintenance program, they'll never get to 25 that.
I) Heritage Reporting Corporation (202) 628-4888
,=
- \ '.
A' 32
)- 1 And I think when you do your value impact 2 statement it's going to be left, it sounds to me, you better 3 put in a lot of man hours of maintenance --
4 MR. DEY: Well in some sense programs that are out 5 there have done this already and they have derived their 6 maintenance activities, they have made a judgement as to how 7 much maintenance is required for the component.
8 MR. CARROLL: Yes, but it's one thing to make a 9 judgement, Moni, it's another thing to have what are the 10 words that inspectors like to use, objective audible 11 evidence that you've done this. That implies a lot of paper '
12 cnd analysis.
13 MR. DEY: Yes, our intent is not to create paper 14 We want the licensees to go through this exercise and
() work.
15 to some extent have 3t documented. But we're not looking 16 for volumes of auditable material and we make_that point in 17 the regulatory guide.
18 MR. MICHELSON: Where do you make that point?
19 MR. KING: Ii ll find that page here.
20 MR. MICHELSON: That's a foot note on page 43, 21 MR. KING: No, that's not a foot note. Page 3 of 22 the reg guide.
23 MR. MICHELSON: Page 37 24 MR. KING: Yes, first full paragraph.
25 MR. MICHELSON: That's oh okay, of the reg guide.
() Heritage Reporting (202) 628-4088 Corporation l
l'
[~ l I
l l l 33 l
() 1 I don't find it overly comforting. It does indeed say that ,
2 you have to figure out an appropriate program but how about 3 the availability of evidence that can be audited and all 4 that sort of thing?
5 MR. DEY: There are some more words in section 4 6 in the preamble to section 4. Which is the section on 7 maintenance activities.
O MR. MICHELSON: Yes.
9 MR. DEY: Because that is certainly an issue we've 10 been deliberating on extensively.
11 MR. MICHELSON: There are a couple of other places ]
12 I marked where you get back to the graded approach idea. So 13 somehow I guess what you expect the utility to do is to 14 figure out what's important to safety and security, identify
(}
15 that in a documented way and then go through and maintain 16 more detailed records on those particular items, and it 17 might maintain that on a lot of other items. Is that the 18 idea?
19 MR. DEY: That is the idea.
20 MR. MICHELSON: And the auditor or inspector would 21 go through and look at his basis which would be his 22 selection process and then look selectively at the items he 23 has -- and look to see if enough items have been covered.
24 That would be the way you would approach it, I guesa.
25 (h
(,) Heritage Reporting Corporation (202) 628-4888
f l
34
! 1 Before you leave that slide, I'm a little puzzled 2 over -- we talk about structures, systems, and. components 3 performance objectives. Now, down at the component level, 4 we're talking about, literally, tens of' thousands of items, !
5 potentially, of which only a small fraction, I guess, are 6 going to be graded out as being important enough to safety 7 to keep detailed records of. Is that, again, the idea, l
8 then?
9 MR. KING: It would certainly be a graded !
10 approach. A licensee would figure out which of those he 11 wants to keep records on, which of those h~ rnts to 12 establish goals on. ,
13 MR. MICHELSON: Yeah, And, having selected the 14 important ones under this process, and you judge his process 15 of selection and audit, maybe, a few of the selected items -
16 -
For. Instance, lets take valves. A given valve, a given 17 gate valve in several different systems might be very 18 important to safety in one system and not quite so important 19 in another system, but you're not trying to use gate valves u 20 as a particular line item, are you? Rather, I thought the 21 whole idea was the availability of important systems of 22 which gate valves are one of the components in a system. ;
23 But your goals are based on system availability,, aren't ;
24 they?
25 MR. DEY: That's right, Mr. Michelson. The idea )
)
Heritage Reporting Corporation (202) 628-4888 l
l t
r i
1 35
() 1 'i s top-down, if I may use that word, to state.
2 MR. MICHELSON: Ok'ay. So you're grading --
3 MR. DEY: On plant safety.
4 MR. MICHELSON: Most of your goals and objectives i 1
5 will be high-level goals dealing, at least, with system-6 availability, although you might-have a lot of valve
-1 7 problems, in which case you might set up some kind of a !
8 program looking to see if you valve maintenance program is 9 doing the right thing.
10 It's just not clear how you do it. It's got to be 11 a case-by-case basis, almost.
12 MR. KING: If gate valves are a particular plant, 13' there may be a specific goal for gate valves. That's not !
precluded.
(])' 14 15 MR. MICHELSON: But, normally, I wouldn't expect 16 that level of detail. I' d expect mostly system goals, 17 wouldn't I? Or would I?
18 MR. KING You would expect some top-level goals, 19 some system goals, and some component goals, t
20 MR. MICHELSON: Okay.
21 MR. CARROLL: So, more likely, you'd say "the 22 feedwater system." That's got to be reliable. I'll set a 23 goal that I shonidn't have more than one reactor trip a year 24 related to.the feedwater. !
I 25 MR. DEY: That's right.
I e
() Heritage Reporting Corporation (202) 628-4888 !
l l'
. - - - _ _ - - - _ _ . _ _ _ - b
1 13 6 l
() .1 MR, CARROLL: Now, how do I take it the next level 2 down. I would look at the components that contribute to 3 this?.
4 MR. KING: If your feedwater pump is the thing 5 that's been causing you the trouble in your system, you 6 might set a specific goal on it.
7 MR. CARROLL: But a pump is made up of components, 8 also, so how far down do I carry this thing?
9 MR. KING: It's up to you.
10 DR. KERR: Would it be acceptable to simply 11 establish system goals, like the feedwater system, and not 12 worry about componenta and setting goals? ,
i 13 MR. KING: No. We have in the reg guide, some 14 component-level goals are important to monitor the
() ;
15 effectiveness of maintenance. 0 16 DR. KERR: But why? If the system performs i
17 correctly, the maintenance must be okay. 1 18 MR. CCING: Do you want to expand on that, Mark?
19 MR. MICHELSON: It's a very important point, and 20 it's a puzzling one to me how you're really structuring this 21 thing.
22 MR. WILLIAMS: Well, there's a couple of things.
23 First of til, I'm only speaking in the area of indicators 24 and monitoring of indicators.
25 MR. MICHELSON: Mark, excuse me. Would you get
-( ) Heritage Reporting Corporation (202) 628-4888
4 L .y 37' a
it your.name into.th4 record, please.
2 MR. WILLIAMS: Mark Williams. AEOD.
3 In the area'of indicators, I.think your point 4 about the association between the risk significance and the 5 maintenance'is something that we struggled with quite a bit.
-6 Next week,we'll be discussing the progress on safety system 7 . unavailability indicators. .
8 Initially, we tried to use that kind of a thing, 9 which is a risk-based indicator, for assessing maintenance 10 effectiveness.- It didn't work out. We couldn't get .
11 correlations. We'couldn't show an association with the 12 effectiveness at.the' system level ~for auxiliary feedwater 13 systems, for example. The unavailability is controlled by 14 other things.
()
15 , In order to assess the effectiveness of 16 maintenance or the quality of maintenance, we had to take .a 17 different approach.
18 MR. . CARROLL: Back up on what you just said.
19 You studied auxiliary feedwater systems, looked at i
20 their unavailability. It was controlled by things other )
21 than maintenance?-
22 MR. WILLIAMS: Well, LCO entries, the 23 unavailability of the systems was controlled by, in many 24 cases, exposure times, exposure times that would result from 25 discovery after a 30-day test interval of some particular I i
i
() Heritage Reporting Corporation (202) 628-4888
l 38 i
l( ) l' problem for.which they entered an LCO on the system. .THe 2 association of that with maintenance effectiveness was not L 3 clear, I guess would be the best way I would put it.
- 4. ~ Carl, would you like to expand.
5 MR. MICHELSON: Well, it's fairly possible that i
6 th ere's isn't much relationship .between the two, you know.
7 What do you mean by maintenance effectiveness?
l 8 MR.' WILLIAMS: When we thought about maintenance 9 effectiveness, what we really thought of was the quality:of ;
10 maintenance. The ability of the maintenance to maintain-the 11 performance of the component. In order to get an j 12- understanding of that, what we had to do was go to the l 13 component level and look at the change in failure rates at 14 the component level. By doing that, we could come up'with
()
15- deterministic and empirical correlations that convinced us 16 that we had an indicator that measured the effectiveness of 17 ~ maintenance.
18 MR. MICHELSON: How did you separate out just poor 19 equipment? It's originally designed efficiency?
20 MR. WILLIAMS: Well, one of the things -- we'll 1
21 discuss this in detail on Thursday, I think. 1 I
22 MR. MICHELSON: Oh, okay. I 23 MR. WILLIAMS: One of the things that we do is -- )
L
!. 24 MR. MICHELSON: What's the discuasion on Thursday?
25 That's with us or some other group?
I
. l
() Heritage Reporting Corporation (202) 628-4808 ]
1 j
l' a I'i L. 39 '
1 .MR. ~ WILLIAMS: That's with you. Thursday is the 2 discussion of performance in'dicator development.
p .
3 FEL. MICHELSON: I think it's a fairly ~short i
4 discussion, though.
5 MR. WILLIAMS: It's about an hour. That's all we 6 had.
i 7 MR. MICHELSON: On performance indicators alone?
8 MR. WILLIAMS: That's it.
9 MR. MICHELSON: I thought there were two subjects, 10- of which that was one. Is that all? One hour on 11 performance indicators?
12 MR. WILLIAMS: I think it's one hour on
, 13 performance indicators.
14 MR. MICHELSON: Okay. That ought to be adequate.
() !
15 MR. WILLIAMS: We can go into it in detail then, i
16 I think, in response to the question, one thing that we do 17 is we monitor the change in the failure rate, eliminating i
18 say, the inherent failure rate of the equipment. !
19 MR. CARROLL: We were talking components. '
20 MR. WILLIAMS: Right.
21 MR. CARROLL: We were talking auxiliary feedwater 22 systems.
L 23 If the availability of the auxiliary feedwater )
i 24 systems is. consistently good, and it's up here, but I've got 25 some component that's failing all the time, and I've looked
() Heritage Reporting Corporation (202) 628-4888 !
1 i
- - - - - _ _ _ _ - _ _ _ _ _ _ _ _ - _ _ _ _ \
40 (m) 1 at, and I say, "Well, I'll live with that. I'll replace 2 this thing once a day or once a year." Are you saying that 3 that's a bad way to do maintenance?
4 Because it's what happens in the real world.
5 MR. WILLIAMS: Well, no. I'm not trying to take 6 issue with that point.
7 I think what we were trying to do is, if there was 8 a particularly troublesome component that had a certain 9 failure rato, I think the model that we're using would tend 10 to ignore that for maintenance effectiveness monitoring.
Il In the area of safety system unavailability 12 monitoring, and Carl Johnson from Research has been doing j l
13 the work right here, I think, if that component impacts the
( }; 14 system level availability or the train level availability, 15 then that monitor would come into play, but it wouldn't, 16 from what I understood you to say.
17 So it wouldn't be picked up by either.
18 MR. MICHELSON: You don't seem to be advocating 19 component level objectives, then, but rather, system level 20 objectives, or at least that's where you're finding your 21 proper ability to correlate.
22 MR. WILLIAMS: Well, for maintenance 23 effectiveness, I think we are saying we should monitor 24 component failure rates. j 25 MR. MICHELSON: Well, no. I'm talking about
(,) Heritage Reporting Corporation 3 l
(202) 628-4888 l
l
)
41
_1 component objectives, as opposed to system objectives. Are 2 you advocating prescribing component objectives, which is 3' what. the slide seems to indicate?
4 MR. WILLIAMS:- Well, it's a top-down approach, 5 again.
6 The way the program is laid out is that it's a 7 7:!aded apprcach with overall objectives, which may be plant 'l 8 availability and in safety systems it may be overall 9 failures of tests or whatever it is. But it's a graded 10 approach, and I think~it's just.as Tom said: when a plant 11 reviews their experience for troublesome components or for 12 certain selected components, they may set component 13 performance objectives. It would be difforent for different
] 14 systems and different equipment.
But it would seem to me that the system 15 DR. KERR:
16 -performance would be much more important and much more of an ,
17 indicator than component performance. I'm lost here.
18 Is risk what we're trying to reduce?
19 MR. JOHNSON: Yes, certainly.
20 MR. MICHELSON: Would you identify yourself for 21 the record.
22 MR. JOHNSON: I'm Carl Johnson from the Research 23 office.
l 24 The performance indicator program that NRC adopted 25 is based as close as we could on a risk point of view, of i
V l
Heritage Reporting Corporation L
(202) 628-4886 }
l 1
l \
l 42 m
() 1 monitoring, unavailability of systems and initiating events i 2 and this sort of thing. But it had to be a qualitative sort 3 of logic because we weren't able to provide good models for 4 interpreting data in that fashion. So it's evolving.
5 One of the things that it's evolving along is a i
6 research effort to try to come up with a way of doing just l I
7 what you're saying, of how could you monitor the 8 unavailability of important safety systems. f 9 We worked up several ways of doing this in theory, 10 we presented it to the Commissioners a year ago, and they 11 said, that sounds nice in theory; try it out on some real 12 plant data, historical data. We found some of these things 13 Mark mentioned, that in some ways the indicator behaves like I'T 14 we thought it would. In one case, we tried it out on a
%)
15 plant that had improved its maintenance program, and the 16 indicator didn't indicate improved availability of aux I
17 feedwater, for example.
18 What I'm trying to say is, we're not done. We ,
19 need to do some more work in this area.
20 DR. KERR: How did you conclude that the 21 maintenance program had been improved if it didn't improve 22 the availability of the systems.
23 MR. JOHNSON: Well, this is a perception.
24 MR. WILLIAMS: What we were trying to do in these 25 programs was to obtain a correlation between some -- In 4
(._) Heritage Reporting Corporation (202) 628-4888
r 43 y
() l' Carl's program, for example, we were trying to obtain a 2 correlation between an improved performance, as determined j 1
3 by other NRC indicators, being SALP or inspections or l 4 whatever.
5 DR. KERR: You're pulling my leg.
6 MR. WILLIAMS: No.
7 DR. KERR: Surely. Because all the SALP reports 8 say that they're nc,t objective, that they are subjective, 9 that they vary.
10 MR. WILLIAMS: Well, that's a vulnerability, and 11 we recognize that.
12 DR. KERR: But tell me why system performance is 13 not a good indicator of risk.
14 MR. JOHNSON: It is a good indicator, and we're
()
15 trying to develop a good indicator for that. We're not to 16 the point where --
17 DR. KERR: Now, wait a minute. The indicator of 18 that is simply whether the system is available or 19 unavailable, isn't it?
20 MR. JOHNSON: Yes.
21 DR. KERR: And you don't have data on that?
22 MR. JOHNSON: That sort of data is not reported to 23 the NRC.
24 DR. KERR: I thought any time a safety system was 25 unavailable and the licensee knew about it, it had to be
() Heritage Reporting Corporation (202) 628-4888 i
44
() 1 reported. ,That's not the case?
2 MR. MICHELSON: Not necessarily.
3 MR. JOHNSON: . That is true, but not at the train 4 level, so you could make a probabilistic --
5 DR. KERR: Well, but it's the system that I'm 6 concerned about.
7 MR. JOHNSON: That's one of the indicators right 8 now.
9 DR. KERR: Not the component or the train.
I 10 MR. MICHELSON: By system, do you mean both trains 11 unavailable or one train unavailable?
12 DR. KERR: I mean whatever it is that will make j 13 the thing function when it's called upon. It would have to
/~\ 14 be the sort of thing that you'd look at in a PRA if you were V
15 trying to determine whether ECCS is going to work or not.
16 MR. JOHNSON: Those data are not routinely 17 reported to the NRC.
18 MR. WILLIAMS: Again, we had to separate the 19 issues. We had to separate risk monitoring from safety 20 system unavailability monitoring, and that is reported to us 21 pursuant to 5073, from attempting to monitor the 22 effectiveness of maintenance. It's difficult. Initially, 23 we tried to use safety system unavailability in that, but it 24 was in vain.
25 MR. CARROLL: What did you find? Why was it in
() Heritage Reporting Corporation (202) 628-4888
45
("% .
t ; 1 vain?
ns 2 Guys that you thought had bad maintenance programs 3 had different safety system availability? Was there no 4 correlation?
5 MR. WILLIAMS: Well, we can show you some graphs, 6 I think, next week, but we found is that we just couldn't 7 mnke a case for it. We couldn't make a case that safety 8 system unavailability was determined to be maintenance 9 effectiveness. You have standby systems --
10 MR. CARROLL: I wonder what that tells you?
11 (Laughter.)
12 MR. JOHNSON: I believe that there is a connection 13 --
14 MR. WILLIAMS: It has a lot to do with the model,
()
15 too.
16 MR. JOHNSON: -- and so our objective is to try to 17 show this with work we're doing, but we're not to the point 18 where we can see, here, see.
19 MR. MICHELSON: We're apparently at the point 20 where we can write a regulatory guide about all of this.
21 That's what we have in front of us, and I'm still puzzled as 22 to what do we really know about the component level 23 objectives, and how do we write component level objectives, 24 or how does the utility really write component level 25 objectives at this stage of the game, when we're having a O
(,) Heritage Reporting Corporation (202) 628-4888
g.
y 4 46
() -1 little trouble trying to understand even system level:
'2 objectives. f
' 3: MR. KING: I don't think it's as complicated as,;
4 maybe, you_think it is. If you have a~ system'1evel -f 1
5 objective, and you're finding out that one or two or three- ]
6 components in that system are the cause of.your 7 unavailability problems, you could set some. specific-goals 8 'for those components.
9 MR. MICHELSON: Well, I guess I could, but you're 10 beginning to get to be a very large program indeed if you 11 start dealing now with'amall, individual components and 12 their particular objectives and then a hierarchy to this 13 system and a further hierarchy to-the entire plant.
NR. KING: A lot of problems. There could be a'
( } ' 14 15 lot of monitoring.
16 MR. MICHELSON:' Could be a lot of modeling,
~17 NR. KING: Again, it's something that we're trying.
18 to let the licensees figure out what's the best way to do ]
19 this.
20 DR. KERR: Are you willing to permit the licensee 21 to only work at the system level and to set a system level 22 objective and show that he meets that?
23 MR. KING: I think the reg guide allows the 24 flexibility. If he can make a case that that's all he 25 needs, the reg guide allows that.
Heritage Reporting Corporation (202) 628-4888
- ' - - ~ ~
.; ;g j; T'.
. 5- 1 '
f
, '47
.t -1 . MR. WILLIAMS: I think the-reg guide has a graded
.2 approach..
3 ., MR. MICHELSON: It does.
4 MR. WILLIAMS: In the area of maintenance, and
, 5 it's difficult. to cope with maintenance and risk in'a 6 quantita.tive sense and try t'o tie the two together via 7 safety system unavailability, but in the area of maintenance 8 --
9 DR. KERR: Look. Risk doesn't have anything but a
[ 10- quantitative sense. If you can't deal with risk in a 11 quantitative sense, you're not dealing with risk.
12 MR. WILLIAMS: Okay, but in the area of 13- maintenance --
14 DR. KERR: If there does not exist a relationship
(}
15 between maintenance and risk, we're going to a lot of 16- difficulty.for no good reason.
17 MR. WILLIAMS: I agree with that, and in the 18 regulatory analysis --
19 DR. KERR: And if we can't establish at'this point-
~ ;2 0 that there's a relationship between maintenance and risk, I 21 . guess I wonder.
22 MR. WILLIAMS: I think qualitatively, it's 23 something I think we can all accept. Safe operation is tied 24 to good maintenance. To quantify that --
25 DR. KERR: But that may simply mean, in the l
O Re ie - R o ein, (202) 628-4888 cergeratien l
1 1
i 48 j l
(( 1 context in;which we're talking about, that there are pretty f 2 good maintenance programs out there, and they're' good enough 3 so that the system availability is acceptable. I don't 4 think there's any plant that has zero maintenance. Do you?
5 MR. KING: No.
'l 6 DR. KERR: What we, presumable, have concluded as 1 7 a body is that the maintenance programs are not zero but 8 they're not, good enough, they don't reduce the risk to the 9 level that we want to achieve.
10 MR. KING: We've concluded that maintenance 11 programs can be improved, and they can be improved on a
- 12. cost-effective basis, and that's the rule and the reg guide 13 and the regulatory analysis that we're attempting to put forth.
[ J 14 15 DR. KERR: Come on. We are not setting up 16 maintenance as an end in itself, surely. What we're trying 17 to demonstrate is that we can reduce risk, isn't it?
18 MR. KING: And that they will reduce risk, yes. A 19 cost effective process involves reducing risk.
20 DR. KERR: But if we can't establish a 21 relationship between a maintenance program and risk, how 22 will we ever know whether the maintenance program --
23 MR. KING: The regulatory analysis we put out back 24 in the package that went out in November made an attempt to 25 establish that. We got a lot of comments on the methods Heritage Reporting Corporation (202) 628-4888
49 r
(yj 1 that we used, the data we used. We're trying to factor 2 those into the final analysis.
3 DR. KERR: Okay. You need to get these guys on 4 board and show them how you did it, even though they haven't 5 been able to do it. Is that a fair statement?
6 They haven't been able to establish a relationship 7 between system performance and maintenance.
8 MR. KING: I didn't hear them say that.
9 DR. KERR: Well, I thought they said they were not 10 able to establish a good correlation.
11 MR. JOHNSON: We were able to show a correlation 12 in a couple of cases. In a third case, it did not show at 13 all. We have some postulates as to why, so we are 14 continuing the work to try to prove it.
()
15 We have defined in the rule maintenance, And the 16 purpose of maintenance is to continue or uphold the q 17 availability of syecems that are related -- I shouldn't use 18 those words -- to maintain the availability of systems.
19 Surely the.t is the risk.
20 DR. KERR: Well, the rule says that you ought --
21 it says something about safety, and I would hope that 22 wouldn't be eliminated, and I assume that safety is somehow 23 related to risk.
24 MR. JOHNSON: Yes. And availability.
25 Availability of systems.
A)
(_ Heritage Reporting Corporation (202) 628-4888
4 50
['/l w
1 DR. KERR:
Well, availability is related to risk, 2 yes.
3 MR. JOHNSON: Yes. Directly.
4 So we agree.
5 MR. MICHELSON: Could I give just a slightly 6 different pitch on the same question, since it still 7 troubles me a little bit. l l
8 The plant technical specifications, as near as I 9 can tell, are pretty much system oriented, ,although there 10 are some components specifically singled out in the plant 11 technical specifications. Apparently, we built these 12 specifications on the basis of assuring an adequate level of 13 safety. Why isn't the maintenance rule, then, kind of 14 folded into that level of detail, namely, mostly system
()
15 level of detail, as opposed to now talking about components?
16 Depending on how I read this and how I dream about it, it 17 could include a great deal of component level consideration, 18 way beyond what the plant technical specifications normally 19 worry about.
20 Why wasn't this just tailored to the level that we 21 have tailored the plant technical specifications to?
22 MR. KING: I'm not sure.
23 You're saying, why don't we write the maintenance 24 . rule just to address maintenance on the components and i 25 systems that are in the tech specs?
I k) Heritage Reporting Corporation (202) 628-4888
51 i
f( ) ' 1 MR. MICHELSON: Sort of like that, just as a 2 starting point. Why wouldn't that be a better way to go 3 than just going back and sorting out all the structures, 4 systems, and components and setting objectives, I assume 5 objectives all the way down to the component level. j 6 Admittedly, it's graded according to importance, !
7 but it isn't necessarily that unimportant items don't have 8 any kind of objectives. I just don't know, from reading the 9 material that I've-looked et.
10 MR. DEY: I don't think we're saying that you need 11 to set objectives at'the component level across the board.
12 But you're having problems with the systems, and you 13 determine that it's not meeting its objective'that you set 14 out, then you would look at a lower level, at the component
()
15 level, see what's causing the problem, and set some 16 objective for correcting that, for improving that 17 reliability so that your system --
10 MR. MICHELSON: Before we get too far off, why 19 shouldn't we approach this from a viewpoint of assuring the 20 availability of the items that were considered important 21 enough to be in the plant technical specifications, which 22 are predominantly systems, although some components are 23 there?
24 MR. KING: Well, I think the maintenance rule 25 certainly includes everything that's in the tech specs.
() Heritage Reporting Corporation (202) 628-4888 i
I
52 MR. MICHELSON: Well, undoubtedly it includes all l( ) 1 2 of that, and I think a lot more.
3 MR. KING: And a lot more. -
4 MR. MICHELSON: And what's bothering me is the 5 importance and the effort of getting all the rest of it in.
6 What are the rewards? What would we lose if we only pitched 7 at the plant technical specification level? ;
8 MR. KING: There's a lot of things in the plant 9 that contribute to risk. A lot of failures come from the 10 BOP, which is not covered by the technical specifications. !
11 MR. MICHELSON: You're saying that there are 12 important components that we don't tech spec? ,
13 MR. KING: There are a lot of systems and 14 components --
()
15 MR. MICHELSON: Then I would ask the question, why 16 not? As opposed to going in an expanding the maintenance 17 rule as the mechanism of somehow encompassing these. If 18 they're that important, then I would wonder why they aren't 19 tech specced.
20 MR. KING: You're going tack ~to the philosophy of 21 tech specs, which was established a long time ago, that it .
22 primarily deals with safety systems, safety-related systems, ,
l l
23 and preserves the assumptions from the Chapter 15 FSAR 24 analysis. 1 1
25 MR. MICHELSON: You're saying that that's not --
() Heritage Reporting Corporation (202) 628-4888 j l
l l
53 !
( ]) 1 those.really are inadequate, and that you' re going to take 1 2 part of that inadequacy with an enhanced program of 3 maintenance on certain items that, perhaps, you thought f 4 ought to have been in the tech spec. I'm puzzled as to what j 5 kind of a web we're weaving.
6 1G1. CARROLL: They own up, Carl, in this thing --
7 MR. DEY: I think we rrecognize that the scope of 8 this rule is a policy change.
9 MR. MICHELSON: But why don't we go back and fix 10 the tech specs and then tailor the maintenance program to 11 assure the availability of the items in the tech specs?
12 MR. CARROLL: I think we were there earlier. How 13 come we don't have surveillance testing on this stuff, ,
14 requirements for operation, requirements for QA on spare
( })
15 part.
16 MR. MICHELSON: You'd pick that up if it were in 17 the tech specs.
18 But that's not the approach we use, I realize.
19 I'm just puzzled as to why we're using this approach to get 20 at really a different problem.
21 MR. KING: It certainly is an approach, expand the 22 tech specs to cover a whole bunch more things than they 23 cover now.
24 MR. MICHELSON: Well, expand thom as required to 1
2$ cover the things that aren't covered, whatever amount that (m_) Heritage Reporting Corporation j (202) 628-4888 1
I I
l
y >- . - - - - - - - - - . --- - -
k r<
([1 . .
{
54
(] ' l is,'that are.really that important to aafety, and then do good maintenance, assured maintenance at a high level on j 3 those, and then the other maintenance is me di more of a i
4 balance of the program that would not be very prescriptive.
i 5 MR. WILLIAMS: Carl, let me try to give you 6 something that helped me out a little bit. It's a very 1
7 difficult issue. This issue has troubled us a lot, too, and 8 I don't know the answer to it, but what helped us out a-9 little bit, at least the indicator' folks, is that the rule 10 was originally aimed at maintenance programs, the quality of 11 the program. It wasn't really aimed at systems, structures, 1
12 and components. It was aimed at having an overall l 13 maintenance program.
14 If you look at the NPO guidelines, it doesn't
()
15 really focus on systems, structures, and components. It 16 says your program has to have all these different kinds of 17 elements in it, they all have to be working, you ought to 18 make sure they're working, monitor them, and this and that.
19 The focus of a lot of the current initiatives is 20 programmatically oriented, and I think that was consistent.
21 What happened was that there were a lot of q 22 comments on scope. I thihn one of the major concerns at the 23 outset was scope, what's the scope of the rule.
24 Immediately, one turns and tries to describe that, and 25 that's what has been done. It's very difficult to deal with l
\ -
() Heritage Reporting (202) 628-4888 Corporation l
1 l
'1 ., [
3.i -
y.j ' .i . ,i i
W^ .la, '
?' 55 ]
~
( 1 En those terms.
2 The' industry appears to be' dealing with.it on a. q r ,x , .t1
' 3 programmatic level. . You have a graded program, you j 4 establish a master list of equipment, you apply different I
'5 requirements fer different equipment, standby systems, ,
]
i 6 running systems, and~you have dif'ferent elements of the. {
l 7 . program.
8 When we tried'to monitor;the effectiveness of. ;
I i
9 maintenance, we had to look at it:in pretty much the same 10 way.
11 So the only thing that~has really helped us a
, 12 little bit is-to look at it as programmatically oriented 13 rule and reg guide. That helps us, but it doesn't answer
()'14 your questions. i 15 MR. MICHELSON: You're really doing n. ore than 16 looking at it at a program level. You're taken that cup, 17 and now you're getting right down to components and ;
18 objectives on individual components. That's quito a jump, 19 from a program down to a component level. . It's jump that 20 even the tech specs don't necessarily make, except for a few 21 specific componente. ;
\ l 22 I don't know. Again, the nice words, it's going 23 to be graded, and all that, but 7 don't know how much it's 24 going to be graded, and I don't know how much you're going 25 to'have to prescribe objectives on each component, or l
l -
h Heritage Reporting Corporation (202) 628-4883 L_ _ _- _ _ - _ . i
1 l.4 ^ < G 56 j
1- whether you can keep just. system level objectives or what.
L(])
L, . , 2 It isn't clear.
3 But it could get way out of hand. It could be a 4 mountain of paperwork to sit down and prescribe these 5 objectives and then track to see how you're meeting your 6 objectives. It seems like it would be kind of useless at j 7 many of the component levels.
8 The utility, of course, is' going to decide, I 9 guess, but now you're going to get into endless arguments.
10 with the inspectors as to whether you shouldn't, include 1
11 this, and why don't you include that,.and so on. It could 12 get out of hand easily.
13 It just doesn't seems to fit. We, apparcntly, are quite satisfied with the technical specification as the
(])14 15 controlling document for operation of the plant, and its 16 level of detail. I haven't heard anybody saying that's an 17 inadequate lovel. But on-maintenance we're saying it is an 18 inadequate level.
19 MR. KING: There's a lot of regulations that go 20 along that affect a plant besides technical specifications.
21 MR. MICHELSON: To be sure, but it is really kind 22 of, I gather, one of the controlling documents for the 23 operations of the plant. And maintenance is very closely 24 related to the operation.
25 MR. KING: One could take an approach that we're Heritage Reporting Corporation (202) 628-4868 i
L
57
( 'l ~
going to define all kinds of detail, in terms of what 2 components are covered and what maintenance has to be done, 3 what their.re11 abilities have to be.
4 MR. MICHELSON: Clearly, you wouldn't want to do 5 that.
6 MR. KING: No. We don't want to do that. We 7 think the burden ou7h t to be on the licensee.
8 MR. MICHELSON: But in the words you've put in 9 your document I can easily find that somebody can construe 10 that. I mean, you talk about etructures, systems, and 12 components. There's a lot of components.
12 DR. KERR: But if you place a burden on a 13 licensee, you want to have a fairly clear idea of what it is 14 you're asking them to do. It would be helpful, even, if you
(}
15 knew why, although maybe that's not necessary, 16 It seems to me what one is trying to do is to 17 reduce risk.
18 MR. WILLIAMS: We agree with that.
19 DR. KERR: Therefore, there ought to be a fairly 20 clear line of logic that would connect what it is we're 21 going with risk reduction.
22 Take, for example, the Canadians. They did at one 23 time, and I think still do, have a requirement for control 24 system reliability, availability, a number. As far as I 25 know, they didn't specify anyt.hing about component behavior.
() Heritage Reporting Corporation (202) 628-4888 i
i
1 l..
Y s-58 1 They just said,~you've got to demonstrate that.the
(}
2 reliability of the control system is this.
3 I think it has worked-fairly well. RIt's a number 4 that's high enough so that, within reasonable time, one can 5 probably demonstrate that it's being schieved or isn't being 6 achieved.
7 .By the way, when the ACRS was discussing with the i 8 Staff the effect of control system on safety - you'll 9 remember that generic issue, perhaps -- this was at least 10 discussed, and at that time the idea was not greeted very 11 warmly. It may not be practical in the U.S. environment; I 12 don't know. But it always struck me as being a reasonably 13 logical approach and apparently was workable for the !
l 14 Canadian system.
()
15 MR. KING: I don't think there's anything in this i 16 rule and reg guide that would prevent a licensee from taking 17 that approach.
18 DR. KERR: Well, if that's the case, then I would 19 assume that they could establish system level performance 20 guidelines, and that would be it. They wouldn't have to 21 worry about components.
22 MR. WILLIAMS: I don't thing we set performance 23 level or reliability values for components, and I don't 24 think, at least in the monitoring, the indicator section of 25 this, to monitor the effectiveness of maintenance, we
() Horitage Reporting Corporation (202) 628-?888 i
59 3-
[ 1 ,didn't. ,
2 DR. KERR: No , but' you're going to ask a licensee 3' to-do that.
y 4 MR. MICHELSON: That seems to-be what you're
- 5' asking.for. That's what you talk about up there. I thought 6 that.you were talking about performance objectives for 7- things right down to the component level. SSC, structures, 8 systems, and components.
y 9 MR. DEY: In section 5, we do say that maintenance 10 effectiveness indicators based on component performance 11 should be used. j 12 MR.' KING: We don't say anything in the section 13 that talks about setting goals that it has to be at the 14 ' component level. You're right. In section 5, we do talk
(}
15 about measuring component failure rates.
16 MR. MICHELSON: Well, I think you can measure 17 component failure rates without setting component 18 objectives, necessarily. You need those failure rates to-19 get a system objective identified. Those components make up 20 a system, so you've got to get the data.
21 Maybe I just misunderstood the slide; maybe the 22 slide is not really whr.t you're asking for in the guide.
23 I'd have to go back and study it again carefully. Your 24 slide talked about SSC performance objectives, and C meano l
'25 component, and that means to me, then, component performance I
() Heritage Reporting Corporation (202) 628-4888
60
- . 1 objectives.
2; MR. DEY: Well, maybe we need to put an "or" in 3 there.
4 MR. MICHELSON: I'm looking at a previous from the 5 one that's there at the moment, of course. I'm looking at 6 your slide on criteria and acceptability.
7 MR. DEY: What we're really saying is, the B licensee should set an overall objective related to risk, to 9 improve risk, and set specific objectives. Yeah, it says 10 SSC, but I don't think wa're saying for all components.
11 It's hierarchy where you 'o a systematic !
12 evaluation and set objectives at a higher level. If you 13 aren't meeting it, then you would go down further and see 14 why you aren't meeting it, and perhaps set a gor.1 at the
{}
15 component level to produce the results you war.t.
16 MR. MICHELSON: You may or may not want to set one 17 at the component level. You certainly want to set one at 18 the system level. ;
19 MR. DEY: Right.
20 MR. MICHELSON: But, at the component level, you 21 certainly look at those to decide why you're not meeting 22 your system objective, but you don't necessarily change a 23 component goal. You may just decide to do replacement more 24 frequently, or whatever.
25 MR. DEY: Or hetter maintenance.
() Heritage Reporting Corporation (202) 6L'8-4Pd8
a j
. 61 1 MR. MICHELSON: Which, in a way, changes the goal, 2 I guess.
3 If you really don't mean component goals, then a 4 lot of my pain goes away. But if you're setting individual 5 component goals, only in a few cases would that be
)
6 important, perhaps scram breakers or something; that might. )
7 be important.
l 8 MR. DEY: Yes. !
I 9 MR. MICHELSON: And you would use the tech. spec as 10 a guide to the kind of components that I might think are so 11 important.that I tech spec them, and also I might even stant 12~ to put performance objectives on them.
13 If that was your intent, fine, but it's not clear 14 from reading all these gsneral statements as to how this l 15 could be construed. i 16 MR. KING: Yes, that's our intent. I think we'll 17 certainly look at the words again and see where we can 18 improve'it.
19 MR. DEY: As we have been talhing, we want the 20 licensee to make the relationship to safety and risk and 21' take these issues into consideration.
22 Finally, we expect the goals, the performance, to 23 be consistent with what's achievable in the industry, with 24 what top performers are doing, something that's achievable.
25- (Simultaneous voices.)
Heritage Reporting Corporation (202) 628-4888 i
2 62 1 MR. CARROLL: That's a moving target.
[a) '
2 MR. KING: Well, it may be a moving target. I 3 don't think that's so much a problem. I think the real l 4 philosophy behind it is, there are plants out there that are 5 doing maintenance very well. They're achieving high system 6 reliabilities, low component failure rates. They obviously 7 have found a way to do that that they think is cost 8 effective. I don't think it's unreasonable for us to, in 9 the reg guide, request that the other utilities shoot for 10 similar type goals. If some utilities can achieve that 11 performance in a cost effective manner, I think these other 12 utilities could learn from that, adjust their programs, and 13 also achieve that higher level of performance in a cost
\ 14 effective manner.
(J A.
'S There will be some benefit in terms of reducing 16 risk, and we think that that will be cost beneficial to do l'
17 that. That's the philosophy behind that.
18 MR. CARROLL: Cost beneficial is an interesting 19 concept in as broad a context as maintenance. What may be f 20 cost beneficial to utility A may be anything but cost 21 beneficial to utility B.
22 I can think of one particular utility that, 23 because of a rate case settlement, can spend any amount of l 24 money they.want to have a good maintenance program or a good ,
25 this or a good that. Another utility may be in a situation i
(,) Heritage Reporting Corporation (202) 628-4888 l
63 )
(( ) I where he's,got lots of generation, whether his nuclear plant 2 runs or not is worth a few hundred thousand dollars a day, 3 but not a great deal. He may have a much different problem 4 in justifying spending real mor.ey, particular1'y if he's in a 5 rate squeeze, or whatever.
6 There's no one test of cost effective maintenance 7 that I believe exists in the industry. Each utility has a 8 unique situation. It isn't fair to take some replacement ,
9 power number and say that's true of the whole industry.
10 In a lot of cases, utilities have rate muchanisms 11 where, whether they generate power from their nuclear plant 12 or from a fossil plant or go out and buy it from their 13 neighbor, they come out just as well on the balanca sheet as 14 if their nuclear plant had run 10 percentage points better
('/)
s_
j 15 in capacity factor.
16 There's no one measure of cost effectiveness.
17 MR. KING: I'm not suggesting there's a single 18 measure.
19 To use an example, if you've got a number of 4 i
20 utilities out there where their feedwater systems are able 21 to run for the full length of a gycle, 300 days, 400 days, I
l 22 whatever it is, without a failure that would cause a plant 23 scram, and there are other utilities out there whose I
24 feedwater system is causing them unanticipated scrams and l
25 shutdowns, the one who's having problems, it seems to me,
/
()
y Heritage Reporting Corporation (202) 628-4888 1
1
64
(~'I 1 ought to be able to learn from the ones that are doing it v
2 well and set as a goal trying to do it as well as the ones l
{
3' that are doing it well. l l
4 MR. CARROLL: I may know how to do this, seeing l 5 the experience of other plants. It means a million-dollar {
1 6 backfit. I don't think I want to spend a million dollars. !
l 7 I'll accept the one scram a year. What do you say to me, a l I
8 licensee? Is that bad? j i
9 MR. KING: No. I don't think this would say l 10 that's bad. I think, if he goes through that process and 11 figures out --
12 MR. CARROLL: Okay. I'll accept three scrams a ;
13 year. Am I a bad guy then? I'm just not going to spend the i
{} 14 additional capital bucks. I'm not going to get rate relief 15 for it, or whatever. I'm going to have the three scrams a 16 year. I know I could reduce it to .5 scrams a year, but in 17 this latest and greatest new backfit of the feedwater 18 system, I'm just not going to spend the bucks. What are you 19 going to do to me?
20 MR. KING: I think we'd have to look at it. I'm 21 sure there's going to be a gray area where there's going to 22 have to be judgments like that made. That may be one of 23 them.
24 MR. CARROLL: At ten scrams a year, you probably 25 would have a case.
,s
() ,
Heritage Reporting Corporation (202) 628-4888 I
l
v ;e 65
,L, 1 MR. KING: You mity.
'2- MR.-CARROLL': But there are those real world
- 3 problems. We're not dealing with'a single. utility with.a 4, single resource situation.
5 NR. KING: That's one of the reasons we went to 6 try and.take this' approach.that didn't prescribe how to do- .
7 _ maintenance,. didn't prescribe numerical goals, that sort of' 8- said, what we' re really interested in is this process, that 9 you look at your program, figure out what its goals and 10' objectives should be, and then have a process that. improves' 11 it to meet those and leaves a' lot.of flexibility to a 12 'l'icensee to go figure out the details of hott to do that.
13 MR. CARROLL: What I'was reacting to was that last' 14 bullet, Tom. You're really saying everybody's got to get
()
15- with the program and get consistent with the top performers, i
16 Maybe the top performers have an economic incentive that the ;
17 guy that's performing at the fourth quartile doesn't have.
18 DR. KERR: " Consistent with" could mn9n "one tenth i L !
19 of." It doesn't mean " equal to."
t L 20 MR. KING: It doesn't mean " equal to," but one 21 tenth of, I don't know if I'd buy that.
22 DR. KERR: I don't see anything inconsistent with 23 that.
24 MR. KING: Again, it's a judgment.
'25 MR. MICHELSON: Are you going to review somehow Heritage Reporting Corporation (202) 628-4889 i
j
m ,
4:
! l' s
I; 66 1 the licensee's program? Is he going to write up his' m
(]) . ,
2 maintenan'ce program and send it to you as a document and 3 say'-"Here's my program"?
4 HR. KING: No.
5 MR. MICHELSON: I didn't'think heLwas, either, but 6 --
i 7 MR.. KING: He'11 have-it~on site, and --
8 MR. MICHELSON: He's going .to. prepare a program of 9 some sort. y 10 -MR.. KING: The rule requires that he have a. ,,
11 documented program.
12 MR:. MICHELSON: Right. It's going to be on site.
13 It'll not ever be sent unless you ask'for it. -;
i
('3
\~/ .
14- MR.' KING: Right.
15 MR'. MICHELSON: The inspector, I guess, is going 16 to have to be the sole judge of whether he is abiding by his .
17 program. He's going to inspect against the' program, I {
l 18 guess. q 19 MR. KING: The resident inspector, on a day.to day 20 basis, can look. j 21 MR. MICHELSON: How does he decide, to begin with, ,
i 22 whether the program even meets the intent of the rulo? Is I
j, 23 he the one .nat has to decide that, the inspector?
24 MR. KING: Well, I'm not sure whether, at this l l 25 point, we've decided whether it would be just the inspector j i
s l
l'
() Heritage Reporting Corporation (202) 628-4888
)
]
l
'll ., '%
ti.
P '
67 l
[ L.1 or.whether,there would be some sort of -- like these
'2; maintenance team inspections that are going on now. l
- p .
3 MR. MICHELSON: You haven't decided yet to what 4 extent you will check out the basic framework of this
- 5. Lprocess.
6 MR., KING: No, I don't think that's been decided !
7 at.this point.
8 MR. MICHELSON: You can't expect the inspector to 9' do it.
10' MR. DEY: It is planned to develop an inspection
'll guide.
.12 MR. MICHELSON: So you will do an inspection to j 13 determine that that the basic program is sound, whatever it is. 'Apparently, it can be many things, into many levels of
[] ' 14' .
15 detail, as long as you can satisfy the inspection team that 16 .it's adequate.
17 . MR. KING: Certainly the inspection and audit 18 process will determine whether it'o adequate. How that's 19 going to take place hasn't been determined at this point.
20 MR. MICHELSON: Okay. And, having determined that 21 the program is adequate, then the local inspector just sees 22 to it on an auditing basis from time to time, that you're i
23 doing what you said you were going to do.
24 MR. KING: That's one way to do it. Again, I'm 25 not sure we've figured out exactly how that's going to be
, .O erie e- Re ,ereine cergeraeie-(202) 628-4888
o68
() 1- done yet, the split between what the resident does versus 2 the --
3 MR. MICHELSON: So I guess the most important 4 document, ultimately, will be that inspection guide that 5 indicates how the Staff really expects the utility to do 6 this, i
7 MR. KING: Well, it's certainly an important i
8 document.
9 MR. MICHELSON: And that comes, of course, much 10 later. Several months from now?
11 MR. KING: I don't know.
12 MR. DEY: There's also the other option, where the 13 programs would be reviewed by a third party.
14 MR. MICHELSON.- Assuming that that doesn't go
()
15 through, that the NRC does this, at the time this becomes a !
16 rule, which is six months after, by the end of the year, 17 will this irispection module be available by then?
18 MR. KING: I car.'t give you a date. I'd have to 19 ask the NRR people and see what their plans are.
20 MR. MICHELSON: I would like to put in the record 21 that the ACRS subcommittee would like to see the inspection I-22 module after it's drafted. We would like to look at that, 23 because I think a lot of what we' re worrying about now might 24 becomes clearer one way or the other at the time the i 25 inspection module is issued. If we have real problems, we Heritage Reporting Corporation (202) 628-4888
,s !
.69 I could write a letter at that-time commenting on that' t
(f 2 inspection module, if we so wish.
3 I think it's an important document to see.
4 MR. CARROLL: Do you see that inspection module 5 being much different than the one you have right now for the 1
6 team inspections? That's a pretty comprehensive inspection 7 module.
8 MR. KING: It's comprehensive. My own personal 9 feeling is, maybe in a couple of areas, it would have to 10 have some enhancement, particular in the area of the goals 11 and objectives kind of monitoring. Maybe the feedback 12 process monitoring. But it's very comprehensive in terms of 13 the elements of a maintenance program right now.
14- MR. CARROLL: Getting back to this issue of risk
[}
15 versus system availability. H l
16 Suppose I've done a PRA, and NRC has looked at it 17 and said, " Yeah, that's a good PRA," and in that PRA the 18 assumption of availability of, say, the auxiliary feedwater 19 system, is -- I' d say, "Okay. That's going to be my goal."
20 At least historically, my data says I'm meeting that goal.
21 Would that be a way out of this. Would you say, "Well, no.
22 You've got to do more than that, because, looking at the 23 record, you've got a failure of this component. That could 24 have led to a problem. You've had that kind of failure 25 three or four times."
Heritage Reporting Corporation (202) 628-4888 i
- \ . <
70 .
h .1- MR. WILLIAMS: I think the regulatory guide and 2 the guidance leaves.to the l'icensee the ability to. monitor.
3 the availability of those systems and set those goals 4 suitable 1for the plant.
5 As far as monitoring the quality of maintenance or 6 the effectiveness of maintenance, we have' separated that as
- 7. a separate monitoring item, 'In other words, a simple model 8 would be, if one monitored the amount of rework on a set of.
9 components, and there was a lot of rework, a lot of poor 11 0 maintenance initially that had to be reworked,-that would be 11 a monitor of the effectiveness of maintenance, although it 12 may not be risk-significant.
y '
13 }R. CARROLL: It may not be because, when I do 14 rework, it's because.I've done maintenance, I've now done
(}
- 15. post-maintenance. The damn thing doesn't' work. I'm turning 16 it back to maintenance. That's economic.
17 MR. WILLIAMS: Exactly. Right.
18 MR. CARROLL: That's economic, except for the 19 extended maintenance.
20 Are you telling me I don't have the option of
'21 deciding what my economics would be?
22' MR. WILLIAMS: No. All of those goals are left 23 to the utility, and it's left to the freedom of the licensee 24 to choose those goals. I was simply trying to separate, at 25 least conceptually, the model of what constitutes monitoring Heritage Reporting Corporation (202) 628-4888 l
.. -_____ _ a
l 71 i
.( ) 1 the quality of maintenance or the effectiveness of 2 maintenance and monitoring the availability of safety 3 systems.
1 4 In other words, we've just been dealing with those j 5 as separate items. We haven't had the skill yet and the 6 data to put all that together.
1 Well, if you don't have the skill to 7 DR. KERR:
3 correlate what you call the quality of maintenance with the 9 reduction of risk, how do you know you're monitoring quality 10 of maintenance? Presumably quality of maintenance ought to 11 be related to risk reduction. If it isn't, and you don't 12 know how to relate it to that, how do you know you're 13 monitoring quality? You may be picking the wrong 14 indicators.
(}
15 MR. WILLIAMS: I think one can monitor the 16 maintenance on safety systems. One can monitor their 17 rework, or one can monitor the number of failed post-18 maintenance tests on safety systems.
19 DR. KERR: Unless that's related to availability, 20 how do you know it's related to risk?
21 MR. WILLIAMS: Well, conceptually I think we see i
22 eye to eye. When it comes to proving these with available I
l 23 operational data, then it gets a lot more difficult to show 24 a correlation.
25 MR. MICHELSON: I guess you expect the utility, !
A
(,) Heritage Reporting Corporation (202) 628-4888 :
72
(( ) l' though,'to,somehow know what correlations are valid and know 2 how to .nonitor and how to set its objectives and monitor 3 that he's meeting them. If you don't know how to do it, do 4 you think the utility is even more understanding of this 5 problem than you are and will come through anyway.
6 Possibly, that's the case. I don't know. But it !
7 does puzzle me that, if we don't understand yet quite hov 8 this could be done, why do we believe the utilities will?
9 MR. KING: Well, I think we could propose a scheme 10 to do it. I just think that that's not the right thing to 11 do.
12 MR. MICHELSON: I didn't sense you have a scheme 13 in mind that you think will work.
(~)
v 14 MR. WILLIAMS: We didn't spend any time trying to 15 come up with a scheme to put in here as an example.
16 MR. MICHELSON: But even without putting it in 17 here as an example, do you have a scheme, you think, that'll 18 accomplish all of this. You don't have to put it in the 19 rule, but you could tell us about it.
20 MR. DEY: Well, as we develop the reg guide, we 21 have been working and have ideas on that, but nothing on 22 paper.
23 MR. MICHELSON: Yeah. I expect to have a lot of 24 ideas, but they need to be tested and developed to make them 25 practical enough to implement on a very wide scale.
() Heritage Reporting Corporation (202) 628-4888
s 73 t
() 1 MR. DEY: It's been done in other programs. We've 2 seen examples of this, this scheme and other nuclear power 3 1 programs in. Japan, France, and in the airline industry. So 4 it's not something that --
5 DR. KERR: You mean we're willing to' accept a
'6 level of safety that the airline industry? I didn't think 7 we could get away with that.
8 MR. KING: That's not what he said.
9 MR. CARROLL: I guess I take exception to that.
10 Do you really believe that this scheme that you have up here 11 is something you guys invented?
12 MR. DEY: No. I 13 MR. CARROLL: It seems to me it is exactly what 14 most responsible utilities are doing.
()
15 MR. DEY: That's absolutely right.
16 MR. CARROLL: Back into the days of fossil power.
17 MR. DEY: Sure.
18 We may be sharpening our pencils and --
19 MR. KING: This isn't going to be a big burden on 20 a lot of people. They're probably doing this --
21 MR. CARROLL: Probably be a big burden to document 22 it in a form that allows you --
23 MR. KING: There will be some documentation 24 burden.
25 MR. DEY: We want to minimize that.
(O
_) Heritage Reporting Corporation (202) 628-4888
.. l 1
- 74 i
' I 1 DR. KERR: What fraction of the nuclear utilities 1 2 out there do you anticipate ow have acceptable maintenance.
. i 3 programs?
4 HR. DEY: I think we're going to' respond to that.
5 question later this afternoon. We have some data from the 6- maintenance team inspections, and we'll present that.
7 MR. KING: We have a number later on,'44 percent, 8 based upon the latest maintenance team inspections, that we !
9 think would have minimal impact by this and be consistent 10 with that.
11 MR. MICHELSON: And they're doing performance 12 objectives down at the component level, which is what this 13 . talked about?
-14 NR. KING: Well, it doesn't require.you to go down
(
15 to the component level.
16 MR. MICHELSON: Well, I don't know what it 17 ' requires .. I'm puzzled, confused by what it really requires, 18 but it talks about it down to the component level. SSC 19 means structures,. systems, and components. And you talk 20 about performance objectives, down to the component level.
21 Do utilities out there now keep performance objectives kind 22 of across the board down at the component level, or is there
-23 an occasional component that's giving them so much trouble 24 they're keeping tabs on that component?
25 MR. KING: I don't know to what extent they go to Heritage Reporting Corporation (202) 628-4888
I y 75
() I the component level.
2 MR. MICHELSON: I' don't, either. That's a big 3 part of the program, trying to prescribe down to the j 4 component level. System level, I kind of think that we may 5 not be in too bad a shape, but I'm not sure at the component j 6 level that we do much at all, in terms of documented and 7 tracked, not just out of the hip pocket.
8 MR. KING: I guess the only message I was trying 9 to say was, there is some recent inspection data on a 10 sampling of plants, and from that sampling of plant, about 11 44 percent got the top rating from the maintenance 12 inspection team. We think those plants with those kinds of 13 programs would have a minimal impact by this rule and reg 14 guide.
(}
15 MR. MICHELSON: It would be interesting to take 16 what you think are even the best of that sample and send 17 them this guide and ask them, do they think they comply with 18 it.
19 DR. KERR: How big a sample was that?
20 MR. KING: Eighteen sites and 27 plants.
21 DR. KERR: Twenty-seven units.
22 MR. MICHELSON: That's a large sample.
23 MR. DEY: I'm sure we'll get responses during the 1 24 public comment period.
25 MR. KING: NRR headed that up, but that's the
() Heritage Reporting (202) 628-4888 Corporation
76 1 number that are done that are documented today. There's
[v) 2 others that have been done where the documentation's still 3 in preparation.
4 MR. CARROLL: The two I looked at in some detail 5 and actually spake to the utilities that were subjected to 6 theic I found a wide variation in the approach the people 7 used, even having the inspection module in front of them.
8 In one case, I think, the people doing the 9 inspection wanted to make that particular utility look good.
10 In another case, I think they had exactly the opposite l 11 intent, reflected in the quality of the inspection report.
12 MR. KING: Well, I can't comment on that. It's 13 not the same group of people that goes to each plant, so
("T 14 there is some variation.
V 15 MR. CARROLL: I know.
16 I don't take a lot of comfort in your 44 percent 17 that are good. I don't know what that means, really. ,
18 MR. KING: All right.
19 MR. DEY: We'll come back to that.
20 I have just one other slide, and that's on the 21 scope.
22 As we mentioned, we modified the rule to indicate 23 scope for those components that would impact safety and 24 security. .In the guide, we have identified those structures I l
25 and systems relied upon for cooling pressure boundary, safe ?
O(,, Heritage Reporting Corporation (202) 628-4888 i l
t 77
'( ) 1 shutdown, prevention and mitigation. This is what's 2 normally the scope of Appendix B.
3 Secondly, those SSC whose failure would initiate a 4 transient or challenge a safety system.
5 Three, other systems required for safety, fire 6 protection, security, safeguards, emergency preparedness, 7 post-accident monitoring.
6 D is one which we talked about this morning.
9 DR. KERR: You wouldn't consider saving D for the 10 next reg guide, would you? I think that's so important it 11 ought to be saved for a special reg guide and not put in.
12 here.
13 MR. DEY: Which item? D?
14 DR. KERR: Yes.
(}
15 MR. DEY: You think it's so important that it 16 should be saved --
17 DR. KERR: Put in some other reg guide that you're 18 going to develop about 10 years frora now and left out of 19 this one.
20 MR. DEY: So you're saying we should take it out.
21 DR. KERR: I think it's almost meaningless. It's 22 in the eye of the beholder. It can be anything.
23 MR. KING: Well, I'll tell you why it was put in 24 there. It was put in there to address the things that cause 25 a lot of nuisance alarms in the control room, that not only l
Heritage Reporting Corporation (202) 628-4888 i
l
- l 78-() 1 distract' operators but eventually become ignored when maybe 2 there's a real problem.
3 That was really the thought behind putting the 4 words in item D in there.
5 DR. KEFIR: You could say "whose failure 0
6 deliberately attracts the attention of plant operators" and 7 be talking about the same thing.
8 MR. KING: You could.
9 DR. KERR: It doesn't describe, it seems to me, 10 what you're trying to describe. j 11 MR. CARROLL: So some once in the plant lifetime 12 kind of failure out in the switchyard, he probably was 13 talking about, that has nothing to do with nuclear safety, 14 isn' c in the category that would distract the attention of
()
15 plant operators.
16 MR. KING: This thing could oe read to include the 17 stuff out in the switchr* d. I think there has to be some 18 judgment on what that includes.
19 MR. CARROLL: Let me try something else.
20 Let's try the building on site where you keep the 21 lawnmowers and fertilizer for the lawn in front of the 22 administration building that has a probability of catching 23 on fire once in the lifetime of 'ne -31 ant. That's going to 24 distract the operators, because taay've got to go put the 25 fire out. Does the role involve the maintenance of that
() Beritage Reporting (202) 628-4888 Corporation j
79
['J w
T 1 building?
2 If you're real int'ent is just the nuisance alarm 3 issuc, I'd say it more clearly.
4 MR. KING: If you want to abuse item D, you could 5 go as far as what you're talking abcut.
6 MR. CARROLL: I know.
7 DR. KERR: I don't want to abuse it. I want to 8 save it. I think that it's important enough that it ought j 9 to be --
10 MR. WYLIE: You could just take item B -- for 11 Baker, B -- and the switchyard is included, because a short f
12 circuit in the switchyard will shut the plant down, and 13 that's a transient. Just by item B, you'd take the 14 switchyard in there.
f~Nd 15 MR. CARROLL: Well, there can be things in the 16 switchyard that really have nothing to do with the plant. !
17 MR. WYLIE: Oh, I know that. But I can take the 18 potential transformer in the switchyard that's full of oil 19 that blows up and knocks out the switchyard.
20 MR. CARROLL: Oh, okay. I can do that, yeah.
21 In your C list, which is sort of my important to 22 safety list, not safety related, I also have in my grand 23 scheme of things radiation protection, chemistry. What else l
24 goes in there?
)
25 MR. KING: That's not intended to be a complete !
I) Heritage Reporting Corporation (202) 628-4888 l
_-_ ---- --- -- - - i
)
80 t
Those are examples. Radiation protectien's
(]) 1 list under,C.
2 covered under E.
3 MR. MICHELSON: We have item E.
4 MR. CARROLL: Okay. It is. All right. j L
5 Might want to add process chemistry.
6 MR. MICHELSON:. Might almost be easier to identify 7 what isn't covered on site. What isn't covered by this.
8 MR. KING: Well, we tried that ao part of t
9 developing this thing.
10 MR. MICHELSON: That didn't work, either?
11 MR. K2NG: We took a crack at identifying what is 12 and what isn't, i 13 MR. MICHELSON: I wasn't fruitful, either?
14 MR. KING: We thought it was better to try and
(}
15 just put it down in words. Each plant is different, and 16 licensees with these general words figure out what is in and 17 what is out and justify it. ,
18 MR. WYLIE: Do you really think that the scope as 19 written here is in keeping with the spirit of the scope as l
20 written in page 55 of the rule?
i 21 MR. DEY: You mean -- j 22 MR. WYLIE: To me, they don't compare. Maybe 23 somebody else -- )
24 MR. KING: They're intended to keep with the 25 spirit of page 55.
. l
/3
() Heritage Reporting Corporation (202) 628-4888
I i: i 81 L
() '
1 MR. WYLIE. Well, it says those things that j
'l 2 21gnificantly impact safety and security, and some of the 3 things we've been tal; ring ab6ut that fell under this scope !
4 are certainly not significant.
5 MR. DEY: Okay.
6 1 just wanted to summarize -- this is the last i 7 slide -- what our intent is in the regulatory guide: give y 8 gu:fdance on the procesa and not give a lot of details.
9 DR. KERR: By the way, I want to say that I think 10 that's a good alide.
11 MR. DEY: All right.
12 We've gotten as far as --
13 MR. CARROLL: I want him to add '" graded" to the
~'
/l LJ 14 middle box, if that's okay. I think it's a good slide, too.
15 MR. DEY: We've covered up to C-3. After lunch, 1
l 16 we'll get into C-4. That concludes this part of the 1
17 presentation.
18 MR. MICHELSON: By saying it's a good slide, you 19 mean the scope slide was the good one?
20 MR. CARROLL: No. I meant the flow chart of the 21 process.
22 MR. MICHELSON: Oh, okay.
23 MR. CARROLL: Except the middle box needs to say 24 " graded."
l 25 MR. MICHELSON: All right.
1 lju Heritage Reporting Corporation (202) 628-4888
u n_ , r - c
?, ;
_f-
. 82 O> 11 ; -I believe we're still about on schedule for the W.y"
- n. .. "
-2.', ' afternoon, as near as=I can tell. .Is that your
'3 .
understanding, too? !
a I
f '41 MR. DEY: Yes. I think wo are ahead of schedule.
'5 MR. MICHELSON: A little ahead,' if anything, 3
!' .6 right, but we won't think about that too hard. ]
7- I'd like to reconvene at 1:15.. :
i 8 (Whereupon, 12:20 p.m., the proceedings were-
[
9 adjourned until 1:15 p.m..this same day.)
10 1 11 12 13 14-15 16 17 18 19 20 21 22 23 l 24 l t
25 i Heritage Reporting Corporation (202) 628-4888 2 1
f 63
([) I hEIEEHQGH EEEAIQH 2 (1 :15 P .M. )
3 IE. MICHELSON: I believe we're ready to reconvene 4 the meeting.
5 And do you want to pick up the ney.t part of your 6 program? l 7 MR. KING: Yes. We're going to have a 8 presentation now by Ken Hart of NRR to talk about Section 9 C(4) of the Reg Guide which deals with the description of i
-10 the maintenance activities. NRR was heavily involved in l 11 putting that together for us.
22 And then following t. hat Meni will finish up, I 13 describe Section 5 and 6 of the Rag Guide and then we'll get I
f'h J
14 into the regulatory analysis and justification.
15 MR. MICHELSON: Do you have a copy of the 16 regulatory analysis for us?
17 MR. KING: No. That's what's being worked on.
18 MR. MICHELSON: It's being worked on.
19 MR. KING: We're going to talk about the status 20 and some of the qualitative arguments that--
21 MR. MICHELSON: Okay. We will get a copy of that 22 next Wednesday.
23 MR. KING: That's next Wednesday.
24 MR. MICHELSON: Okay.
25 MR. HART: My name is Ken Hart. I'm with the
() Heritage Reporting Corporation (202) 628-4888 1
I
84
(~')
s-1 Division of Licensee Performance and Quality Evaluation, the 2 Office of Nuclear Reactor Regulation.
3 I'm going to be discussing Section C(4) of the l l
I 4 Draft Regulatory Guide Conduct of Maintenance.
5 The proper conduct of maintenance is an essential ,
6 element of an effective maintenance process. I think Moni 7 gave us a good overview of what the process was that we were 8 talking about and really provided a good lead in to the 9 conduct of maintenance.
10 The program for the conduct of maintenance should 11 require and document application of these activities in a 12 manner consistent with the complexity and importance to 1
13 safety of the maintenance task to be accomplished. And 14 consistent with achieving the maintenance goals and
(}
15 objectives.
16 It is the responsibility of each licensee to 17 determine the degree to which he chooses to apply the 18 following activities in the conduct of various maintenance 19 tasks commensurate with complexity of the task, importance 20 of the equipment to safety and the established program goals l 21 and objectives.
22 DR. KERR: I'm sorry. I guess 1 don't understand 23 that last statement. "It is the responsibility of- " That 24 sort of says everybody is on his own and--
25 MR. HART: It's the responsibility of the licensee
() Heritage Reporting Corporation (202) 628-4888 3
1
~
l 85
() 1 to determine the application of these elements to the 2 conduct of maintenance.
3 DR. KERR: Okay. I'll try again next time. ,
l 4 MR. HART: The next slide--
5 MR. KING: Let me say one thing about that.
6 Section 4 was not an attempt to describe all the things that 7 have to be done on each piece of hardware that's covered by 8 the Maintenance Program. It's really more of a general 9 description of those things that we think contribute to 10 doing good maintenance, but a licensee will figure out 11 exactly how to put his program together and what elements 12 that are described in Section 4 are going to be part of his i 13 program and the degree to which they are a part of his 14 program, and which components they apply to.
()
15 MR. HART: Okay. The next slide is simply an 16 outline of Section C(4) and I'll be covering each of these 17 sections of the outline in more detail.
18 MR. MICHELSON: Can I ask a general question?
19 I noticed 3' the preamble portion of Section 4 the 20 same terms that I've noticed elsewhere and that is you are 21 introducing the idea of importance to safety and security.
22 I don't recollect that I've seen that "and security" added 23 in the past.
24 You are now introducing it here for the first j 25 time, is that the idea?
1
) Heritage Reporting Corporation (202) 628-4888 l
86
{) 1 MR. DEY: I didn't understand what you meant.
2 Past? Past regulations or-3 MR. MICHELSON: I'm talking about Section 4 of the 4 Rule now. Pardon me. Of the Regulatory Guide.
5 MR. DEY: Oh. I think it was an intent to include I
6 that throughout. 1 s
7 MR. MICHELSON: But in the past you haven't 8 included the security aspects--
9 MR. KING: In the proposed rule back in November 10 that word was not in there. That's right.
11 MR. MICHELSON: Yes.
12 MR. KING: But you'll see it show up in a bunch of 13 places--
14 MR. MICHELSON: Apparently you mean to include all
()
15 the same good things about items important to safety also 16 apply to items important to security. And my thought was 17 that I thought this was all covered by the rules that have 18 been written in past or the Reg Guides dealing with 19 security, equipment and so forth. And I thought maintenance 20 was a part of that. But I didn't have time to go back and 21 research the subject. ,
\
22 In other words, is this a duplication of something 23 you've already prescribed or is this something new that's 24 being added to the litany of things for the first time.
25 MR. MIZUONO: I'm Gary Mizuono from OGC, Heritage Reporting Corporation (202) 628-4888 l
l l i
7 'l 87 1 Rulemaking and Fuel Cycle Division. And it is true that the 2' Regulations, Part 73 in part'icular, have specific 3 requirements for maintenance and testing of security.
4 However, if you look at the requirements, they are not 5 inclusive of all the activities that arc within the bounds 6 of the maintenance rule.
7 And I believe it was the intention of the staff 8 that in addition to the specific requirements for 9 maintenance and testing that are contained in Part 73.and 1 1
10 then the Reg Guides that implement that, that the remalcing 11 activities should also be applied. And, again, according to 12 the graded approach. l a
13 Things with a high importance to security would be l O 14 determined by the licenso. _, quire certain activities, so V l 15 probably a very broad scope of the activities that are i l
16 mentioned, specifically listed within the Rule and the Reg 17 Guide. And things which are only indirectly related to 18 security or perhaps where there are backups. You know, the #
19 graded approach would allow them to apply less of thoso 20 activities.
I 21 MR. MICHELSON: Well, that ideazof the graded (
22 approach didn't come through on the security aspect quite 4 23 clearly, but certainly I hope it's graded.
24 But what bothers me is that apparently we have q 25 reason to believe that in the caso of the security aspects I
Heritage Reporting Corporation (202) 628-4888 !
~.- _
g
~
l l
i
'/ 88 7
l
[ 1 that'the maintenance history, et cetera,-has been such thAt 2 we believe there's a problem there and we ne.ed to do more. i 3 Or why was it'even included, since it's already covered by a 4- different part of thr> Regulation?
5 Well, why did we think we had to add the j 1
6 maintenance of security-related items to this? I can !
7 understand in a case .of the important-to-scfety ttoms 8 because we've got a lot of history uncertain of those items
, . 9 that maintenance has been a problem. Is there a comparable 20 history on the escurity side? I'm not aware.of it, but f'
11 maybe there hr.s been.
12 MR. CARROLL: It's been an ongoing thing between 13 the regions and the power plants. At least a lot of power
/ 14 plants. The regions feel like they are doing a good job of 15 maintaining them.
16 MR. MICHELSON: The security equipment.
17 MR. MIZUONO: Cameras, TV cameras, and the 18 lighting and whatever--
19 MR. MICHELSON: I haven't seen it highlighted 20 particularly, but it may be that down at the regional level 21 it's appreciated. I didn't appreciate there was any.real 22 problem and therefore I was surprised to see the Rule 23 encompassing something I thought had already been taken care 1
24 of to some extent. But apparently not to an adequate 26 extent.
Heritage Reporting Corporation (202) 628-4888
~
i j
J
, ?
I
+89' j) 1 MR. .DEY:- Well,.zif you' read. the first paragraph of 2' lthe Regulatory Guide:in-the Introduction,'it' does state that' s
'9" '3 .this regulation would also codify previously. existing <
]
4 Commission requirements, either explicit or implicit. So-5 the intent is here in this Regulation to take the f 6 regulations that are there, either explicit or implicit,Jand. l 7 include th0m. That is the intent. !
8 MR.-MICHELSON: -Yes. I ' sensed it was clearly your j 9 'i ntent. The statement was quite. clear that it covers.
10 ' security. . It wasn't clear to me why and that d.idn't clarify 11 thtt any, . Why'did this come under<the maintenance. umbrella
'1:2 that you are trying to generate now? It must be because 13 you've had enough; trouble with it, or some good reason to 14 include it.
15 DR. KERR: Well, perhaps it's,.like the reactor- o 16 situation ~in which there is not'a serious problem now but
. 17 one'might develop in tb.s future.
18 MR. MICHELSON: Yes. -Of course, that speculation 19 applies to most anything we do. There's always'the chance ,
1 I
20 it will go snur.
21 MR. MIZUONO: May I try to respond to that?
22 >
I believe part of the rationale including--and 23 specifically calling out the security systems is because the i 24 Commission's scope of authority in the Atomic Energy Act ;
25 covers not only safety but also the common defense and L
O erit e- R gereine cerebration (202) 628-4888
( 1- security _-type concerns.
2 And although there is some debate legally as to 3 whether the security concerns fall within the health and 4' safety _aut.hority grant _within the Atomic Energy'Act or the i
(
5 Common Safety and Security, it was felt that it would be .l 1
6 best to call it out specifically as something that we are' -
7 concerned about.
8 And, again, I wanted to stress--I think Tom I
9 mentioned it before, that the maintenance rule is being i 10 promulgated on the basis of the authority of the Commission 11 to set forth standards for the operation of the plant. And 12 it doesn't necessarily represent the fact that we have a
! 13 current problem with security. It just simply represents l
\
14 the fact that we have authority there.
15 We believe that it's an important part of carrying
.16 ' out our authority under the Atomic Energy.Act, and therefore 17 we believe that maintenance standards should be set up for
( 18 the security aspects of the plant as well as those important 19 to health and safety.
20 DR. KERR1 I don't think you understood the point 21 that Mr. Michelson was making, either that or I didn't. I 22 thought his point was that requirements for maintenance have 23 already been formulated and exist in Part 73.
l 24 Why do you need additional requirements in this l
25 rule?
\
Heritage Reporting Corporation (202) 628-4888
, l
91 .
(')
1 HR. MIZUONO: I think I tried to respond to that !
2 in my initial ccmuents which was that in looking at the 1 3 requirements in Part 73 and the Regulatory Guide, the )
1 4 specific requirements for maintenance and testing there were 5 not sufficiently broad. Did not cover all aspects of the l 6 maintenance activities and the maintenance process that is 7 being called out in the Regulatory Guide and in the rule. l 8 MR. MICHELLON: There is some reason to believe 9 that it should I guess?
10 MR. MIZUONO: Yes. Cecause we believe that l 1
11 security is very important.
1 12 HR. MICHELSON: Oh, I don't doubt that. But has 13 experience then indicated that we have a deficiency that l 14 needs to be corrected? That the present process isn't
()
15 working? y l 16 MR. MIZUONO: I would put it in terms of saying if 17 you didn't have these activities in terms of maintenance of 4
18 security systems and the health and safety SSC's, that you
! 19 could have a problem develop in the future. In fact, there 1
l l 20 might be problems now. We're not making a determination q l
21 that a problem exists now. That's not the basis for l 22 promulgating the rule.
23 DR. KERR: I see this argument used at other 24 places in this. And it's as if the Commission isn't sure 25 that a problem exists, but we'd better set up a rule just in
[ \
( ,)
'~
Heritage Reporting Corporation (202) 628-4888 I
l l
, 1
l 92 l
()m 1 case it might.
2 Now, that may be great logic, but it's not very 3 good use of resources. It's almost as if all the problems 4 have been solved that we know about and now we're going to l
5 set up rules to solve some of the problems that we aren't !
l 6 sure exist. And we have plenty of problems on which we l i
7 ought to be putting in resources. So surely we aren't at )
8 the point where we just set up rules--I can't believe that, f 9 You don't mean that surely.
10 MR. MIZUONO: No, I think what the staff is--and 11 what the Commission, as far as we can understand, is saying 12 is that if you don't do these things, you will have a 13 problem.
("') 14 In one sense you can say we have already R.,
15 determined that a problem exists because if in the sense 16 that we do not have a regulatory basis to timely act and 17 take action if we don't--if licensees don't pursue these 18 maintenance activities.
19 I think we were trying to make a distinction 20 between the fa'ct that a licensee might not have a current 21 problem. T mean, a current problem with maint6 nance. But 22 that doesn't necessarily metn that we don't feel that there 23 is no problem at all, j 24 DR. KERR: I am simply saying that I didn't know 25 that we could afford the luxury of solving problems that n
(_) Beritage Reporting Corporation (202) 628-4888
1 l
I 93 i
() 1 don't exist when we still have a good many problems that do 2 exist on which we ought to be working. )
i 3 MR. DEY: I think what Gary is saying that, you j 4 know, the reason and philosophy beyond regulations, having a i
5 regulation, is not solely based on having a problem. It's 6 based on-- J 7 DR. KERR: I agree.
8 MR. DEY: --what the Commission feels it should 9 have in its regulations to avoid problems. So that safety 10 and security is achieved.
i 11 DR. KERR: Yes. But that assumes that one can set 12 up a regulation that will avoid the problems and that is a 13 part of the logic that seems to me is not very well 14 developed in this presentation.
(~J) m.
15 MR. DEY: Well, we were going to come to that in 16 the last session of the afternoon.
17 DR. KERR: And it's going to be markedly different 18 from what I have read up to now?
19 MR. DEY: I think it's going to be different from 20 what you've read, yes.
21 OR. KERR: I shall look forward to it.
22 MR. CARROLL: One minor editorial point. In 23 reading the Introduction to the Reg Guide the amendment of 24 the regulations, a couple of purposes. One is to codify 25 blah, blah, blah. And it does not mention security anywhere
) Heritage Reporting Corporation l (202) 628-4888
l 94 j 1
1- in Part.73_or'anywhere else.
)
2' MR. DEY: -Okay.
3 MR. CARROLL: This would probably be a good place i
4 to tell people _about_this. l 5- MR. DEY: Yes. That's a good point.
6 MR. HART: The conduct of maintenance includes a )
7 clearly defined maintenance organization with specific lines )
8 of authority, responsibility and accountability.
9 Effective management.of maintenance' requires 10 effective written and oral communication between the l i
11- Maintenance Department _and other supporting groups.
12 Communications within the department and between the 13 Maintenance Department and plant and corporate management 14 are also essential-to an effective program.
[}
15 Criteria--
i 16 MR. CARROLL: Do you know how to measure that?
17 MR. HART: No. And I don't think we're trying to 18 measure that more than we're trying to say it ought to be 19 incorporated.
20 MR. CARROLL: Okay. Now, how do I incorporate it?
21 One way to do it is to have some very formal procedures that 22 says "If A happens, Joe talks to Charlie," and that kind of 23 stuff. - "And Charlie talks to Pete."
24 MR. HART: And it may be as simple as an 25 organizational chart,
() Heritage Reporting Corporation (202) 628-4888 l'
95 (m) 1 MR. CARROLL: It may be as simple as an 2 organization chart and my st'atement to the inspector that 3 " Hey, we talk to each other, but it's on an ad hoc basis.
4 There's a reason. And we do have a staff meeting once a l 5 week."
6 Is that good enough? How is this Resident !
7 Inspector going to judge this particular off issue?
8 MR. HART: I think the way to try to judge 9 something like that is to try and come up with a problem 10 that you can trace back to a lack of communication as 11 opposed to specifically spelling out that, you know, these 12 two guys need to talk every other day or something like 13 that.
I 14 MR. KING: Yes. That gets back to the purpose of f'/)
x_
15 this whole session. It's not to tell people how to 16 structure their program. It's to more be used as a guide if 17 there is a problem in maintenance, to go back and say, okay, 18 these are the things you ought to be looking at to see if 19 you can find a root cause of the problem.
l 20 If it's communication, you know, what could make 21 it better. And figure out what he could do to make it a l 22 better.
! 23 MR. CARROLL: Following your point up though, you .
24 could find a problem out there where a root cause analysis ;
1 25 shows it was poor communication, and you could say, okay, f
Heritage Reporting Corporation i
(202) 628-4888 l
1 i
L-___ - _--
l 96
() 1 I'm going to cite you for a violation of the commitment you 2 made in 4.12 of the Reg Guide.
3 And the licensee turns around and he says, " Wait a 4 4!nute. You know, I know ten other plants in this region 5 that do exactly like I do. What's your real basis for 4
6 citing me. I have a program. You just found it's not )
)
7 perfect. Nobody's is. You just found one instance where I 8 messed up. But I'm not doing anything any different than 9 the plant next door."
10 MR. HART: Does the plant next door have the same l 11 problem? l l
12 MR. CAEROLL: "And it doesn't have a problem. And 13 you've never cited them."
14 MR. HART: Obviously then their communications may
()
15 well be effective and in different cases and different 16 levels maybe judgment being effective.
17 MR. CARROLL: It's really coming down to inspector 18 judgment. ;
19 MR. HART: And I think in almost every case you 20 are going to involve inspector judgment.
21 MR. KING: But if you remember Moni's side on how 22 we would determine acceptability, compliance with this Rule 23 and Reg Guide. It wasn't whether they have a problem. It's 24 whether this process is trying to find out why they are 25 having a problem and fix it. So if they had a problem and
() Heritage Reporting Corporation (202) 628-4888 l
1 l
I I
1
1 , 97
()
f 1 it turnedLout to be communication, I think as long as they L
2 are taking steps to try and correct that so it doesn't 3 happen again is.what we'd be looking.for. Not.the fact that 4 they had a problem, 5 MR. HART: Criteria and procedures for the 6 selection of. personnel with acceptable qualifications to 7' perform the designated assignments are necessary for 8 effective staffing.. Resource allocation should include 9 adequate staffing of support organizations to provide for 10 expected contingencies. l 11 DR. KERR: Tell me something. I almost hesitate 12 to ask the question, but the stuff you are saying is so 13 obvious that it seems to me that anybody who didn't 14 . understand this would be completely incapable of setting up l
(}
15 a maintenance program even if he had this stuff. What's it 16 for?
17 MR. HART: Put it in writing. And I agree with 18 you. I think it is obvious.
19 DR. KERR: But I mean--
20 MR. HART: And I think a lot of the utilities are 21 already doing it.
22 MR. DEY: I think the intent here is to provide in 23 one document the basic guidance and if there are problems j 24 out there that this provides the basis for making an 25 evaluation.
(/ Heritage Reporting Corporation (202) 628-4888
4 98
) 1 It may very welllbe that a lot of these things'are
~
^
2 very practical and already exist in all.the plants, but what 3- we'kliketodo-isputitinoneregulatoryguidealongwith 4 this process so that it's a template the person could use'to 5 make these evaluations. That these are the type of 6 activities that are important--
7 DR. KERR: But I get the impression this is being 8 written for a bright fourth grader. Not somebody who is i
9 responsible'for.a maintenance program.
.10 MR. KING: Well, it is very general. I think 11 you'll' find it parallels probably pretty close to the'INPO 12 Guidelines but the INFO Guidelines have a lot more detail in 13 each of these areas. We intentionally tried to make it 14 general and state the principles in the broad concepts but
()
15 not get into the details of how to do things.
16 MR. ZWOLISKI: Once again, these are attributes, 17 the seventeen attributes, that we feel comprise the 18 satisfactory maintenance program. It parallels or 19- correlates one to one with the INPO Guidelines.
20 DR. KERR: Would you identify yourself for the 21 Reporter?
22 MR. ZWOLISKI: I'm John Zwoliski.
23 MR. CARROLL: Well, I think this is though where 24 the difficulty comes into th>s. My question this morning
'25 about Licensee 1, Licensee 2 and Licensee 3. Each of those I
Heritage Reporting Corporation (202) 628-4888 ;
1
._-_- _ _ _ _ - _ _ _ _ _ _ = _ _ _ _ . . _ _ _ _ .
-4 99.
. 1 guys can say, " Hey, I meet this." And with Licensee 1 you 2 probably--you might'want to take a look and say,' yeah, I've 3 confirmed that what he says is true.
4 With Licensee 2, you've got kind of a "never-5 never" problem. If your goal is really to make everybody as 6' good as Licensee 1, you'11' chase Licensee 2.. You'll cite .j 7 him for not meeting this and that. I don't know what you do 8 with Licensee 3.- I guess you use this to say, no, you don't 9 really meet this. We interpret these words to mean this.
10 And he says, no, I do this, I do this, I-do this. That's 11 what those words mean.
12 I'm not sure we've helped the regulatory process a 13 whole lot by'having this document. I'm not sure what it 14 does for me.
15 MR. FOLEY: Sir, I'm Tom Foley. I'm an ex-Senior l 16 Resident Inspector at Yankee Rowe, Indian Point and Calvert 17 Cliffs. For ten years practically I've been a Senior 18 Resident Inspector.
19 We do exactly what you are saying with all our 20 rules and regulations with each of the various utilities, 1, 21 2 and 3. That is, with some of them we can clearly say that j 22 you don't meet this and it's quite obvious.
1 23 With others, it's a matter of interpretation and 24 it's our interpretation that counts in the long run and we--
25 MR. CARROLL: That's what I'm worried about.
Heritage Reporting Corporation (202) 628-4888
/
100
( ); 'l- Because I know a whole spectrum of Resident Inspectors.
And
- 2~ some are good-and some'are very, very bad. And, you know--
3 MR. FOLEY:- That's a matter that we're going to-
-4' have toD11ve with in interpreting our regulations. ]
5 MR. CARROLL: Well, I just hoped that there was j 6L some,better answer than to leave'it up to the Resident's ,
7 judgment.
8 MR. DEY: There's two ways to look at it though.
9 One is to prov1de for flexibility in the regulation and put-10 the objective of being non prescriptive. And that's a-good 11 . objective. But that does create room for interpretation.
12 So it's.a balance one has to achieve.
13 MR. CARROLL: But where I'm coming from is that,
()'14 you know, hypothetically a Resident Inspector can be a real 15- detriment to safety. If in fact he keeps the plant
.16 organization so stirred up on trivial issues that they don't 17 mind the store.
?l8 MR. KING: He.could do that on almost any issue 19 under the existing regulations currently.
20 MR .- CARROLL: I agree with what you are saying. l l
21 MR. KING: But then again-- l 22 MR. CARROLL: But on the other hand, he could be a .l 23 very positive benefit towards nuclear safety. 1 1
24 MR. KING: That's true. !
25 MR. CARROLL: And I guess what's bothering me is
( Heritage Reporting Corporation (202) 628-4888
l 101
~
we're putting one more piece of paper out for Resident
()s. 1
~
2 Inspectors to deal with and--
)
1 3 MR. KING: And it's an extremely wide-ranging one. I 4 MR. CARROLL: And it's an extremely wide-ranging ]
j 1
5 one.
6 MR. FOLEY: Hopefully our regional management and j l
7 our headquarters management will keep an eye on the Resident 8 Inspectors and evaluate them constantly to insure that they 9 are implementing the rules and requirements the way the l l
10 headquarters and the regions want them to, as an agency, 11 from an agency perspective. That's their job and that's 12 what Residents typically do. If they are not performing 13 that then they shouldn't be Resident Inspectors and they 14 ought to be out of there.
()
15 MR. MICHELSON: Is there going to be~some kind of l
16 an. inspection module written around this Regulatory Guide i 17 that gives the Inspector further guidance on what to look 18 for, what to expect? And how to interpret or whatever? l 19 MR. KING: As we talked about it this morning, t
f 20 there already is an inspection module for maintenance. When l l
21 the Rule and Reg Guide are finalized, it would seem to me 22 there may have to be some expansions in certain areas in i
23 there to make sure everything that's talked about in the Reg 24 Guide is covered. ;
25 MR. MICHELSON: That would have to be revised I (
)
()
i Heritage Reporting Corporation (202) 628-4888
i l
102
() l' 'think and written more around the Regulatory Guide for 2 maintenance. j 3 MR. FOLEY: In the back of the I&E Manual 1
.4 currently there's a section called Interpretations and l 5 whenever there are several questions regarding any 6 particular rule or regulation, the agency does come out with 7 an interpretation of that or we come out with policies also, !
8 a position, regarding a particular statement.
9 MR. MICHELSON: Well, if this document is viewed
'10 to be an extensive expansion of the regulatory inspection 11 process then I would think you have to revise the inspection 12 module and then start adding the special cases as they come 13 up as further clarification.
MR. KING: I don't think it is an expansion.
() 14 ~ I 15 think the' current inspection module for maintenance tracks 16 very closely with the stuff we are talking about.
17 MR. MICHELSON: Oh, you mean we already require 18 goals on components--
19 MR. KING: Not the goals. I'm talking about the 20 Section C(4) that we're talking about now that gets into the 21 various attributes--
22 MR. MICHELSON: Oh, I was thinking of the total 23 document, not necessarily specific parts, some of which I 24 hope are already being done. But as a whole, will this 25 document prompt the revision, a new inspection module or Heritage Reporting Corporation (202) 628-4888
103 p) q 1 revision of the present one?
2 MR. ZWOLISKI: As mentioned earlier, the document 3 parallels INPO guidelines. It actually evolves--the Reg 4 Guide actually evolves from the Maintenance Team Inspection 5 Guidelines that currently exist.
6 It would be my view that upon codification of a 7 rule, the staff would reevaluate this particular Team I
8 Inspection Module in light of the rule being on the books.
9 MR. MICHELSON: Does the Resident Inspector use 10 that Team Module or does he have a separate one? I thought 11 he had his own manual.
12 MR. ZWOLISKI: This Team Inspection Module is 13 geared for a six-man team--
I'i 14 MR. MICHELSON: Yes. But it's the Resident
(_/
15 Inspector that's going to enforce this one, isn't it? There 16 may be an occasional--I didn't sense there's going to be a 17 routine--
18 MR. KING: As I said this morning, I'm not sure 19 we've worked out exactly who is going to inspect it when.
20 It's something to be determined yet.
21 MR. MICHELSON: But clearly the Resident Inspector 22 has to have guidance about this document if it is issued.
23 MR. ZWOLISKI: Let's add one more point. The 24 current inspection modules that e.ist that are utilized by 25 Resident Inspectors target maintenance activities.
O C/ Heritage Reporting Corporation (202) 628-4888
i 1
)
104-l
() 1- The Maintenance Team Inspection that we've been 2 talking about was an enhance' ment of those products. So 3 Resident Inspectors currently have guidance to work with.
'4 MR. MICHELSON: Well, of course, I guess I'll have l
~
5 to say that if they've already got all the guidance that's 6 in here, you don't need this document. I assume this is l 7 something new, an addition, an expansion, whatever, isn't 8 it? Or is this already fully covered by everything the 9 Inspectors are already inspecting against, and if it is, why 10 is it.being issued?
11 MR. FOLEY: This goes beyond what we typically 12 have.
13 MR. MICHELSON: I'm sure this has got a lot more 14 nuances than he's used to seeing.
(}
15 MR. FOLEY: As Resident Inspectors we have l 16 approximately three maintenance modules we call them or 17 inspection procedures to go by. We typically do a--we do a 18 monthly inspection, a monthly maintenance inspection, that 19 the Residents commonly do. And then there's another module i.
l 20 that Specialist Inspectors from the region come out and they 1
21 look at the Maintenance Program on a yearly basis and they 22 also have a Special Maintenance Program that they look at.
23 This covers some of those aspects but this goes 24 into it quite a bit more.
25 MR. MICHELSON: But I assume those modules have to l
Heritage Reporting Corporation (202) 628-4888
105
- ( '1. .be revised, if'this is a significant departure from what 2 we've:been doing.- And if it isn't, then 'I ask why we're 3 oven in this business if it's just a minor' departure.
~
4 MR.. CARROLL: How much different do you view it?.
5 MR. FOLEY: _ In my view, it gives us another tool 6 to look at more of the balance of plants. There's so many.
7 ' things out there that cause transients--cause reactor plant 8 transient, and we have no--
9 MR. CARROLL: Let's exclude balance of plant. The 10 . process' involved.
-11 MR. FOLEY: Well, the maintenance modules that we 12 have now, they don't require as much management involvement. ,
13 They don't require the communications, the stsff. None of 14 that's specified in the maintenance modules that we have
(}
15 now.
16 Some utilities that I've been inspecting for ten 17 years or so, they have maintenance organizations and because 18 they are so small, so close knit, their procedures are very, 19 very weak. It's not specified. These things are not called 20 out or specified in their procedures. They do have them to 21 some extent. But there's nothing required about them.
22 And, of course, in some of the larger utilities, 23 the more sophisticated util tes, these things are all well 24 laid out, spelled out.
25 DR. KERR: In your view, even though a system is Heritage Reporting Corporation (202) 628-4888 r
L________._____._
o
- il 106 working well, if they don't have requirements,.it's not a
( ). 1 2 good system? Is that--
3 MR. FOLEY: That's not necessarily true, no, cir.
4 No, sir. In my view, if.it's working and it continues to 5 exist--
6 DR. KERR: I thought you were being critical of 7 these small groups that didn't have written requirements--
8 MR. FOLEY: No, sir.
l l
9 .DR. KERR: --just because they--
10 MR. FOLEY: They have better programs than the 11 ones that are all laid out.
12 DR. KERR: Oh, okay. All right.
13 MR. MICHELSON: If the program's in the head, 14 which it could be if it's a small operation, and I think
_ ()
15 that's sort of what yvu are suggesting--it's in the process, 16 it's working well, but it just isn't on paper, would it pass-17 the muster of this new regulatory guide?
18 MR. FOLEY: The problem with those small utilities 19 where it's not on paper, is that there's nothing to assure 20 that it will continue to exist with management changes, 21 personnel turnover, things like that.
22 MR. MICHELSON: So I guess what we're saying is .
23 that even though it's working well, you noV must document it 24 formally?
25 MR. KING: There is some base of documentation Heritage Reporting Corporation (202) 628-4888 i
.i
.\
1 107
'( ) 1 required by this.
2 MR. MICHELSON: Oh', there's quite a bit. I mean J
3 you start talking about creating goals and so forth, I ]
i 4 assume you write them down and-don't keep them in your head. '
5 MR. KING: You write down your program, what it is j 6 you're doing.
7 MR.'MICHELSON: You write down your program. You 8 don't keep it in your head. A number of these objectives 9 and so forth I think have to be spelled out if you expect 10 them to be. accomplished.
11 So no matter how well the program is working, l 12 there.will be a documentation job associated for any utility 13 that is now not already documented.
14 MR. ZWOLISKI: There's a great deal of sensitivity
(}
15 on the staff that this will not be an enhanced burden as far 16 as additional paperwork for licensees that are performing 2
17 well. We're probably more driven to licensees that are not 18 performing well'in this area of maintenance.
19 Some of your facilities have a minimum amount of 20 paper, skill of the craft is outstanding. They get very 21 high marks. We don't see this activity perturbating that 22 system.
23 MR. MICHELSON: You don't think they--
24 MR. CARROLL: You are not going to make them write 25 it down?
. Heritage Reporting Corporation (202) 628-4888 .
108 1 MR. ZWOLISKI: Essentially every licensee has a (v) 2 maintenance program in which selected goals and objectives 3 have been identified. We're not asking licensees to change 4 a program that's working fine already. We are targeting 5 this at the licensees who are your poorer performers.
6 MR. CARROLL: Okay. Now, I've got two comparable 7 utilities, one next to the other. A is how you've been 8 describing. He does a good job but there's not a lot of 9 formalization and book cases full of policy guide statements 10 and whatever.
11 B is another small utility and management hasn't 12 given him the resources and he's doing a lousy job on 13 maintenance. A lot of safety system challenges and all j 14 kinds of problems. You are going to make him write up or f*e) x-15 put together a very formalized program but hold him to it.
16 I can understand why you might want to do that, i 17 but it seems a little inconsistent that you haven't made A 18 do the same thing. Why don't they both need the same level 1 19 of documentation.
20 MR. ZWOLISKI: Your second licensee is going to 21 have increased regulatory oversight in which one area would l l
22 be increased attention to maintenance. And the development
\
1 l 23 of their program and how they implement their program. )
l 24 Utility A was implementing their program l 25 satisfactorily so we didn't do anything with it, in your l
(,,) Heritage Reporting Corporation (202) 628-4888
1 i 109 J("T
.v-1 example. 'Over time we would expect Utility A, if their' 2 program was to fall back, to have problems and not maintain 3 the standards, we would expect Utility A to begin enhancing 4 their program.
5 MR. CARROLL: When all is said and done with these 6 two cases, I go in and-I look at the paper, assuming both of 7 them, or that A stays okay and B gets enhanced. B is going e to have a 1oe bigger book she1f of paper, right2. ;
9 MR. ZWOLISKI: That's possible, but now you've 10 driven our. inspection to be paper reviews when in fact 80 to ;
- 11. 85 percent of our time is performance-based inspect!ons.
12 MR. CARROLL: I'm not worrying about your time.
13 I'm worrying about the time of the utility and the time taken away from the maintenance people for the real world of
( ) .14 15 activities and--
16 MR. ZWOLISKI: The bottom line is to have a 17 licensee implement some number of these seventeen elements 18 appropriately. That's the goal of the regulator. Minimize 19 challenges to the plant and how the licensee wants to 20 implement any number of these is on their shoulders. With 21 goals and objectives to be established by each individual 22 licensee.
23 MR. FOLEY: If a utility is not performing well in 24 maintenance, and it doesn't have a well-defined 25 organization, and doesn't have the paper, and then we have
() Heritage Reporting Corporation (202) 628-4888 l
i 110
() 1 another utility that is performing well, whether it has the 2 paper or not doecn't matter. It's this utility that's not 3 performing well we need to help them get organized and to 4 comply with the organization.
5 And it's not who's got the most paper or, you 6 know, how much time they are spending on the paper. It's 7 performing well. How do we get them to perform well--
8 MR. MICHELSON: I think you are introducing a >
9 concept here now of graded performance that I simply never j i
10 read out of thl3 Regulatory Guide. I read the concept of 11 graded importance to safety and then clearly you adjust your 12 Maintenance Program according to the importance of the 13 components and whatever that came through I think was (v~'; 14 reasonable. But nowhere did I ever get the feeling that a 15 good utility with a fine Maintenance Program did not have to 16 address each and every one of these elements and do what it 17 required. I never got that feeling.
18 MR. CARROLL: I think he's saying they are 19 addressing them.
20 MR. MICHELSON: Well, I think they--
21 MR. CARROLL: There is the amount of 22 documentation--
23 MR. MICHELSON: Well, I think the documentation 24 goes with it. I don't think that you are excused from 25 documentation because you've got a good SALP rating.
4
/N
(,) Heritage Reporting Corporation (202) 628-4888
i I
l 111 MR. FOLEY: You are not really excused from it,
'l' 2 but when we have utilities that are typically performing 3 very well in all areas and we look at'the end results, they 4 typically don't get the same degree of inspection as one 1
L 5 th'at's performing poorly.
16 MR. MICHELSON: I know that, but I don't 7- understand why--what in this document says that'you are L8 going to relieve certain licensees of detailed documentation
-9 because they've got good performance. It's not in here- is 10 it in-here?
11 MR. YOLEY: No.
12 MR. MICHELSON: I didn't sense it at'all.
13 MR. FOLEY: That's not what I'm saying.
i f~)
\_/
14 MR. MICHELSON: Well, I that's what I thought you 15 just got done saying.
16 MR. FOLEY: I'm trying not to convey that. The 17 degree--everything is done in degrees.
18 MR. MICHELSON: We're talking about documentation i
19 now.
20 MR. FOLEY: Documentation in degrees.
21 MR. MICHELSON: Okay, o
22 MR. FOLEY: If they have a well-performing 23 . organization, you are certainly not going to go and 24 scrutinize their organization and tell them that they need 25 more in this area or they don't have enough in that area.
Heritage Reporting Corporation (202) 628-4888 L
t L _ - - _ _ - _.___ _ - __ - _____-.-________ - =___._.______.--.___--______- _ _ _ .
)
112
() 1 MR. MICHELSON: But you expect that they're 2 already doing it.
3 MR. FOLEY: You would expect that they have a 4 certain level--
5 MR. MICHELSON: Some minimum level at least and I 6 thought this was proscribing that level somehow of 7 documentation and that everybody was going to document.
8 There's only--that wasn't the question I thought at all.
9 Everyone must document every one of these seventeen items.
'10 MR. KING: No, no. I think you are reading more 11 into this than--
12 MR. MICHELSON: Where does it say that you don't 13 document all of this? Where is the proscription-- !
(' } 14 MR. KING: Page 7 of the Draft Reg Guide, at the 15 very bottom of the page, the last six or seven lines.
16 DR. KERR: You know, you just made a statement 17 that seems to me is extremely important and I'd like to 18 follow it a bit. And you said it was up to the Commission 19 to get them to performing well. And'that's an interesting 20 philosophy of the regulation.
21 An alternative and one which I would have thought 22 the Commission might be using was to determine whether an 23 organization is operating a plant well, and if they aren't 24 operating it well enough, shut it down.
25 But I did not realize that the Commission had the Heritage Reporting Corporation (202) 628-4888
113
() 1 responsibility of seeing that they got it back up to 2 operating well.
]
d 3 MR. FOLEY: An acceptable level I meant by "well." J 4 DR. KERR: Well, I mean your comments seem to 5 indicate you felt it was the responsibility of people who 6 weren't performing well to tell them how or to get them up 7 to where they are performing well.
8 MR. ZWOLISKI: And we do that by our SALP process.
9 DR. KERR: I'm simply saying that an alternative 11 0 approach would be for the Commission to judge whether things i l
11 are performing well or not, and if they are not, shut them 12 down.
13 But you don't necessarily have the responsibility
/~T 14 of telling people how to run their plants so that they will V
15 be running them well.
16 MR. ZWOLISKI: I understand your point. And it's 17 correct. I will counsel that if in a SALP process give a 18 licensee a Category 3 in a particular area I have put him on 1
l 19 notice that compared to his peers he's ranked lower and 20 should pay additional attention to a particular area.
21 DR. KERR: I would like for us to spend a day or 22 so on the SALP process. I have serious misgivings about 23 that process, but that's another day.
24 MR. ZWOLISKI: We' d be happy to come down and talk 25 to you.
/3 Heritage Reporting Corporation
'V (202) 628-4888 l
l 114
() 1 DR.-KERR: Because I think SALP'is a step towards 2 the URC managing plants. And while I'm not sure . that 3 utilities can manage' nuclear plants, I'm very sure that the 4 NRC can.
5 MR. ZWOLISKI: I agree.
'6 MR. CARROLL: Getting back to what you were 7 saying, maybe I can say it this way. You are saying that if 8 somebody's program is weak, the fact that you are getting an 9 additional attention, it's probably naturally a result of 10 producing more documentation.
11 MR. FOLEY: That's our experience in track record 12 to date. Yes, sir. .
13 MR. CARROLL: You don't find that inconsistent to 14 the guy next door who has and continues to maintain a good
(}
15 program--
16 MR. FOLEY: Many times we have troubles
.17 reconciling how one plant with a limited number of staff 18 seems to be able to outperform plants with huge staffs. And 19 we go around in circles on that particular point. Very 20 difficult to answer.
21 MR. MICHELSON: I've read the bottom of page 7 22 which I think is where you referenced where this point would 23 be covered. I don't read the bottom of page 7 apparently 1
24 the way you do.
25 It does indeed say there that some activities
) Heritage Reporting Corporation (202) 628-4888
115
() l' require little if any documentation, where other will
~2 require documentation consis' tent.with 10CFR 50, Appendix B..
3 And, of course, I thought that statement related only'to the 4 importance to safety question. Clearly systems not very 5 important to safety get little or no documentation. Those 6 very important to safety get full documentation. I thought 7 that's what it meant.
8 I didn't . read into this that people who are doing 9 a good job don't need to document. People who are doing a i
10 poor job have to document more. I didn't read it that way i 11 at all.
12 Now, maybe the words need to be changed. If 13 indeed you mean that people doing well don't have to
(~3 14 document. As long as they show they are doing'well, why
%/
15 don't we say it, if that's really what you intend.
16 MR. KING: I think we'd expect people who are 17 doing well to have some level of documentation that is 18 helping them do well.
19 MR. MICHELSON: It may not be--
20 MR. KING: It may not be to the degree that other 21 people have and it's not our intent to come in and have 22 everybody have some minimum number of notebooks on the s
23 shelf.
24 MR. MICHELSON: Well, I thought you just explained 25 to me that some people do very well with little or no O
, (_/ Heritage Reporting Corporation (202) 628-4888 A
J
116 l
- 1 1 documentation, but they got the right kind of supervision, j Iv) 2 'the right kind of crews. It's a small compact operation.
3 It's very successful without much documentation.
4 But that message didn't come through to me on this 5 statement, which if that's what you're citing as the basis, 6 I don't read it the same.
7 MR. ZWOLISKI: You should be aware, there is a ;
8 large volum6: of material on the shelves relatable to 9 maintenance of all these systems and components. Vendor 10 specs, vendor manuals, quite elaborate.
11 MR. MICHELSON: Yes, I'm aware of it. <
i 12 MR. ZWOLISKI: --are implementing procedures that 13 we find at the facilities very dramatically. Some
(] 14 procedures have every step, cross every T, dot every I.
v 15 Other procedures are very global, use vendor inspection l 16 manual, or vendor manual, whatever, No. X, X, X.
17 Now, licensees typically have their own culture, 18 own climate en the level of procedure and the amount that is i 19 in a procedure so you immediately have a gradient in paper.
l 20 Some licensees tnat rely on the skill of the 2.1 craft, have demonstrated over time, seemingly are using less 1
22 paper than those that don't have the same skill of the 23 craft.
f
> 24 We've identified that, by the way, in several of l
l 20 our maintenance inspections.
I O nerieeee aegerties corgemeeien (202) 628-4888 l
w - _ _ _ _ _ - _
1 1 117 l l
1- MR. MICHELSON: I don't think there's any doubt of
({)-
' ' 1 2 that, but I guess what.you are saying is that degree to l
3 which this: Regulatory Guide will be implemented will be j 4 performance based. q i
5 MR. ZWOLISKI: Yes, sir. )
d 6 MR. MICHELSON: .That message never came through to 7 me in reading anything so far. And although I don't think 8 it's a bad idea, but that idea never'came through. So could 9 you show me3where you-say'that, that that's your intention 10 to use a' performance-based enforcement of this Regulatory .
- 11. Guide and Rule?
12 'Now, if it is, I don't take issue with
- 13. performance-based, but I do wonder what the guy who's doing
'14 well"does with this paper now if he's'doing it with a small 15 ' operation that's very efficient and very. understanding--it's-16 great.
17 MR. CARROLL: I take exception if it's performance 18 based until somebody. tells;me how it's going to be 19 implemented.
20 MR. MICHELSON: Yes. It's the other extreme, of 21 course. 1 22 MR. ZWOLISKI: I've been referred to page 5, 23 paragraph B.
24 MR. MICHELSON: Of what? !
25 MR. ZWOLISKI: Conduct of Maintenance.
1 Heritage Reporting Corporation (202) 628-4888
1 i
l l
118 i 1 MR. MICEELSON: Of the Reg Guide?
[^J) 2 MR. ZWOLISKI: The last sentence in Conduct of 3 Maintenance.
4 MR. MICHELSON: Okay. That's not a bad sentence.
5 I think that's--so what you are saying then is, indeed, if ,
1 6 you can show me you're doing a good maintenance job with I 7 minimal documentation, that's acceptable?
8 MR. ZWOLISKI: Yes. "Should also be aware within
}
9 the context of NRR's Maintenance Team Inspections, those are 10 performance-based inspections." )
'l(
11 MR. MICHELSON: Uh-huh.
)
12 MR. ZWOLISKI: Those six-man teams, i
13 MR. MICHELSON: But if you're not doing so well in f (3 14 performance, then we'll use these seventeen elements to l wJ 15 judge where you need to shape up or whatever? Is that sort 16 of how you use this? t 17 MR. KING: We would expect the licenaee to use 18 that assessment feedback process to judge his own where his 19 own weaknesses are do something about them. And, yes, we 20 can come in and overview that anytime we choose.
21 MR. ZWOLISKI: Independently if we found 22 conflicting results or differing views, that's when we would 23 sit down with the licensee and compare our results. That
) 24 obviously there's a problem in the feedback mechanism, in 25 the way they scrt their goalc and objectives. j l
l 1
<~.
(,) Heritage Reporting Corporation (202) 628-4888 l
_ _ _ _ _ . . }
i
,. l 1 -
119 l l
() 1 gm. DEY; And what's being presented in Section 4 2 is a template for.this assessment and. evaluation. l l
3 DR. KERR: So you will have or will develop l
')
4 criteria which will permit y9u to determine that a )
3 5 maintenance program is or is not satisfactory?
.6 MR. ZWOLISKI: The Maintenance Team Inspection l
? that's currently underway uses a TI which does have criteria j q
8 in it. .And'that being a performance-based inspection, as I J s
9 said earlier, will be revisited in light of the rule and the i 10 full' process. I don't know how much it will change, if at 11 all. I think it-does need to be changed in selected areuse 12 Tom mentioned earlier our goale and feedback 13 mechanism--much of this Reg Guide evolvos from the l 14 inspection tree that we developed to perform these
(}
15 maintenance inspections. Therefore I don't believe there's 16' very much disconnect between the two documents, But it is a 17 living document and we are aware of that and we' re certainly 18 aware of your point. It's also been made internally.
1 19 MR. HART: I think that brings us to Qualification 20 and Training. Program 50--
21 MR. CARROLL: We can apend our time better than 22 having you read what's already in here, i 23 MR. HART: Well, is there a better way to discuss !
- 24. the points?
25' MR. SMITH: Have all of us read this? f
() Heritage Reporting Corporation (202) 628-4888 1
l l- 1 L'L _ _- _- - - _ - _ _ _ _ b
..e
)Y '
120
'kj) 1 MR. MICHELSON:
Yes, but of course when you read 2 it and then you reread it, it comes out differently 3 'sometimes. Sometimes better, sometimes worse. Refreshing V
4 the memory doesn't hurt.
5 MR. .ZWOLISKI: Now, it's certainly not our intent S te read a viewgraph to you. However, in browsing through if 7 questions or comments exist within the context of our 8 slides, we could use a Q and A format, if you would prefor?
- 9 DR. KERR: Let me Q 4.2 Is INPO accreditation now 10 required of all maintenance programs?
11 MR. HART: No, it's not going to be required.
12 It's stated that the INPO accreditation is an acceptable !
13 method of meeting the training and qualification
'N 14 requirements. It's not necessarily meant to be the only way i
- (G 15 a licensee can do it.
16 DR. KERR: That's tantamount to requiring it, so 17 that answers my question.
18 MR. CARROLL: Well, you are answering from a 19 Commission perspective. From a utility perspective, being a 20 member of INPO does require you to commit that you are going 21 to meet their accreditation requirements.
l 22 MR. ZWOLISKI: You'll be accredited for
- l. 23 maintenance training, yes.
24 MR. CARROLL: So this is kind of stating what 25 already exists here.
. i
/~T Heritage Reporting Corporation j
(,/
(202) 628-4888 !
I i
m-_______ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
121
( .
1 MR. MICHELSON: Uh-huh.
2 MR..ZWOLISKI: Excuse me. This is a regulatory 3 guicke, not a requirement. The Rule--
4 DR. KERR: Look, we have been over this regulatory
.5 guide business over and over again. You know and I know 6 that in 99 percent of the cases the utility will decide that 7 the easiest way to do things is to commit to the reg guide.
8 Once he does that, for all practical purposes it becomes a 9 rule And that's what will happen. I don't know if that's 10 good or bad, but that's.certainly a fact.
11 MR. ZWOLISKI: Recognizing the industry's l 12 perception of our regulatory guides, we tried.very.hard to 13 parallel existing materials that were in open literature for 14 licensees. We recognized the relationship between the 15 Commission and INPO and what kind of documents are easily 16 referenceable and are not.
17 That's a document that I cannot append to a guide, 18 for example. If the industry was to come to the Commission 19 with something like that as a guide for the industry, 20 certify using a third party, I think we would find that 21 approach acceptable.
22 MR. SMITH: Outside of a rule?
I 23 MR. KING: No, within the rule. Within the 24 context of.that new option to the rule.
25 MR. SMITH: You may have to step down what you Heritage Reporting Corporation (202) 628-4888
122 f
f1 'j ust.said., It isfpart of the. rule that you are asking the 2 industry to belly up to the Reg Guide. It is enforced by 3 the Rule which is the INPO guideline which means then that 4 you have made INPO a part of the regulatory process, and 5 that's our major concern.
6 16R. KING: I understand that's your concern.
7 That's not our intent.
8 MR. SMITH: .It's not a problem of intent. It's 9 the facts. ?t's the fact that you put facts in that guide 10 and attach it to the Rule. It is, it is. It's mot a 11 question of what you intended.
12 MR. ZWOLISKI: Not to belabor the point, the 13 Regulatory Guide will stand on.its own merits, with or 14 without the INPO Guidelines.
(~}
15 MR. HALL: Excuse me. My name is Warren Hall and 16 I'm with NUMARC. Based on your-statement, how can you put 17 that into accreditation as part of the title?
18 MR. CARROLL:~ We're simply saying that that's one 19 way of accomplishing this.
20 MR. ZWOLISKI: It's an example.
21 MR. HALL: But the accreditation process changes, 22 End the NRC doesn't particularly cars for it.
23 MR. MIZUONO: And the Reg Guide is revised to take 24 out that reference to the INPO Guidelines and as a person 25 that's being involved in the continuing involvement of O
y Heritage Reporting Corporation (202) 628-4888
123
() 1 various Reg Guides which reference differ Unt industry 2 standards, that's done all the time.. The fact that it might 3 represent an INPO standard as opposed to a quote " industry 4 standard," I know represents a concern for INPO.
5 But the process of endorsing something that is out 6 there and the withdrawing that support because it's evolved 7 in a direction that the NRC is not willing to endorse 8 happens all the tim 6:.
9 MR. CARROLL: One general comment I had about this l 10 whole section and having been beat over the head when I was 11 helping write AeS standards some years ago is the old 12 "should" and "shall" argument. This thing ira just loaded 13 with "shoulds" and "shalls."
14 MR. DEY: There aren't any "shalls."
(' }
15 MR. CARROLL: What?
16 MR. DEY: There aren't any "shalls." l 17 MR. CARROLL: Okay, okay. So "should" really l
18 means "should" as defined in an English dictionary?
19 MR. FOLEY: In a regulatory guide, it's not a 20 requirements. It's guidance. Which the licensee--with the 21 option of following it on that.
22 MR. CARROLL: I.can remember some reg guides that 23 were produced a few years ago. The only thing that they 24 said was everyplace in the industry standard where it said 25 "should," read "shall."
() Heritage Reporting Corporation (202) 628-4888
(
124 IT is, 1 M.R. FOLEY: Yes, that's when you endorse 2 standards..
3 MR. MICHELSON: But I guess a Reg Guide becomes a 4 "shall" at such time as you commit to following it.
5 MR. FOLEY: Yes.
6 MR. MICHELSON: Then the "shoulds" become "shalls" 7 by your commitment. But not until.
8 MR. FOLEY: The utilities that I've dealt with, 9- and when they commit to Reg' Guides or industry standards, 10 they typically commit to them with many, many exceptions in 11 their Quality Assurance Manual. We will commit to the 12 Maintenance Rule Reg Guide with the exception of such and 13 such, and such and such. And all the things that they don't 14 want to do are simply inconvenient for them to do. And I'm
(}
15 serious. Very just inconvenient and for no other reason.
16 And when you put down "should," they hold you to 17 that. They say, "We don't have to do that. That's a 18 'should.'" Or, you know, "We don't have to do that." See, 19 "Here it says 'shall," and we'll do that." And that's only 20 when we hold their feet to the fire.
21 They take them very seriously and I think wo do 22 too, the "shoulds" and "shalls."
23 MR. MICHELSON: But the "shoulds" do become 24 "shalls" at such time as you agree to follow the Reg Guide 25 in total.
/'s
() Heritage Reporting Corporation (202) 628-4888 1
-125 I' I l'- MR.-FOLEY: Yes, sir.
2 MR. MICHELSON: Now, if you take an exception and 3 .those exceptions still become "shoulds" and not "shalls."
4 MR. FOLEY: Now, I've seen this someplace. When
-5 .they do .cosmnit to it, if it says that, if it simply states 15 that-the "shoulds" will become "shalls," then it does.
7 But if it doesn't say that, I don't know of 8 . anyplace where it requires that automatically it, turns a 9 "should" intoLa "shall." I haven't seen that anyplace. Has 10 anybody else?
11 MR. MICHELSON: If you agree to do something then 12 I think it becomes automatic.
13 MR. FOLEY: Yes.
' 14 MR. MICHELSON: Only if you agree to do.it though.
15' HR. FOLEY: Oh, it should.
16 MR.-MICHELSON: Okay. Why don't we proceed.
17 Unless there are other questions, let's proceed.
18 MR. ZWOLISKI: For the Subcommittee's indulgence, 19 I've asked Warren Puesko to join me at the table. Warren is 20- also a past Senior Resident Inspector, having spent quite a 21 bit of time at Susquehanna in Region I.
-22 The reason I've asked a couple of Senior Resident 23 Inspectors to join me at the table has been to bring a more 24' of the field experience to the document and they have been 25 principal contributors.
Heritage Reporting Corporation (202) 628-4888
1 l
126
()- 1 Looking at the section from the standpoint of 2 implementation, as an Inspector, how would I march through a 3- plant with this kind of document?
4 And our Maintenance Team Inspection criteria and 5 guidelines.
6 MR. CARROLL: I don't know whether they were here 7 this morning, but I'm sorry they were not.
8 My first question this morning and I said I hoped 9 I got an answer sometime today, I was talking'about three 10 licensees. 1, 2, and 3. And one of my questions was if 11 they came in and said, "Okay, I've got a good maintenance 12, program. It certainly meets the adequate protection I
13 requirement. It's not perfect. I make mistakes once in a
/~ 14 while, but I'm improving all the time, b) 15 The reason I think it's good is that I've done a 16 maintenance self-evaluation and although I've found a few 17 little things, they tended to be things I could fix," and 18 maybe they fixed them. "NUMARC's come in and looked at me--
19 MR. MICHELSON: You mean INPO?
20 MR. CARROLL: I mean INPO. "I don't get too many 21 negative findings. Therefore I certify, based on the above, 22 that I meet 65 and comply with the Reg Guide."
23 If I said that, what would you do? How would you 24 deal with 1? Would you believe that story?
25 MR. FOLEY: As a Resident Inspector you are there Heritage Reporting Corporation (202) 628-4888
1 1
127
() 1 ten hours a day, sometimes six or seven days a week. At 2 least five days a week. You live with the people. You see 3 their maintenance programs. Often times, if you are there 4 for a number of years, you have more corporate memory than i
5 the Plant Manager or the Maintenance Superintendent and j 6 you've seen the history of maintenance.
7 And it's very difficult to buffalo a veteran 8 Resident Inspector who has been there for a long time. You 9 can tell whether they are just handing you a line or not.
10 That's for either 1, 2 or 3. They're just not going to pull 11 a line like that on you. If they do you can tell--
12 MR. CARROLL: Okay. Let's take 1--
13 MR. FOLEY: If it's a utility that truly is a 1 14 and they are doing a superlative job in maintenance, I look.
(}
15 I mean we've got other things to look at. We look at 16 problems. We are out there to protect the health and safety 17 of the public. To look for plant problems, things that are 18 causing transients. We are out there to look at safety--
19 there are one or two Resident Inspectors on site and two 20 thousand people or one thousand. It depends on construction 21 or operation type.
22 If there's no problems in that area, if our 23 specialists have come out and looked at it, if we've done 24 Maintenance Team Inspections, if INPO has come in there, if 25 everybody else is coming in and saying they are great, why Heritage Reporting Corporation (202) 628-4888 I
l 128 l) 1 should I spend my time disagreeing with them. I've got too 2 many other things to do.
3 MR. CARROLL: So you are not going to argue with 4 somebody over the meaning of the words compared to what he 1
5 has in a manual on a shelf. This is the 1 guy we're talking 6 about.
7 MR. FOLEY: That's right. If the 1 guy is truly 8 performing that way, I probably won't have time to read the 9 maintenance rule. Much less go out there and try to enforce 10 it, as a Senior Resident.
11 MR. CARROLL: Okay.
12 MR. FOLEY: You focus your efforts where you 13 perceive or where you have indications that there are j 14 problems.
()
15 MR. CARROLL: Okay. Now, let's go to the 3 guy.
16 I think I know what you are going to do there too.
17 MR. FOLEY: Darn right. I'm going to take that i
18 rule and I'm going to make him adhere to the letter of the l 19 law.
20 MR. CARROLL: Are you going to try to make him as 1 21 good as 1.
22 MR. FOLEY: No. I want to bring him up to an 23 acceptable level. And then I'm going to look for other 24 placee that are below acceptable levels.
25 MR. CARROLL: Well, 3 is an acceptable level
() Heritage Reporting (202) 628-4888 Corporation
129
) 1 according to all the SALP reports that I have read.
2 MR. FOLEY:' Well--
3 MR. CARROLL: In the sense that they are not 4 violating. regulations, the plant is safe.
5 MR. FOLEY: I'm going to emphasize that-- I 6 MR. ZWOLISKI: It's a minimally acceptable. level 7 and we do.believe there's' room for-improvement.
8 DR. ICERR: Look, come on. There isn't anything 9 called minimally acceptable. Something is either acceptable 10 or it's not. acceptable.
11 MR. FOLEY: That's easy to say.
12 DR. KERR: That's because it's true. That's easy.
13- to say too.
14 MR. FOLEY: But it's not that easy. When you are 15 out there inspecting, I mean--
16 DR. KERR: Look, it'may not be easy to tell when !
.17 something is acceptable or not. I mean something is either
- 18 acceptable or not acceptable.
19 MR. FOLEY: You mean there is no gray?
20 DR. KERR: I don't see how there can be.
21 MR. FOLEY: There.is. There's gray in everything, 22- in all the rules, in all the regulations. There are 23 interpretations. Your interpretation and my interpretation.
124 And especially when you are trying to go to power and I'm e
25- trying to protect the health and safety of the public.
Heritage Reporting Corporation (202) 628-4888
49 p
'I
, j l
130
.-1.
DR.' KERR: There is not any--
- 2. MR. FOLEY: .Which.means in this case to shut down.
3 DR. KERR: There's.not any gray between zero and
-4 1.
5 . MR . FOLEY: .It depends on how you look at'it.
6 DR. KERR: You can invent some if you want to.
7 But-- l 8 MR. ZWOLISKI: There would be a significant amount 9 of regulatory intervention in Category 3 and a much reduced 10 amount intervention in Category 1. The balance being at a 11 Category'2.
12 DR. KERR: Do you expect--
13 MR. ZWOLISKI: We allocate our inspection
.14 resources purely based on SALP processes.
(
15 >
.MR.. CARROLL: Okay. Now, what I'm worried about 16 with the guy who consistently--he gets to know the people 17 pretty well and the Team Inspectors and, you know, he gives 18 a sad story. They'd like to do better. If their Senior
- 19. Manager will give them more resources somebody will pay for 20 them. They'd really like to be a 1 plant.
21 You're not going to touch that, are you?
22' MR. FOLEY: It depends upon--
23 MR. CARROLL: They could be better. They know 24' they could.be better.
25 MR. ZWOLISKI: Well, we recognize that some Heritage Reporting Corporation (202) 628-4888 1
131
-'v ') 1 licensees are very happy or content with SAL:?--they've told 2 us they don't aspire to all Category 1. Or maybe can't 3 afford it.
4 MR. CARROLL: Or maybe can't afford it.
5 MR. ZWOLISKI: We don't find anything wrong with 6 that.
7 MR. CARROLL: I read words in here--maybe I 8 haven't read them in the right context--that led me to 9 believe that you are trying to get everybody up to being an
~
10 excellent performer.
11 MR. KING: We have words in here that say as part 12 of setting their objectives and goals, they ought to be 13 shooting to be up with the Category 1 folks.
(
%j 14 MR. CARROLL: I don't think that's what these guys 15 are talking about. They would accept a guy who says, no, 16 I'm happy to be a 2.
17 MR. MICHELSON: I don't think you are trying to 18 make them l's. But you are trying to keep them, I would 19 gather, in the 2 Category.
20 MR. KING: The words in the Draft Reg Guide--
21 MR. MICHELSON: Now, which words are you referring 22 to that--
23 MR. KING: The ones under this. The section 24 called " Goals." It starts on page 6. Basically say in 25 setting their goals they ought to be shooting for fs l I) s_ Heritage Reporting Corporation (202) 628-4888 l
i
. 'I s .
132'
' performance. systems structures'and. components,. consistent I l' 2 with the top performers in t'he industry. It doesn't put a 3 numerical'value in. 3 4 DR. KERR: What you are saying is that your r 5 regulations are not adequate to provide a performance that 6 you want to set as a goal. I 7 MR. KING: No, you are back on adequate 8 protection. We're not on adequate protection anymore.
9 DR. KERR: I have not used the term " adequate 10 protection" at all. I am talking about your regulations 11 which' determine the way a plant should operate and certainly 12- your top people, your 1 people, are, according to the SALP 13 ratings, much better than is required by regulation. That's-14 why you give a SALP rating. 3 is still acceptable.
15 So what you are saying is that.you want plants to 16 be better than your regulations require them to be. Now, it 17 seems to me if that's the case, there's something wrong with 18 the regulations.
19 MR. KING: No. I think what we're saying is 20 there's room for improvement, cost effective improvement, 21 and this regulation--
22 DR. KERR: It's not a question--
23 MR. KING: --is a direction toward doing that.
24 MR. KING: The regulations are supposed to protect 25 the public health and safety. They say nothing about Heritage Reporting Corporation (202) 628-4888
l 133 whether there's room for-improvement or not.
..h . 1 --
2 MR. KING: A: lot.o'f regulations these days are 3 based upon enhanced safety.- . Generic issues'primarily are l
4 based upon are they cost effective to do to enhance the ,l 5 safety of the plant. They are not' based upon adequate 6 protection. q 7 DR. KERR: But when you require them, they become 8 part of a regulation..
9 MR. KING: Yes. And then a plan is required to 10 meet them.- And that--
11 DR. KERR: But now you are using a non-regulation 12 to require--I mean you are saying that we need to set'a goal 13 which is beyond the requirement of even this regulation. So 14 you are requiring something that the regulation doesn't 15 require.
16 MR. KING: The regulation requires assessment, 17 feedback, corrective action process. The Reg Guide is 18 expanding upon what that means.
19 DR. KERR: Well, I guess I don't know what it 20 means either because do you ever try to stop encouraging 21 improvement? And you know, pushing people to improve?
22 Because, you see, if you do, you are saying that the 23 existing level of activity is not adequate. Now, I'm not 24 talking about adequate protection in the legal sense in 25 which the NRC uses the term.
Heritage Reporting Corporation (202) 628-4888
.. . - _ _ _ _ _ _ _ - 1
134
() 1 I'm saying it's not good enough to satisfy a 2 regulatory agency. I'm not against excellence. But I 3 don't--it seems to me it is not the responsibility of a 4 regulatory agency to try to push people beyond the 5 regulations that exist to require--or are supposed to be 6 adiquate to protect public health and safety 7 I mean it's fun to do that. It's exciting to it, i
8 It's a challenge to do it. Some people will do it. But it 9 doesn't seem to me that that's the responsibility of a 10 regulatory agency.
11 MR. KING: Most of the regulations that have been 12 put out recently, particularly in the Generic Issue area, 13 have been based upon doing better. Not adequate protection, 14 but doing better.
(}
15 MR. ZWOLISKI: Over time our standards have 16 changed.
17 DR. KERR: Of course. And that should be 18 expected.
19 MR. ZWOLISKI: And there has been an increase.
20 DR. KERR: And then when they change, you change 21 the regulations.
22 MR. ZWOLISKI: Our expectations also fall. We 23 could have put in capacity factor or radiation exposure or 24 typical--
25 DR. KERR: No. Your--
() Heritage Reporting Corporation (202) 628-4888
135 l 1
1 MR. ZWOLISKI: Typical goals that licensees shows-(( ) '
4 2 DR. KERR: No. I'm not talking to you about that. i 3 I'm talking about a situation in which you have a level so 4 that presumably if you go below that level you are 5 unacceptable. Above it, presumably you are acceptable.
6 But there seems to be, certainly in the SALP 7 ratings, and here, an effort to push people on beyond that.
8 And to me that is not the responsibility of a regulatory 9 agency.
10 MR. FOLEY: The way I see it, from what I 11 underscand, you are saying that if they don't meet the 12 regulation they are unacceptable. If they are unacceptable, 13 they perhaps ought to be shut down, we'll say.
14 And you say there is no gray matter, it's black or
[v~}
15 white.
16 DR. KERR: I'm saying it ma:r be difficult to reach 1 17 a decision that something is acceptable or unacceptable.
18 But once you reach that decision there's no gray matter.
19 Either they are or they are not.
20 MR. FOLEY: But in regards to regulations--in 21 regards to regulations, there are many, many regulations.
22 If a licensee fails to meet one of those regulations, is he 23 acceptable or unacceptable?
24 DR. KERR: That depends on how you treat the total 25 of all regulations.
() Heritage Reporting Corporation (202) 628-4888
i I
136
() 1 MR. FOLEY: That's right. And that's exactly what 2 we're trying to do. This is my impression of what we're 3 try ng to do here, is to get the maintenance--
4 DR. KERR: But you see, you are not_ insisting that' 5 they--it seems to me you are going beyond insisting that 6 they meet each regulation. You are insisting that they go 7 beyond that. Not just meet it.
t 8 MR. FOLEY: In all areas licensees fail to meet 9 regulations periodically.
10 DR. KERR: But I'm not talking about failure to 1
11 meet. I'm talking about trying to push them beyond meeting i 12 them.
13 MR. FOLEY: We are trying to establish a 14 conservative margin. If they are unacceptable, plants
{}
15- typically get shut down. If they are acceptable, barely 16 minimally acceptable, they just meet the regulations, but 17 periodically falter. We are saying that, hey, if you 18 continue to falter, you are going to be shut down. You are 19 not going to be a Category 3.
20 So what we say is--we encourage a conservative 21 margin such that they don't or that they only very rarely 22 fail to meet the regulations. And that's what we are 23 pushing for. We are pushing for a licensee to meet all of 24 our regulations all of the time. That's the Category 1 25 performer who never fails to meet our regulations. Who L
() Heritage Reporting Corporation (202) 628-4888 i
d 1
137 l surpasses our regulations. And I think that's what we're
. () 1 2 pushing for.
3 DR. KERR: I do too.
4 MR. FOLEY: And I think that's what the goal is.
5 DR. KERR: And I think that's inappropriate. I 6 think you should only insist that operators meet your 7 regulations. If meeting your regulations is not enough to 8 ensure the public health and safety, the regulations ought 9 to be changed.
10 MR. CARROLL: And my question, and I guess I got 11 the answer, a 2 guy is -- a 1. Even though you are doing a 12 2 better.
13 MR. MICHELSON: I thought that's the answer I 14 heard and that bothers me because the statement you cited on
(}
15 page 7, in that paragraph at the top, the last sentence, 16 seems to indicate that you really want to go beyond that, 17 unless you think that--the latter part of the sentence says, 18 well, the sentence says, "In general, goals for SSC's should 19 be established to achieve a level of performance consistent 20 with that. achieved by those U.S. plants with rocognized 21 effective maintenance programs."
22 My first question is, are those meant to be the 23 Category 1 plants or are those also Category 2 plants? Are 24 Category 2. plants recognized as having effective maintenance 25 programs?
l i -
() Heritage Reporting Corporation (202) 628-4888 l
138 MR. ZWOLISKI: Recognized--
( ). 1 2 MR. MICHELSON: In the context of this sentence.. '
3 MR. ZWOLISKI: We feel that much of the industry 4 have adopted effective maintenance programs. The programs 5 are instruction in all the elements. Implementation of--
6 MR. MICHELSON: Well, effective is the word, not 7 very effective.
8 MR. ZWOLISKI: Effective.
9 MR. MICHELSON: Okay.
10 MR. ZWOLISKI: Implementation of that program 11 which embodies all the elements, is now routine for graded--
12 and the Level 2 performer or Level 1 performer, going back 13 to SALP, has earned his mark based on how well he has 14 performed over the grading period.
~}
15 MR. MICHELSON: Well, I think it would be around 16 it. Let me ask you a specific question. Does a Level 2 17 performer have a program with a recognized effective 18 Maintenance Program? Is that a Level 2? Or is that only a 19 Level 17 Or what did you mean?
20 MR. ZWOLISKI: Once again, it depends on how 21 effective he was in imple;. tenting--
22 MR. MICHELSON: I know there are degrees of 23 effectiveness. But I'm just trying to ask you what is the 24 meaning there on page 77 25 MR. ZWOLISKI: I can tell you what I meant. I l
) Heritage Reporting Corporation (202) 628-4888
139 4
() 1 wrote that, sentence. In general what I meant was a Level 1 2 performer.
3 MR. MICHELSON: Well, I thought that's what you 4 meant.
5 MR. ZWOLISKI: That's what I had in mind when I 6 wrote that. It doesn't mean there's no flexibility there, 7 but, you know,.that's generally what I had in mind.
8 MR. CARROLL: These guys seem to be saying that--
9 MR. MICHELSON: Level 2 is not good enough by your 10 definition then.
11 MR. CARROLL: They seem to say Level 2 would be 12 good enough.
13 MR. MIZUONO: I think you have to make a 14 distinction.
{}
15 MR. MICHELSON: They are saying Level 2 is good 16 enough.
17 MR. MIZUONO: I think you have to make a f 18 distinction between the performance that's achieved and what 19 is acceptable versus the goals that you have. And I think 1 20 what the sentence with regards to goals means, even though 21 that you might be satisfied with achieving a performance, 22 which is equivalent to a Level 2, that your goals should be 23 Level 1.
24 That might present a logical inconsistency and I 25 think the staff will have to address that.
=
j .
() Heritage Reporting Corporation (202) 628-4888
__ mm._____. ____m _
140
() 1 Thank you for pointing that out to us. But I 2- 'think in the course--and you've drawn our attention to that, 3 but-I believe that the intent when we wrote that was that 4 .the objective has to be a Level 1. And I think we're still 5 unsure as to how we actually evaluate them in terms of 6' performance.
7 It's clear though that as long they are--what we 8 are trying to do is focus in on the performance. That's the l
9 mein objective'.
10 MR. MICHELSON: Is there any reason--
11 MR. CARROLL: What's wrong with these guys is it's 12 been out in the real world said there are other utilities 13 out there that are perfectly happy with the SALP ratings of 14 1 or 2 and maybe as individuals they would really like to
(~))
15- have the resources to do a better job but the reality is 16 they're not going to get them.
17 MR. MICHELSON: But the Regulatory Guide says you 18 must set goals equal'to 1 perhaps. Depending on how you 19 interpret what was meant here. I don't think you meant that 20 you have to set a goal equal to 1, did you? Or is that what 21 you meant?
22 MR. ZWOLISKI: That was the intent. Our intent in 23 developing this Guide was--
24 MR. MICHELSON: I'm talking about this sentence 25 now only. On page 7.
() Heritage Reporting Corporation (202) 628-4888 i
I
141'
^
--was'to have an impact on. industry'
()[ l' MR.-ZWOLISKI:
1 2 to aspire to a-level of performance superior to what I find
- 3. in SALP--
4 MR. MICHELSON:. Does that mean Level 1 is what you 5' are striving for?
6 MR. ZWOLISKI: Superior.
7 MR. MICHELSON: To what? Level 1 even?
8 MR. ZWOLISKI: Reliance on preventative and 9 predicted maintenance, trending, doing maintenance--rather 10 than a fixed mentality that has permeated the industry these 11 many, many. years--
12- MR. MICHELSON: Excuse me. If you want to have a 13 conversation, please, I can't hear him and hear you right 14 behind him. Thank you.
[}
15 MR. ZWOLISKI: --that'run, break, fixed .nentality 16 is what we are trying to target by stating in this manner or 17 a like manner, shoot for a level of excellence,~ people 18 understand the SALP Category 1. language.
19 Now, what should be the Agency's goal in this 20 area? We try to regulate with a minimum threshold. And 21 we've been mixing words here back and forth.
22 It was our intent to try to hit a baseline that no -!
23 matter who would read this, the minimum cut I would get )
24 would be an acceptable level of maintenance for the industry 25 and there would be noticeable payback to the individual 1 Heritage Reporting Corporation (202) 628-4888 t
i
142
) 1 licensees. Via enhanced maintenance.
2 And it would be seen hopefully in capacity 3 factors. It would be seen in reduction in radiation 4 exposure, reduction in trips, ESF actuations, what have you.
5 MR. MICHELSON: Well, I'm mainly struggling to j l
6 make sure I understand what we ware asking for.
7 MR. ZWOLISKI: I wanted to put on the tabla that ]
i" 8 we are shooting very high. We recognize there needs to be a c
9 compromise position to some extent.
10 MR. MICHELSON: But that targe.t that you are 11 shooting for is the equivalent to what you normally give 12 when you give a Level 1 rating.
13 MR. ZWOLISKI: - your better performers that have
's 14 in the past received a SALP--
(^J 15 MR. MICHELSON: Is there some reason why we can't 16 talk about Level 1 SALP in a Regulatory Guide? You know, 17 that term I understand a little better because then the l
18 utilities that get No. 1, say, gee, this is how I-- l 19 DR. KERR: That's a good idea too because then 20 everybody would be above average.
21 MR. MICHELSON: And this is the next problem.
22 Well, it was never--I don't think it's defined as average.
23 This is just talking about a goal. And your goal really is 24 Level 1 then. Is that right?
25 MR. KING: Yes, sir. >
Heritage Reporting Corporation (202) 628-4888
q l
i 143 :
1
,] 1 MR. MICHELSON: I didn't read it originally that 2 way. I guess people underst'and it even though I didn't read 3 it the same way. I thought the goal would be Level 2 which 4 I've always thought to be what I'd call " fully acceptable."
l 5 Not necessarily excellent. But fully acceptable. And that t 6 you must have at least a fully acceptable program. j) i 7 And' here you' re eaying, no.
8 MR. ZWOLISKI: In the context'of our inspection f I
9 program, that licensee would earn a SALP rating of 2.
10 MR. MICHELSON: Yeah.
11 MR. ZWOLISKI: It would ask him, are.you setting I
12 your goals and assessing yourself forthrightly? Are you 13 satisfied? As a regulator we' re satisfied. He's earned a' 14 grade of 2.
15 MR. MICHELSON: But that should not be his goal. 2 16 His goal should be a grade of 1. And he should set all of 1 17 his goals with that in mind.
18 MR. KING: That's what we had in mind.
19 MR. MICHELSON: But that message doesn't come 20 through here.
21 MR. KING: We'll check the words--
22 MR. MICHELSON: Maybc you can fix it up somewhere.
23 It only needs to be said once in a few sentences, but really 24 you are striving for moving people to a Level 1 SALP rating.
25 And then what does Level 1 mean when everybody gets up O erie ee Revereine cereeretion (202) 628-4888 r 1
1
o ; ,,
y di L, j .
144
'. 1 there? I cJuess it means only you no. longer need to look at
> 'A 2 it careful'ly from a regulatory viewpoint.
.'3 MR. KING: 'Aell, you still.want to look at it to .
4 'make sure there's no backsliding.
5 MR. MICHELSON: But you don't look at it in the j
- 6. depth--
7 MR. KING: But it's a term of trying.to state, 8 you know, what is the objective? How much is enough? Where 9 are we trying to get to?
10 MR ., MICHELSON: Well, I thought your goals were 11 going to be a little lower than Level 1 for this. And I ;
12 didn't realize it was that high.
13 MR. KING: We've put some high goals. Like John i
14 mentioned. We took'a line in this that was trying to move '
15 people up. !
16 MR. MICHELSON: But what do you say to the guy ,
i 17 who--I mean the Chairman of the Board of Podunk Light and 18 Power comes to Adm. faech and he says, "Look, I can't afford f
19 to be a 1 in maintenance. I just want to be adequate. I-20 don't have the resources. I don't have the money. How am I 21 going to run my utility company if I have to be excellent in 22 everything I do?" !
23 MR. KING: I think there's going to have to be 24 some flexibility to look at those situations and recognize l 25 that--
L Heritage Reporting Corporation (202) 628-4288 E_ _ _ _ _ - - - _ _ _ _ _ _ _ - _ _ _ - . - 4
4>u .
I 145
() 1 .MR. ZWOL'ISKI: SALP 2 is a fair score for the l
?2 individual to earn.
l 3 MR. MICRELSON: It's not a fair objective J 4 apparently. John King just told me it wasn't. )
i 5 MR. ZWOLISKI: We'd ask the industry to push 1 J
l 6' themselves. l l
7 .MR. MICHELSON: Yes..
8 MR. ZWOLTSKI: In seeking standards'of excellence, l
9 w0 would see that--
10 MR. MICHELSON: That's a very important measage, i
11 if that's what you want. It really ought.to be flagged in a 12 conspicuous and clear way.
13 MR. KING: A guy has a goal consistent with the 14 l's--
(}
15 MR. MICHELSON: Yes, and then the Commission 16- changes sides. }
17 MR. KING: Or for whatever .eason he can't get 18 there, resource c6nstraints or whatever. I think wo said 19 before that' compliance with this rule is not judged by 20 whether the guy is meeting his goal or not. It's judged by. i 21 is he taking reasonable measure to.try and achieve that 22 goal, to fix his problems, make improvements where his 23 assessments say improvements need to be made. Recognizing
, 24 that there is a graded approach. There's a cost benefit
(
25 tradeoff.
() Heritage Reporting Corporation (202) 628-4888 1
[
L-_--___-________ ._ . _ - _ . . _ _ _ _ _ _ _ _ _ _ _
146
( ); 1 We tried to get some words in to get that message 2- across. And I think those kinds of situations have to be 1 3 recognized and have to be dealt with.
. 4 Other than stating that in general terms, I don't 5 know what else to say in the Reg Guide.
1 6 MR. MICHELSON: Wellr you can use the word-- ]
7 NR. ZWOLISKI: As I mentioned earlier, over time '
8 performance of the industry looked at harder and harder,- and '
9 the SALP rating of a 1 in 1980 may be a grade 2 today.
10 ThereHis that creep that has taken place it our inspection ;
)
11 program in the process and there is some recognition of that 12 within the context of this statement. l j 13 MR. MICHELSON: That's perfectly all right. I ,
I
/~1 14' just was trying to find out what our goals are though, and
\s/
15 our goals are to achieve and maintain a Level 1 performance 16 however that's defined by the Agency with time. What I 17 think I'm hearing is your intent of the goal.
18- DR. KERR: Let me try to explore this from a l k
19 slightly different perspective. We have from the Commission 20 a Severe Accident Policy Statement which says that operating ?
21 planta are acceptably safe except that there might be some i 22 outliers and we're going to set up a program to look at l 23 those and it's now about to go into existence I guess as an 24 IPE program. ,
25 Now, it apparently will be the case th&t t
) Heritage Reporting Corporation (202) 628-4888 1
l l
- _ _ _ _ _ _ __ 1
.-, . w
. c 147
- r( f
~
g.
1 independently.cf what the IPE program indicates a-2 maintenance rule 11s.necessary in order to produce;a risk 3 that is deemed appropriate by the Commission. Even though
- ~
4 the' Commission is on record as having said that the plants, 5 are acceptably safe.
6 i. Will the Commission go back now and say we really l -
.7 shouldn't have set any Severe Accident Policy Statement that
'8 plants are acceptably safe? We have discovered that they 9 are not and they need a maintenance rule in order to be.
10 MR. KING: I think there's'something mixed up 11 here. j l
' 12 MR.' KING: I do too. .)
13 MR. KING: The Severe Accident Policy Statement 14 said plants that are out there today.are acceptably safe, (T}
15 but we believe there are some cost effective things that 16 could be done to make them safer. ;
1 17 DR. KERR: No, it did not say that. The Severe 18 Accident Policy Statement does not say that. That came 19 along later.
20 ~ MR. KING: No. The Severe Accident Policy i i
21 Statement has the words " cost effective" in it.
22 DR. KERR: Dut it says that one needs to look for 23 outliers and if it is cost effective to fix those, then they 24 should be fixed.
25 MR. KING: Yes.
() Heritage Reporting Corporation (202) 628-4808 i
.-_--.-a_-----------___--.----w--
a .~
]
j i
148
'1 DR. KERR: But it doesn't say that we.believe'that 'd
,there are cost effective things that can'be done to improve l
.2 '
.3- . safety. It talks about--
4 -- MR . KING: No,.it states it the way you stated it.
5- DR. KERR: Yes.
6 MR. KING: Look for outliers and if it's cost, 7 effective to fix them, fix them.
s 8 DR. KERR: That's right. Now, this is not--unlesa l t
9 maintenance is an outlier in all plants, and you tell me'it f 10 isn't, this is not a look for outliers, the maintenance I
- 11 rule, it's an add on. It may be necessary, but if it is, it 12 seems to me the Commission owes it to the public and to the 13 licensees to say, after having looked at.this we've decided 14 that really plants out thera aren't adequate.'.y safe. They 15 need a maintenance rule.
16 MR. KING: That's not what we're saying. I mean l 17 if we were taking an adequate protection route with this 18 rule, that's what we'd be saying, but we're not. We're 19 saying that we believe that maintenance can be enhanced-and 20 it can be enhanced in a cost-effective manner and--
21 MR. ZWOLISK r : If you read the Introduction--
22 DR. KERR: But the Commission has already said 23 that operating plants are safe enough.
24 MR. KING: But they continue to put out 25 regulations with generic issues, with IPE, with whatever it ,
4 O Remit ee Revereine cergeraeiem (202) 628-4888 1
- __ _ - _ __- __-_ _ -_- _ _ _ _ __ _ _ ._ _ h
l l
, , 1 149 !
'() - l' .may be, that use cost-effective basis to do more, to enhance v
2 safety. It's being done all the time. Here's another case
[.
3 where it's being done.
4 DR. KERR: I am not saying that because it's being )
1 5 illogical in this case, it can't be illogical in other 6 cases. I'm simply talking about the situation at hand.
7 MR. ZWOLISKI: I want to bring you back to the 8 thrust of this and it's caught right in the Introduction, 9 codified previous existing Commission requirement.
10 DR. KERR: But it's doing much more than that.
11 And we both know it. Previous commitments, for example, do 12 not include balance of plant.
13 Now, I'm in favor, by the way, of including some
,i'3 14 balance of plants, but it's not codifying that. That's a V
15 very significant and new requirement.
16 MR. ZWOLISKI: Second point. In current !
.17 regulations--
18 MR. KING: It's doing that, but it's doing it with 19 a cost-effective basis.
20 DR. KERR: We don't know whether it's cost 21 effective or not, Tom.
22 MR. KING: Well, understand, you haven't seen the 23 package yet and--
l 24 DR. KERR: Even after I see the package I won't l
25 know whether it's cost effective.
/
I
( Heritage Reporting Corporation (202) 628-4888 l
4 150
-1 MR. KING: You may disagree with the package.
([ }
2 DR. KERR: Nor wil'1 you. It's a case-by-case 3 application.
4 MR. ZWOLISKI:- We were trying to go through some 5 of the individual points contained in Section 4.
6 DR. KERR: But with a lot of1 interruptions. I'll 7 keep quiet for a while.
8 MR. MICHELSON: If anybody has any questions, 9 raise them.
10 -MR. CARROLL: We've all got a copy and have read 11 them. I don't really have any questions. The words are 12 fine as far as they go.
13 'MR. ZWOLISKI: Shall we continue with the next
/"T 14 item in the agenda?
V 15 MR. MICHELSON: Sure. Unless there are questions 16 yet on the items on those two or three pages.
17 DR. KERR: Just one question. What is " predictive 18 maintenance" in the sense in which it is used on paragraph 19 137 I see the word but--
20 MR. ZWOLISKI: Ken, do you want to field that?
21 DR. KERR: " Monitor trend, analyzed." Does it l t
22 mean that somehow by doing predictive maintenance you can 23 prevent any failures?
24 MR. HART: That predictive maintenance, in a sense 25 we use it in the Regulatory Guide--it's meant to monitor the
() Heritage Reporting Corporation (202) 628-4888 i
i
151 Lt l' performance of the operating characteristics of a component 2 such that;you can predict when a particular. component is 3 -going to fail such that.you can take action prior to an 4 actual failure.
5 DR. KERR:' Well, that is contrary to anything I-6 know about_ reliability' theory. Something that says you can 7 predict when a component is going to fail.
8 MR. DEY: Well, an example would be vibration 9 monitoring.
10 DR. KERR: That's not predictive maintenance.
l 11 That tells you when the thing is vibrating. Nothing 12 predictive about that.
13 NR. HART: No, but you are monitoring a condition 14 that will allow you to take action before you actually have 15 a failure. And that's the intent.
16- . MR . CARROLL: With some forms of vibration 17 monitoring, Bill, you can'get some real insights as to what 18 the heck is going on in the piece of machinery.
19 DR. IGEIUR: Well, you.could certainly get some 20 insight that it's not performing the way it should perform.
21 But can you predict when it's going to fail?
22 MR. DEY: No, not predict when it's going to fail, 23 Predict if preventive maintenance is required. i 24 DR. KERR: I just heard the statement I thought 25 that said you could predict when it's going to fail.
() Heritage Reporting Corporation (202) 628-4888
152
() 1 MR. DEY: In some. instances, yes.
2 MR. KING: I think the point is you have a better 3 fee $whenyouneedtogoin-anddomaintenanceinsteadof ~
4 waiting for the failure to occur. You've got some precursor 5 that says I'd better get in there and do something.
6 DR. KERR: But for most--many, let me say, of the 7 predictions you make, you have something like a bathtub 8 curve in the simplest case. You can make some predictions i
9 that failure is most likely to occur at the beginning and at 10 the wear-out stage, and you can predict the likelihood of 11 failure over an interval during this flat portion. But you 12 can't predict when failure is going to occur.
13 MR, FOLEY: That's pretty much correct. The term 14 ought to be revised a little bit. More like what you said.
(}
15 The likelihood of failure or something like that. But not 16 precisely when it's going to fail. Not to the moment, not I
17 to the day.
- 18. MR. MICHELSON: I think we are going to have to 19 proceed unless you can assure me that your last two items 20 are not going to take very long.
21 MR. KING: Okay. We'll move on to the next 22 session.
23 MR. MICHELSON: We'll take a break now. This will 24 be a good time to do it. Until three o' clock.
25 (Whereupon, there was a short recess.)
() Heritage Reporting Corporation (202) 628-4888
153 1 MR. MICHELSON: All right, let's get going.
-(])
2 You got one hour left, that's it and we're-going 3 to get 10 minutes out of that one hour.
4 MR. DEY: What I would like to do is, we have 5 described the first 3 stages of a process I mentioned 6 earlier, the Policy and Establishment of Objectives and 7 Goals Deriving Maintenance Activities--the conduct of 8 maintenance.
9 What I would like to go over is the.second half of 10 the process, the monitoring and assessment of the program 11 and feedback and corrective actions and the last, not being 12 the least important. Certainly feedback and corrective 13 action which is a significant part of the process.
/ 14 But I would like to do this fairly quickly because 15 we want to leave sufficient time for the last session and 16 discussion and have an opportunity for feedback on that.
17 So with that, Section 5 of the Regulatory Guide 18 deals with Performance Monitoring and Assessment.
19 Essentially what we are saying in that section is that as a 20 result of setting objectives and goals, one should use 21 indicators for tracking the performance of the maintenance 22 program towards these objectives and the licensee has the 23 flexibility, of course, of what parameters he chooses, but 24 he should include the definitions for these quantities so 25 that is defined as to what parameters are being tracked.
() Heritage Reporting (202) 628-4888 Corporation
(
c.
154:
.1 We looked at'three types of--
2 - MR . MICHELSON: . Excuse me. . Is that slide.in your 3 handout or am I just having trouble finding it.
4- MR. DEY: The packages are labeled IV and.VI and 5 it's Page 9.of;that. Roman Numeral IV and VI on the top 6 here and it's Page 9. It's the continuation of a previous
'7 package I'was using.
8 MR. MICHELSON: Okay, thank you.
9 MR. . CARROLL:- Given that you're dealing with the 10 four smartest guys on ACRS today, you ought to have these 11 things-better numbered when you come and deal with the. rest 12 of the members.
o 13 MR. DEY: Oh yes, we would like to have them 14 numbered serially, yes.
{}
I 15 We look at three types of indicators. We talked
, 16 about the hierarchy and having, you know, overall objectives L 17 _and then get into.the system level and the component level.
18 So we look-at three types of indicators: Plant level 19 indicators, process indicators, which indicate the maintenance program itself, items such as re-work, backlog, 21 ratio, preventive to corrective maintene.nce and.the last 22 item here is effectiveness indicators that are based on 23 component failure rates itself.
24 We encouraged the use of NPRDS and we talked at 25 length about that this morning. And the second part of
() Heritage Reporting Corporation (202) 628-4888 '
u 155
) 1 monitoring, we see that as being management over sight,and
'2 self assessment or, in other words, qualitative monitoring.
3 The management involvement sampling of maintenance 4 of maintenance activities, of their being done, looking at-1
~
5 .the trends and using that as input toward determining the 6 effectiveness of a program.
7 The-last section in the Regulatory Guide is 8 Feedback'and Corrective Action. As a result of the 9 Performance' Monitoring and Assessment, the last step in.the 10 process is to determine the need for corrective action.
11: Again using Section IV of the Regulatory Guide as a template 12 for areas that may need to be addressed.
13 MR. MICHELSON: Can I ask you a question. I'm still puzzling on the previous slide and I had to'look it up.
(O~Y '14 15 again.
16 You talked about right off the top, establishing 17 ' goals, objectives and use indicators. But that doesn't seem 18 to be the set that you're talking about in the Regulatory 19 Guide when you talk about' goals and so forth related to the 20 structure, systems and components. It's not necessarily the 21 same set or is it?
- 22 If it is, then I don't understand the rest of the 23 slide. Are these the objectives that you are talking about 24 when you talk about, " define the overall policy and 25 objectives of the program," and then, "The maintenance !
t Heritage Reporting Corporation f (202) 628-4888 !
)
r i
i 156
( E 11 required,~on the various structure system components shall be l
2 directed toward achieving these objectives." !
3- So which--for instance, on a component like a:
4 . motor operated valve, you got a lot of them in'the plant.
.I 5 That is one kind of'an objective, maintenance'of valves. l i
(i Another kind of objective,though, is availability of- H
- 7. . auxiliary feedwater.
8- Which kind are you talking.about here?
9 MR. DEY:- These parameters, these indicators.could j l
10 be at various levels of the plant. Some would be at the '
i 11 ' level of the feedwater--auxiliary feedwater system. Some i 12 would be at the level'of the valves,.if necessary, and-13 determined to be so by the licensee.
- 14. MR. KING: They're the same goals and objectives
()
15 we were talking about at the beginning of the Reg Guide.
16 Section C-3.
17 MR. MICHELSON: Yes--okay. I always thought those ~[
18 to be system level and component level objectives. They 19 seem to be a little bit different. Process indicators and 20 so forth--
21 MR. DEY:' Yes. The process indicators are a 22 little different, you're right. These are programmatic 23 indicators and not--
24 MR. KING: There may be a little confusion here in
' 25 that I guess what we are trying to say is there is more than h Heritage Reporting Corporation (202) 628-4888
7= ',
157 1 just E looking at quantitatively,'are you achieving your goals
( )~
2- and objectives that you set'back in Section 3.
MR. MICHELSON: On the structure, systems and 4- components?
5 MR. KING: In terms of assessing the effectiveness 6 of maintenance, there is more than just looking at, are'you 7 achieving those numerical values. It involves self 8 assessments. It involvec some qualitative evaluations and 9 assessments and you could use other indicators besides just 10 the things that you have established for goals and 11 objectives, things like process indicators.
12 MR. MICHELSON: Now those indicators have to be j 13 related to something like a goal, 14 MR. DEY: That's right.
(}
15 MR. MICHELSON: They indicate how close to the 16 goal you're getting, is that right?
17 MR. DEY: Yes. Your goal, for example, could be 18 to minimize a certain type of transient and you set a 19 target.
20 MR. MICHELSON: So by process indicators, did you 21 mean, system performance indicators?
22 MR. KING: No. Process indicators don't 23 necessarily have to be related to a goal. They probably 24 wouldn't be.
25 MR. MICHELSON: So there's more in here than there
() Heritage Reporting Corporation (202) 628-4888
158 i is that we.have talked about in.the Regulatory Guide where-
'2 _We talked about only structure systems-and components.and 3 not objectives:and goals and--
4 MR. KING: There's more'in Section V, that's 5 .right.
6 MR. MICHELSON:' This is something bigger than-7 that?
8 MR. KING: Yes.
L 9 MR. DEY: Yes. And, you.know--excuse me for not 10 having a schismatic. What this cection would do'is you.
11 would monitor the performance of the plant and your program, 12 your maintenance activities and try to assess if you need 13 ' corrective actions and this process indicator is really that
'14 link towards how the equipment is performing and what your
-15 ' program'is doing. That provides that link ~.
16 Maybe I will try and put together--
! _.1'/ MR. KING: I think this slide is confusing. We
.' ~
18 need to present-this a little better for.the full committee.
E19 MR. MICHELSON: You need to relate it- you know, I 20- see you can go-beyond what you talked about.in the 21 regulatory guide and you make it clear that'you've gone 22 beyond that--
23 MR. KING: Yes, I see your confusion.
24 MR. MICHELSON: --but this has to relate to the
25 regulctory guide as well--
O '
erieae. Re,orei- , co ger tion (202) 628-4888
159 f)/
A-1 MR. KING: Yes.
2 MR. DEY: Yes.
3 MR. MICHELSON: --and its goals, but these are 4 goals beyond the Regulatory Guide that you also want to 5 track somehow.
6 MR. KING: Well, these are things that you would 7 do as part of assessing the effectiveness of maintenance, 8 beyond just looking at your goals is what it is.
9 MR. MICHELSON: But you didn't give us any ,
1 10 background on these, you just--
11 MR. KING: I understand.
12 MR. DEY: Yes, that's right. We'll try and make a 13 better presentation on that.
14 MR. MICHELSON: Is there any written material?
/}
15 Are you going to--that's not going to be in a Regulatory 16 Guide, is it?
17 MR. WILLIAMS: Let me give you, maybe a 5 minute 18 discussion on that and it might help a little bit.
19 MR. MICHELSON: All right. I think it might be 20 worthwhile, unless I'm the only one confused by it.
21 MR. WILLIAMS: It's a valid point. The front and 22 the back need to be put together better, Carl.
23 These goals--these indicators are really focused 24 on maintenance and back in June, AEOD had a program to try 25 to determine some maintenance effectiveness indicators.
(O
,/ Heritage Reporting Corporation (202) 628-4888
E 160 l 1
1 We visited 23 plants, talked to a lot of
])
2 utilities, a lot of people and wrote a report back in j 1
I 3 November and we wrote a second one in January.
4 Basically the way that we wound up structuring 5 things was that we set the process indicators that utilities 6 were using. These were things like the backlog 7 measurements, preventive maintenance to total maintenance S measurements, different criteria that were really management 9 -process controls.
I 10 So we came back and we tried to correlate those to 11 some other overall planned performance measures. Maybe the 12 equipment costs forced outrageous rates and the like.
13 We wound up deciding that there wasn't any 14 correlation based on the analysis, but these indicators were
}
15 very useful-to control backlog and there is a structured set 16 of definitions out there and they all have a good use as j 17 management controls. l 18 The area we had the most success in finding a 19 correlation to overall planned performance was in the 20 analysis of component failure rate and we picked up 21 component failure rate information from the plants. We 22 brought it back, put it in data bases and ran SATS, 23 Statistical package hundreds of times.
24 Then we decided that the component failure 25 information provided the best avenue to pursue to try to j
() Heritage Reporting Corporation (202) 628-4888 l
161 l() 1~ find some correlation to overall plant performance.
2 Then the. task was to decide how we could connect 3 that or understand how it was connected to maintenance 4 effectiveness.
5 . We went through another analysis which is 6 documented in a Commission paper dated February 6th or so, 7 that was sent over, maybe two weeks ago and this ia on 8 Maintenance Performance Indicators. And, in there, we said 9 that there are indicators that we decided are valid to 10 monitor the effectiveness of maintenance. The process 11 indicators are valid to monitor the maintenance process, to 12 control it. The effectiveness indicators monitor the results i
13 of maintenance.
14 We did not find plants using effectiveness
(}
15 monitoring of maintenance. We did find, engineering groups 16 doing RCM analysis or small efforts in small certain 17 systems, but we didn't find kny integrated efforts to 18 monitor the effectiveness of maintenance at the plants we 19 visited using quantitative measures.
20 Now IN?O does have some things now involving 21 analysis using the NPRDS data base and we also turned that 22 data base to get out best results.
23 So the effectiveness indicators- put the other 24 slide back up. This slide basically boils down into three 1
25 areas, overall plant indicators, which are availability,
. i
() Heritage Reporting Corporation (202) 628-4888 l
l
162
~ l j
1 forced outage rate, some overall plants trends and then you l 2 get to the process indicators which are backlog
. I 3 measurements, ratios of maintenance kinds, preventive to i 4 total and what not.
5 And then the effectiveness indicators which are 6 really equipment oriented indicators and we can discuss an 7 example of that next week with you.
8 And the NPRDS again, we--
9 MR. MICHELSON: I think I understand--
10 MR. WILLIAMS: --I don't want to get into that 11 again, but that is the data base we're trying to--
12 MR. MICHELSON: I think I understand where we're 13 at. What you're saying is that there must be a program as (3 14 required by this regulatory guide, the utility must have t/
15 some kind of program for not alone tracking their structure 16 systems component performances and establishing goals and so 17 forth for them.
18 They must also have a program which is an overview 19 of the maintenance process, per se--
20 MR. WILLIAMS: Right.
i 23 MR. MICHELSON: --and that there are certain goals 22 thrJc must be established and methods of assessment and that 23 is what you are discussing now.
24 MR. WILLIAMS: That's exactly right.
25 MR. DEY: That is the link between--
() Heritage Reporting Corporation (202) 628-4888
163
( ) 1 MR. MICHEI SON: I'm back with you. I'nt sorry we--
v 2 MR. DEY: --monitoring before determining the 3 corrective actions and that is the aim of the process.
4 MR. MICHELSON: This is another step in the 5 process?
6 MR. DEY: Yes.
7 MR. MICHELSON: This now is essentially--
8 MR. DEY: It's essentially--
9 MR. MICHELSON: Are utilities doing this or is 10 this mostly new?
11 MR. DEY: I think--
12 MR. WILLIAMS: I think they are pretty much 13 universally doing process monitoring. IMPOs indicators are
("T 14 oriented towards process information. INFO has an overall !
(/ l 35 indicator gross heat rate that is somewhat tuned in to 16 maintenance effectiveness of the plants as would be 17 availability and other large measures. i 18 At the component level, in a systematic way, I 19 think things are just beginning and it involves a lot of 20 training that they are doing in the last year or so, 21 industry wide.
22 And tha . ads of things that we are trying to have 23 utilities do t , monitor their actual equipment performance 24 to assess maintenance in a structured way. I think they're 25 just beginning from what we found in our site visits and
(~T Heritage Reporting Corporation
(_/
(202) 628-4888
164
() 1 other staff reviews.
2 MR. MICHELSON: Go ahead.
3 MR. DEY: Okay. Feedback and Corrective Action, 4 the sectioS in the Regulatory Guide, I think it is fairly 5 straight forward. The purpose of feedback is to determine 6 the need for corrective action. The people responsible 7 should be identified and corrective action should be 8 directed toward the deficiencies.
9 The process of corrective actions determine the 10 cause of deficiency, provide timely action, and provide 11 documentation of what actions have been taken.
12 MR. MICHELSON: Let me ask. We pursued this 13 question, if you have got a real good program, then you 14 don't have to have so much documentation.
[
15 MR. DEY: Yes.
16 MR. MICHELSON: We were talking structure system 17 and component documentation, you did that. Does that same 18 argument apply here? If you got what people view as a good 19 program, do you still have to go through and perform this 20 part of the regulatory guide?
21 MR. DEY: I think the documentation for corrective 22 actions could be very minimal but identify exactly what is 23 being done. I don't think we' re talking about volumes of 24 writing.
25 MR. MICHELSON: Well some of this is volume, some l'h
(_/ Heritage Reporting Corporation l (202) 628-4888 j 1
f
I '1 L ,
- )
165
() 1 of it isn't.
2 MR. DEY:- I think'this is simply an identification 3 of the' deficiency and what. corrective action is being taken 4 and tracking whether the action you have taken is causing an 5 improvement.
6 MR. MICHELSON: But the same argument will apply I
7 here as well. If you have got good SALP ratings, so to 8 speak, then you don't have to do so much documentation, 9 which means that you don't have to do a lot of this.
10 DR. KERR: Presumably 44 percent of the plants, 11 unless they changed drastically, wouldn't have deficiencies, 12 .so they won't have to do anything.,
13 MR. MICHELSON: And they won't be doing this 14 trending and all the other things talked about in here.
{}
15 MR. KING: Well, I think they would be certainly 16 doing-some trending. I think the extent of what they have 17 to do over and above what they are currently doing, to keep 18 their programs good, would be minimal, is the way we would 19 see it.
20 MR. MICHELSON: Th6y may calculate them in their 21 heads.
22 MR. DEY: They may be.
23 MR. CARROLL: Realistically I think, most of them 24 have a trending program.
25 MR. DEY: Yes, I think so.
Heritage Reporting Corporation (201) 628-4888
. . . - _ _ _ _ _ - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ - _ . _ . - _ _ J
166
() 1 MR. CARROLL: And that is one of the reason's that 2 they are in the 44 percent.
3 MR. DEY: That's right. The last two bullets were 4 timeliness and management involved in what corrective 5 actions have been taken and whether they are actually 6 impacting--improving the effectiveness of the program.
7 And that is all that I had to say in this section.
8 At this time I will give it over to Tom King for discussion 9 on the backfit analysis.
10 MR. KING: I'll go to the slides that are in the 11 last section of the handout. .I think they should have a 12 Roman Numeral VII on them. I am going to talk about the 13 first three and then Brian Richter will talk a'little bit 14 about the status of the regulatory analysis.
{}
15 Basically what I wanted to talk about was the 16 supporting basis for the rule and the reg guide. As I 17 mentioned earlier, we're not purs.ing this on an adequate 18 protection basis anymore, we're pursuing it on an enhanced 19 safety.
20 We have basically two thrusts in looking at 21 enhanced safety. One the qualitative discussion and second 22 the quantitative or the regulatory cost benefit analysis. I 23 DR. KERR: Excuse me, which Roman Numeral section?
24 MR. KING: VII. It should be the last stapled i
25 package in your handout.
I .
() Heritage Reporting Corporation (202) 628-4888
. _ _ _ - _ _ _ - _ - - _ - _ - _ _ __-______- _____ - - _____- - - __=________ _ - ______-____ ____ _ _ _ _ - _ _
l F
167.
( )l 1 DR. KERR: Thank,you.
2 MR. KING:l We received many comments on the-3 previous regulatory analysis. As you recall, we did a cost 4 benefit look'at the proposed rule. We are currently 5 considering those comments, trying to revise the regulatory 6 impact. analysis. We don't have a bottom line yet to 7 present. Hopefully in a week or so we will. I do want to 8 discuss the qualitative argument today.
9 Basically, we believe it's fairly clear that 10 . maintenance has a direct impact on plant safety. That can 11 be determined not only from looking at existing performance 12 and past performance of plants and good engineering. judgment ~ '
l 13 will lead you to that conclusion.
14 That impact really has two main components, one
(}
15 affects--maintenance affects the reliability of the safety-
'16 systems as well as maintenance can affect the number of 17 challenges to those safety systems and to the operators 18 which are certainly a line of defense in the plant.
19 Therefore, we believe it is reasonable and proper 20 for NRC to regulate in the maintenance area provided that 21 it's in a fashion that will result in an improvement in 22 safety.
23 We currently have some existing rules that 24 implicitly implied maintenance requirements, particularly in 25 the safety systems. We feel that, in addition, this rule j .
Heritage Reporting Corporation (202) 628-4888 1
168
() 1 will help provide some more explicit guidance as to what we 2 want to see in terms of maintenance, maintenance programs, 3 performance of maintenance, as well as include the BOP, 4 which is probably an extension of what the current 5 regulations cover.
6 The second point is, there is a variation across 7 the industry in maintenance. We have used in the proposed 8 rule some data from NUREG 1212 to talk about that variation.
9 Currently we have got the maintenance team inspections under 10 . way. I have mentioned these percentages before as to the i 11 currently completed maintenance team inspection,.what the 12 percentages of the rankings of the plants are.
13 MR. MICHELSON: Do you know, off hand, how many, i 14 in each one of those 3 categories might have had a SALP 1 j 15 rating already, if you went back and looked because that 16 " good" doesn't mean SALP 1 necessarily.
17 MR. KING: We can get that information. I thought 18 generally when we talked about this before, that generally 19 the goods were l's and the fairs vere 2's and the poors were 20 3's.
21 MR. MICHELSON: I would hope that is not the case 22 because I thought SALP 1 was considered more than just good, 23 but rather excellent and SALP 2 was certainly fully adequate 24 and SALP 3 was adequate with deficiencies or something.
25 But I don't think that this is 1, 2, 3, in those Heritage Reporting Corporation (202) 628-4888
169
() 1 numbers,.but I don't know, that's why I asked you.
2' .MR. KING: We can get the numbers--
3 MR. MICHELSON: Okay.
4 MR. KING: - from the plants that have been looked 5 at, that's not a problem.
6 MR. MICHELSON: Okay. Thank you.
7 DR. KERR: It's reasonable to assume that 11 of 8 those 27 plants were considered good, is that--
.9 MR. MICHELSON: Yes, that's the way-I read it.
10 MR. KING: I think the percentages are based upon-11 the site, as I recall, not on the number of plants. The 12 maintenance team inspection reports are written on a per 13 -site basis.
14 MR.-MICHELSON: Well, the good is a good site for
)
15- a site?
1:6 MR. KING: Is that right, Ken?
17 MR. HART: Yes.
18 DR. KERR: Then it will be 8 of 18.
19 MR. KING: The site may be 1, 2 or 3 sites. i 20 DR. KERR: Phich 8 were those?
21 MR. KING: Pardon me.
22 DR. KERR: Which 8 were those plants?
23 MR. KING: Did you bring a list? Who has the 24 list? Do you have the list?
H25 DR. KERR: I don't need it now. If I can just--
O~ rie e- eerei e correr eie-(202) 628-4888
4 170
() 1 MR. KING: We can get you the list. We have the 2- list of plants.
3 MR. CARROLL: And also, what was their SALP 4 rating?
5 MR. MICHELSON: Yes, on maintenance at least.
6 MR. CARROLL: And surveillance.
7 MR. MICHELSON: That's right, those two.
8 MR. ZWOLISKI: If I could interject, moving 9 forward with the entire maintenance rule process, the office 10 directors were concurrently developing a status report card 11 on maintenance team inspection results to date. That memo 12 hasn't been issued yet. It does contain much of the 13 information that I understand you're seeking.
14 When it becomes available, I will assure that the
,{}
15 subcommittee is provided the information.
16 MR. MICHELSON: Will that be by next week, 17 perhaps?
18 MR. ZWOLISKI: If it's not, I will be able to l
19 provide the specifics as far as plants that have been 20 inspected and inspection reports issued and their SALP 21 ratings.
22 Was there a third question?
23 MR. MICHELSON: No. I think that was it.
l 24 MR. ZWOLISKI: I would sensitize you to not draw 25 the immediate one to one correlation that this SALP 1 SALP 2 Heritage Reporting Corporation l (202) 628-4888 l~
l
_- -- - - - - - - - - a
I 171
() l' and 4 to SALP 3.
2 The inspection procedure was not initially 3 developed to rank plants SALP 1, 2 or 3. We are in the 4 process of-re-evaluating, should we score on our inspection 5 in that mode or methodology tends to have 3 categories that 6 is overlapped with SALP. It's a little premature is what I 7 am saying.
8 MR. CARROLL: As long as you're making a table, 9 you might add to that automatic SCRAMS, safety system 10 unavailability for some period of time. j 11 MR. KING: Finishing up the second bullet, given 12 that there is a variation across the industry in 13 maintenance, we believe the rule will facilitate the l
l 14 Commission's'taking action in the maintenance area to l
(" }
15 achieve improvement where just such improvement is needed.
16 The third bullet addresses the point that we think 17 a regulation will, over thu remaining life of the plants 18 provide some stability and better define the requirements 19 for maintenance that will help insure good maintenance 20 practices are achieved and sustained and I think " sustained" l
21 is the key word there or backsliding in the future and to 22 help insure the affects of aging are factored into plant 23 maintenance programs.
24 You're aware that the NRC is doing quite a bit of 25 aging research. They are starting to issue reports. We want
( Heritage Reporting Corporation (202) 628-4888 I
L L.--_-___-.
a O' <
- y. l 172
]
{) -1 to make sc e that those things are being considered by the 2 industry and factored into t'he plant maintenance programs to j j
3 help correct any deficiencies that aging would bring about. !
I 4 And finally, we think it is cost beneficial to' f 5' have a good maintenance program. Just qualitatively some 6 plants are already doing it. There are plants out there 7 that we judge have good maintenance programs, have good 8 system availabilities and so forth. To us, obviously, if it 9 wasn't cost effective for them to do that, they wouldn't be 10 doing it.
11 From a qualitative sense, we think that 12 improvement can be achieved in a cost beneficial manner. I 13 know we discussed this a little bit this morning and there 14 may be some disagreement that not everybody could reach that
{}
15 same conclusion, but--
16 MR. MICHELSON: Has there been any attempt to 17 correlato or maybe you have, correlate the good maintenance 18 to--pardon me, a good availability to a SALP 1 rating on 19 maintenance or something of that sort?
20 Is there some attempt that has been made to take 21 the operating history of the plant in terms of availability 22 or power production, whatever and relate that to how their I 23 SALP ratings have been coming out?
24 MRc KING: We have done that in the proposed 25 package.
() Heritage Reporting Corporation (202) 628-4888 j
~ ,
J
! N.M,
- l Y '
173' (1
1 MR. MICHELSON: Okay.
.2 MR. WILLIAMS: We have done that in the proposed- '1 3 package, yes. ,{
4 MR..MICHELSON: .Okay.
'5 MR. KING: Not only have we done it, we sent 6 'information to.2NPO and INPO did it.
7 MR. MICHELSON: Could you tell us, in summary 8 form, how that came out,or is it possible to summarize it?.
9 MR. WILLIAMS: We didn't find any correlation.
10 MR. MICHELSON: No correlation.
11 MR. WILLIAMS: And in addition, in safety system 12 unavailability, the things that influence that heavily are
'13 design basis,. reconstitution efforts, where we find--
14 MR. MICHELSON: Are mostly focusing on power 15 production because that was the thrust of the bullet, was 16 the economics, 17 DR. KERR: What is the design versus 18 reconstitution?
19 MR. WILLIAMS: If a plant is trying to reconstruct 20 its design-basis and is going through a systematic review of I
21 a system and finds a, whether it be a single failure or l 22 whether it be a condition that has existed for some time I
23 since the modification, they would wind up with a safety 24 system unavailability report likely, and that would be what 25 you get in this data table.
Heritage Reporting Corporation (202) 628-4888
3
!' 1 I
a ,
174 I 1
[ , 1 MR. MICHELSON: So the last bullet hasn't been j 2 justified by any kind of correlations then, is that it, just ,
l 3 an intuitive feel? j 4 MR. KING: If you recall, in the proposal making 5 package, there was a cost benefit analysis that had some 6 correlations in it and as I-"
7 DR. KERR: But can it be not cost beneficial to 8 have a good maintenance program because.of going cost 9 beneficial it wouldn't be a good maintenance program.
10 MR. MICHELSON: But you could, perhaps, go 11 overboard and have a very expensive maintenance program that 12 is not really enhancing--
13 DR. KERR: But then it wouldn't be a good 14 maintenance program.
(' )
15 MR. MICHELSON: Well it depends on how you define 16 good.
17' MR. CARROLL: Overboard maintenance program.
18 MR. MICHELSON: Yeah.
19 MR. KING: It wouldn't be optimal then.
20 MR. WILLIAMS: The problem is that if you take the 21 strength of a quantitative correlation. that is one piece of 22 evidence that--my answer was, we haven't been able to 23 produce clearly statistically significant evidence that 24 shows that, however, we do and in our development work on 25 indicators, we did have clear cut cases where you could say
() Heritage Reporting Corporation (202) 628-4888 f
i.
, Ji l ' ,
k 175 o .- )
} l there appeared to be evidence, but overall, industry wide, 2 if.you did it,'I don't think we could give you a rl -3' quantitative rule.
4 MR. MICHELSON: But it's a pretty good gut 1
5 feeling. )
6 HR. WILLIAMS: And.in some cases there is a clear 1
7 correlation, but not in all cases or enough to stand on its J 8 merits.
9 MR. MICHELSON: Okay.
I 10 MR. KING: The last thing, which really is an 11 introduction to what Brian Richter is going to say, is that 12 the quantitative basis, we are re-evaluating, they can look 13 at what we did at the proposed rule making stage, the 14 comments we received, the new information that'has come out 15' since November, in trying to revise the regulatory impact 16 analysis.
17 Basically we're taking a look along the lines that 18 good maintenance equates to reduced risk, trying to
-19 estimate, with the rule and the reg guide, the reduced risk 20 due to improvement of the poor and fair performers and that 21 it should be cost beneficial to achieve a good performance 22 level and good maintenance, hopefully, would equate to some I-23 net cost savings by reduced down time and so forth.
24 We don't have any bottom line to give you any 25 numbers today but Brian is going to summarize some of the Heritage Reporting Corporation (202) 628-4888
~
176 majorfcomments and some of the areas that are being revised
'(v"') -1 '
2 in the regulatory analysis.
1 3 MR. RICHTER: I'm Brian Richter from the Office of '
4 Research.
5 As Tom said, I will be talking about the public 6 . comments received on the draft reg analysis and also the 7 status of the' revised analysis.
8 Extensive comments were received. It was decided 9 to follow basically--to address the comments that NUMARC had 10 expressed concerns about. Most of.the other comments we 11 received, we were able to fit into the generrl concerns that 12 NUMARC had.
i 13 The first point was a concern on the over reliance !
14 on NUREG 1212 which was from '80 through '85 data. The
)
15 second point was a concern that the' costs were 16 underestimated. The third point related to the benefits or 17 the risk reduction of the rule being over estimated, and the 18 last concern was that many of the assumptions were highly 19 uncertain.
20 This table shows a comparison between the draft 21 reg analysis and the revision that we're working on now.
22 For the cost elenients, in the draft reg analysis, NUREG 23 1212, was the source. In the revised Reg analysis, we're
!~
24 working with the draft regulatory guides.
25 With respect to the number of plants requiring i
I
() Heritage Reporting Corporation (202) 628-4888 1
1 J
- l s
(.
h 177 l : 1'. improvements.in the draft that was originally NUREG 1212 as -
2 well, this time.we have some maintenance team inspection 3 data to work with.
)
11 A key-point, I think, is the number of performers
]
I 5 in the three categories. .In the Draft Reg Analysis, we 6 estimated 20 to 30 poor performers based on NUREG 1212. !
l 7 with the maintenance team inspection, we're estimating 7. l 1
8 The Draft Reg Analysis did not address any of the )
l l 9 fair performers. In this case, there are roughly 60 of them 10 and the good' performers originally estimated at 90 to 100, 11 we're now estimating about 53.
12 MR. MICHELSON: Now, what is good, fair and poor?
13 There is-some: kind of criterion and inspection module?
14 MR. RICHTER: This is based on the maintenance team
{
15 . inspection-data and---
16 MR. MICHELSON: And in that module for the 17 inspection or the procedure for the inspection, they define 18 what would make a good performer, fair and a poor, is that 19 the idea?
- 20 MR. ZWOLISKI: There is a criteria established and 21 -there is a graded scale on each evaluation area and valuate 22 the program and the implementation of the program.
23 MR. MICHELSON: And did not necessarily relate to 24 SALP ratings--
25 MR. ZWOLISKI: Correct.
O erita,e Regertin ,
(202) 628-4888 cergeraeien
{
1
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ . . _ _ _ _ ____ _ _ _ . _ _ _ _ _ _ I
.178-() 1 'IgR. MICHELSON: --of 1, 2 or 37 2 MR. ZWOLISKI: Correct.
3 MR. RICHTER: Correct.
4 DR. KERR: Now, are the 7 poor performers that you 5 are now using typically plants that are already shut down 6 for other than adequate reasons?
7 MR. RICHTER: This was based on the sample of the 8 maintenance team inspections. We just projected it based on 9 the percentage to a 120 units.
- 10. MR. KING: There was one plant that came out as a 11 poor on the maintenance team inspections and given that 12 percentage, we project if they did all of them, they would 13 find 7 of them that fell in that category.
14 MR. CARROLL: Was that a plant that was already
{}
15 down because of other problems--
16 MR. KING: Oh no.
17 MR. ZWOLISKI: We have completed inspection at a 18 plant that is shut down by order and it fell in the good 19 range. I guess on this chart fair, excuse me.
20 MR. CARROLL: Okay.
21 MR. RICHTER: With respect to the risk analysis, 22 as the draft reg analysis pointed out, and as has been 23 pointed out mhny time, maintenance does not really lend 24 itself to the PRA techniques. Again, basically, because of I
l 25 the lack of data, the limitations of the PRA models.
O
(,) Heritage Reporting Corporation (202) 628-4888 l
l
l 179
() 1 The second point is that the risk sensitivity 2 calculations can indicate a broad range of values, again 3 lending to the uncertainty.
4 The third point, as we mentioned, we are 5 considering more recent data and lastly wt a the changes in 6 the number of plants and the different categories.
7 DR. KERR: Does that first bullet the imply that 8 when you talk about cost effectiveness you're not talking 9 about risk reduction, but you're talking about plant 10 availability and how much more money you make or something 11 and how can you use PRA techniques if you're going to 12 calculate risk reduction?
13- MR. KING: You look at both. There is the change
{ . 14 in risk as a result of the maintenance rule as well as the 15 change in the costs.
16 DR. KERR: What does, " maintenance does not 17 readily lend itself to PRA techniques, mean then?
18 MR. KING: I think it means you have to make some 19 assumptions on- you know, PRAs traditionally haven't gone 20 out and tried to identify maintenance problems.
21 DR. KERR: No, this doesn't say PRAs, it says, 22 "PRA techniques."
23 MR. MULLIN: I would like to respond to that 24 question. .My name is Mark Mullin. The idea there is that 25 PRAs can model things such as the frequency of transients, l
( Heritage Reporting Corporation (202) 628-4888 l
180 (f l' the. availability of safety systems and that is fairly 2 straight forward. The problem is in relating those kind of 3' parameters to maintenance, as was indicated by a number of 4 the questions previously asked.
5 DR. KERR: That's what I thought.
6 MR. MULLIN: To draw that link and perhaps, that 7 is strictly _ speaking, not a part of the PRA technique 8 itself.
9 DR. KERR: I thought what you were saying is that 10 you can't really get a good number for risk reduction due to 11' maintenance improvement. Is that what that implies.
12 MR. WILLIAMS: I think, based on operational data, 13' that is difficult. With a lot-of assumptions, it gets less 14 .and less.and less difficult.
{}
15 DR.~KERR: I'm not-trying to disagree, I just want 16 to make sure I understand.or do not understand what the 17 bullet means.
L 18 I thought it meant thatLyou don't'have much; 19 confidence in using PRA techniques to calculate risk 20 reduction due to maintenance improvements.
21 Is that what you mean?
22 MR. WILLIAMS: There are inherent limitations in 23 techniques that make it very difficult to--
24 DR. KERR: I don't disagree with this if that's 25 what you mean.
- O aeritase Re,oreine correlation (202) 628-4888 i
l l
_ _ _ _ _ _ _ _ _ _ _ _ .m __.____._____m________. _.
181 1
1 MR. WILLIAMS: That's right, okay.
{) '
2 DR. KERR: That given that, then there is either )
i 3 quite a lot of uncertainty in your cost benefit calculation, l 4 if you do it on the basis of risk reduction or else you're 1 5 using something else to demonstrate cost benefit and the !
6 something else could be availability. ;
7 So it's risk reduction, but there is a lot of 8 uncertainty.
9 MR. WILLIAMS: That's right.
10 DR.-KERR: Okay.
11 MR. WILLIAMS: I should add, muybe, that--
12 MR. MICHELSON: We will have to keep moving along.
13 We've got about 8 more minutes and then the last 10 minutes 14 are subcommittee time.
[}
15 MR. RICHTER: All right.. On the cost analysis 16 side, the revised regulatory analysis has been restructured 17 to follow the Reg Guide elements so that they are more easy ,
18 to follow--so it is more easy to follow.
19 One key change that will obviously result is a 20 decrease in the cost resulting from the smaller number of 21 poor performers and this is consistent with one of the 22 NUMARC comments. In their re-analysis they suggested, 23 roughly, 7 poor performers.
24 On the other hand, in this analysis, we're 25 considering the inclusion of the fair performers which was
() Heritage Reporting (202) 628-4888 Corporation j
i l
)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . _ _ _ . _ _ I
t 182
() 1 .not addressed in the draft reg analysis.
2 And the last~ point is that.we are considering 3 additional data, and that is post NUREG 1212, as it has been 4 discussed.
5 MR. CARROLL: There I take it, is someplace 6 between a 2 and a 3, some of the pairs might slip down to a 7 SALP 3 and I heard you say, as early as this afternoon, that 8 a 2, we're probably going to leave alone, if they're happy.
9 MR. RICHTER: I think that is one thing that is 10 going to have to be discussed in our fine tuning-the revised 11 reg analysis.
12 MR. MICHELSON: I also heard that the 2's are 13 going to have to have objectives that will get them to a 1
{) 14 15 even though they don't wish to achieve them, if I' understood everything that--
16 MR. ZWOLISKI: We don't want licensees 17 establishing frivolous goals.
18 MR. MICHELSON: But SALP 2 is not a frivolous 19 achievement necessarily.
20 MR. ZWOLISKI: And for that licensee that may be 21 the appropriate goal level for--
22 MR. MICHELSON: I thought it was appropriate for 23 all licensees.
l~
24 MR. ZWOLISKI: We're striving to articulate that 25 we think they shoot for the highest goal attainable. If, A
() Heritage Reporting Corporation (202) 628-4888 I ,
l 183
(} -1 for a given licensee, that does not correlate with resource.
2 allocation and other burdens, a reduction would be 3 appropriate, as long as the basis for their determination is I
4 articulated and they know why they have set their goals and l 5 that is the ones to have.
6 I think that part of our discussion earlier, we 7 failed to realize we expect licensees to understand their 8 own programs'better than we do and they ought to have a 9 basis for the decisions they have made in establishing their 10 goals and objectives, whatever they may be.
11 We're trying to articulate--set them as high as 12 possible. That may not be the real world solution. We may 13 not be able to attain that.
14 MR. KING: We may want to take another look at our
(~J) u 15 words to state better what John just stated right now, so--
16 MR. MICHELSON: Sort of an ALARA principle in the f
17 other direction is as high as reasonably achievable.
18 MR. RICHTER: The last major concern dealt with 19 the uncertainties of assumptions and we were able to nail 20 some of them but there still are others that remain fairly 21 large. !
22 10. . CARROLL: So what is your prediction when all 23 is said and done? Is the value impact analysis going to 24 justify the rule or not?
25 MR. RICHTER: If we had the answer, we would have i
l
() Heritage REJorting Corporation (202) 628-4888 I
184
'( ) 1 given it to you now.
2 MR. KING: We're still cranking the numbers. It's 3 probably going to be like it was in the package that went 4 out with the proposal. There will be a range where it is 5 and a range where it isn't. Depending on whether you agree 6 with the assumptions or not, you can put yourself in any 7 range you want.
8 MR. MICHELSON: Staff has about 5 minutes to say 9 whatever else it wishes to say and then we'll have a 10 subcommittee discussion.
11 MR. CARROLL: I guess I have one more question.
12 .The whole pitch that NUMARC has made is, hey guys, we're 13 doing all of this, we don't need this. Are you looking at
/~T 14 this analysis using that as a base case, that the Commission V
15 doesn't promulgate a rule or a Reg Guide and industry does 16 what they say they're going to do?
17 MR. KING: We do give credit for the industry 18 initiatives in the Reg Analysis. We did in the original l
19 package and we're doing it this time.
20 MR. CARROLL: And do you argue that by having a 21 rule in the Reg Guide, there are going to be more benefits 22 to the industry than if you just let INPO and NUMARC and the l'
23 utility industry do their thing?
24 MR. KING: I think we're looking at that, yes.
25 We're trying to make that argument. We're looking at the Heritage Reporting Corporation (202) 628-4888 s
).
185 l' numbers.
2 MR. ZWOLISKI: I think you're putting'some words 3- -in our mouth.. You have to be careful.
4 We went over the rule this morning, Options A and 5 'B.
6 MR. KING: Yes.
7 MR. ZWOLISKI: One option is follow Regulatory 8 Guide, Staff Development. Another option would be to go on 9 an industry standard and allow the industry to conform to 10 that particular standard. I'm not arguing that it's apples 11 and. oranges.
12 MR. CARROLL: I'm just taking it as a given that 13 industry hasn't played that game. I guess I would like to 14 see in this value impact statement, Case A. Industry does
(}
15 what they say they're going to do over a period of time and j 16 the maintenance gets better because of industry initiatives 17 and you can draw that line with a bunch of different rules, 18 depending on how effective you estimate those programs are 19 going to be,
- 20 I guess I would like to see those cases compared 21 to the cases--how much better it is going to be by virtue 22 of the fact that you have a rule and a rate making.
23 MR. MICHELSON: That's a comparison you have to 24 make, I believe.
25 MR. DEY: You are going to see quantitative l t Heritage Reporting Corporation (202) 628-4888 l
l
186
( )' 1 analysis, but the backfit analysis we will provide you in 2 the final package will have qualitative arguments also.
-3 MR. MICHELSON: By'the final package, you mean 4 next Wednesday?
5 MR. DEY: Next Wednesday, right. As to why the 6 rule would enhance safety.
7 MR. ZWOLISKI: I would caution that plants on E paper, whether it's this regulatory guide or industry 9 developed standard or has it, but paper is not going to make 10 maintenance programs any better in the majority of the 11 cases.
i 12 The honesty and truthfulness that a licensee wants 13 to really buy in and implement the program, those programs 14 exist today. We're seeking improvement now. We're going to
(}
15 attempt to articulate differences as we understand them 16 between what industry is doing today versus our perception I 17 of where the regulatory guide is going.
18 In the context of that difference, recognize that 19 we have a fairly ambitious team inspection activity with the 20 target area currently being mainter.ance. The resuit of that 21 activity appear to be bearing fruit universally as far as 22 industry paying more attention to maintenance.
i 23 MR. CARROLL: I think that's true. I think the
{
l 24 team inspections ought to be put in and I emphasize industry l 25 initiatives and I think that is something that is going to Heritage Reporting Corporation (202) 628-4888
l 187 )
() 1 happen whether this is instituted or not.
2 MR. MICHELSON: Tom, do you have any other words? )
3 MR. KING: If we can just take one minute.
4 MR. MICHELSON: Sure, go ahead.
5 MR. KING: I think, to wrap up, you know, we tried 6 to take an approach where we don't prescribe to the industry 7 how to do maintenance. That is still consistent with the 8 general rule and we tried to put together a Reg Guide that 9 will really concentrate on the process that we would like to 10 see the industry follow in a maintenance program that would 11 include setting goals, having self assessments and making 12 corrective actions where those self assessments say 13 corrective actions-ought to be made and that is the thrust
/\ 14 of what we're trying to get across in this rule and Reg O
15 Guide, not to prescribe how to do maintenance, not to 16 require a pile of paper, but to have the utilities go 17 through and have a program that will cause improvements 18 where it makes sense to make improvements.
19 That is the message I want to leave you with.
20 Maybe you raised a lot of good points and maybe some of the 21 words need to be cleared up, but that is really what we're 22 trying to do with this.
l
( 23 MR. MICHELSON: Any questions to present to the 24 staff from the subcommittee?
l 25 DR. KERR: I would just like to comment that I I-Heritage Reporting Corporation (202) 628-4888 1
1
,i 188
() 1 think that,the presentations today have been well organized 2 and there has been a good response to our questions and good 3 nature responses to our comments and I appreciate that.
4 MR. CARROLL: I guess I would say in my short 5 tenure on ACRS, since June of last year, this is the first 6 time we have had the benefit of having with us a resident 7 inspector--ex-resident inspectors in meetings where, in a 8 lot of cases, I thought they could make a contribution.
9 DR. KERR: I always thought that resident 10 inspectors were like Baptists, you know once a Baptist 11 always a Baptist, and resident inspectors must be in the same 12 category.
13 MR. MICHELSON: Okay, I would--
14 MR. CARROLL: I would encourage what is given on
{}
15 other issues, that have had people that have dealt with the 16 real world problems come talk to us.
17 MR. MICHELSON: Charlie, do you have any comments? l 18 MR. WYLIE: No. Just ask, where do we go from 19 here.
20 MR. MICHELSON: I think I would like to thank the 21 staff for this very fine presentation. I think you held up 22 real well.
23 We will have a short subcommittee discussion which 24 need not be on the record.
25 So, at this time, unless there are other important A Heritage Reporting Corporation
(_f (202) 628-4888 1
i
f .c <
4 189
-O 1'- co- =t , z ~111 co 9 1 t th
- 1=9- :
j 1
2' (Whereupon at 3:53' p.m. , the meeting of the ACRS 3; Su$>committeeonMaintenancsPracticesand- Procedures was 4 concluded.) i i
5
'6' )
1 7
8 9
10' 11 12 13:
O" 15 16 17 18 19 20 1
21 ,.
22 23 f
24 25 Heritage Reporting Corporation (202) 628-4888
i '
i 1 CERTIFICATE 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 0N REACTOR SAFEGUARDS 6 , Name- SUBCOMMITTEE ON MAINTANENCE PRACTICES AND !
PROCEDURES 7
8 Docket Number:
9 Place: Bethesda, Maryland 10 Date: March 30, 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 ne 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/ # S UO N 19 (Signature typed) : RASA VON KIPARSKI 20 Official Reporter 21 Heritage Reporting Corporation l 22 i 23 24 25 Heritage Reporting Corporation (202) 628-4888
~
c L1- CERTIFICATE
!(~T.
\-l 2 3 This is to certify that the attached' proceedings before the 4 United States Nuclear Regulatory _ Commission in'the matter of:
5 Name: ACRS subcommittee on maintenance practices and procedures.
6 7 Docket Number:
8 Place: Bethesda, Maryland 9 Date: March 30, 1989 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear 12 Regulatory Commission taken stenographically by me and, 13 thereafter reduced to typewriting by me or under the direction 14 of the court reporting company, and that the transcript is a
/ .
15 true and accurate record of the foregoing proceedings.
16 /S/
sf IRM N L. FENB R e r>t, 17 (signature typed):
18 Official Repetter 19 Heritage Reporting Corporation 20 21 22 23 24 1 25 j O
Heritage Reporting Corporation (202) 628-4888
LO!
s II-FINAL MAINTENANCE RULE MONI DEY, TASK MANAGER i 0FFICE 0F NUCLEAR REGULATORY RESEARCH O PRESENTED TO THE ACRS SUBCOMMITTEE ON MAINTENANCE PRACTICES AND PROCEDURES MARCH 30, 1989 O
l
PURPOSE OF BRIEFING 0 'PRESENT RULEMAKING PACKAGE:
l -
ANALYSIS OF PUBLIC COMMENTS FINAL RULE .;
REGULATORY GUIDE ,
BACKFIT ANALYSIS 0 OBTAIN ACRS COMMENTS PRIOR TO CRGR AND COMMISSION REVIEW.
O REQUEST ACRS LETTER'AT APRIL MEETING.
O 1
0 1 !
q ,
SCHEDULE- {
ACRS FULL COMMITTEE MEETING APRIL 7 l
l CRGR REVIEW APRIL 12. !
l-RULEMAKING PACKAGE DUE-TO COMMISSION APRIL 21 COMMISSION BRIEFING MID-MAY
- O O
2
p, REGULATORY BACKGROUND' V
1980 - 1985 -
REVIEW 0F' STATUS OF MAINTENANCE IN U.S.
-NUCLEAR POWER INDUSTRY, NUREG-1212.
MARCH 1988 -
POLICY STATEMENT ON MAINTENANCE RULEMAKING INITIATED JULY 1988 -
PUBLIC WORKSHOP ON RULEMAKING NOVEMBER 1988 -
NOTICE OF PROPOSED RULEMAKING (NPRM)
O O ,
l 1
l 3 1 l
L-____-___________
].
NOTICE OF PROPOSED RULEMAKING
.i.
O MAINTENANCE DEFINED TO INCLUDE TRADITIONAL ACTIVITIES PLUS SUPPORT FUNCTIONS 0 SEVENTEEN ACTIVITIES LISTED s
O REQUIREMENTS:
ESTABLISH, IMPLEMENT, AND MAINTAIN PROGRAM
'( ) -
ASSESS PROGRAM EFFECTIVENESS AND EXECUTE CORRE0TIVE ACTIONS 0 IMPLEMENTATION:
3 MONTHS FOR PLAN AND SCHEDULE 2 YEARS FOR FULL IMPLEMENTATION O
. 4
e se 8
- e m -
% M
- % 4 ~=.C.$ -
CJ ~
M 4
P'*
e t .
e e.
W W
--8 .
> 33 W %
=n Db o U W U
E 9 . " &- O.
- % L OE W.M E S W
I E
&n.
O E
O e
0-= e 4 G t-- m > O.
O E D M
d 4@ W Q
Jll K M
wW
- g Lf3
.J % 1 eC 3%9" % DC Ch. N 3 *=
E M 6&W Ch. A e=
m C D C % 9 se. D &
- 7 -"* e 44 % S e M C LD M wr- e NEU W
.J e-o E
W M
m &
C UC m
m ,I.Oo e 6.
% D GA C Ch. U Ch.
W .
G G e== a= .
O D m E d5i: o D.
D @ 2K
% eF >= M O m
,A N
DC 3 *=
CL 6 e K O
F-43*
e 4J
~ %.
134 3 &.
J Q. O W D J 9D
- 3 LAs K CC
3- .
i NOTICE OF PROPOSED'RULEMAKING CONT'D.
y-)
'v - )
SOLICITED COMMENTS'IN 12 AREAS:
- 1. NEED AND COMMITMENT FOR MAINTENANCE STANDARD.
-2. LEVEL OF DETAll IN STANDARD.
- 3. TIME NEEDED 10 DEVELOP STANDARD.
- 4. THIRD PARTY CERTIFICATION.
- 5. SCHEDULE FOR REG. GUIDE AND STANDARD.
- 6. BACKFIT REQUIREMENTS.
i
- 7. SCOPE OF RULE.
i
- 8. WORKER ACCOUNTABILITY. ,
- 9. CRITERIA FOR FULLY EFFECTIVE PROGRAM.
- 10. FUTURE PERFORMANCE-BASED RULEMAKING.
- 11. USE OF NPRDS, COMMISSIONER ROBERTS' VIEWS. l
[]) 12.
6
'J FINAL RULE
(~3 _
t/
- D STAFF RECOMMENDS PUBLICATION OF FINAL RULE 0 DEFINITION OF MAINTENANCE AND REQUIREMENTS UNCHANGED:
NONPRESCRIPTIVE PROCESS APPROACH J
0 SCOPE MODIFIED TO:
" MAINTENANCE PROGRAM SHALL INCLUDE ALL SSC'S WHOSE FAILURE COULD SIGNIFICANTLY IMPACT THE SAFETY AND SECURITY OF THE FACILITY."
O O
O IMPLEMENTAT".JN SECTION CHANGED TO INCLUDE TWO OPTIONS:
- 1. INDIVIDUAL LICENSEE CERTIFICATION.
- 2. THIRD PARTY CERTIFICATION WITH MAINTENANCE STANDARD.
NRC APPROVAL I
TWO-YEAR IMPLEMENTATION c:)
7 1
[_ .
=
FINAL RULE (CONT'D)
- 0.
O BASED ON PUBLIC WORKSHOP AND COMMENTS RECEIVED ON NPR, STAFF RECOMMENDS:
THAT INDUSTRY CONTINUE INITIATIVES AND PROVIDE LEADE.RSHIP.
'THE DEVELOPMENT OF AN INDUSTRY-WIDE MAINTENANCE STANDARD.
THE ESTABLISHMENT-0F A THIRD PARTY CERTIFICATION l PROGRAM.
l-) 0 RULE INTENDED TO PROVIDE REGULATORY EMPHASIS. !
t 0 REG. GUIDE DEVELOPED IN ABSENCE OF INDUSTRY COMMITMENT.
1 i
I
([] l 8
.. i l
O III
SUMMARY
OF ANALYSIS OF PUBLIC COMMENTS R. FRAHM, RES O
O 9
7 PUBLIC COMMENTS r)
L >
y COMMENT 0RS PRIOR TO AFTER 02/27/89' 02/27/89
- UTILITIES 19 26 IND. GROUPS 3 1 FEDERAL AGENCY l STATE GROUPS / INDIVIDUALS 2 --
PUBLIC INTEREST GROUPS /INDIV. _9 _4_
34 31 6 0 O FOR RULE AGAINST RULE 28 31- !
'ALREADY EXTENDED PUBLIC COMMENT PERIOD FROM 01/28/89 TO 02/27/89.
O
I 7- -MAINTENANCE RULE PUBLIC COMMENT
SUMMARY
A. NO RULE l
NO DEMONSTRATION THAT RULE INCREASES SAFETY.
- DIVERT. INDUSTRY RESOURCES, MAY HAVE NEGATIVE IMPACT. -
AUTHORITY ALREADY EXIST TO IMPROVE THE FEW POOR PERFORMERS. q
- If1DICATORS ALREADY EXIST (SALP, NRC l!1SPECTORS).
- INDUSTRY HAS BEEN IMPROVING, PROGRAMS ALREADY EXIST.
B. PROPOSED RULE IS UNBOUNDED NO DEFINITION OF B0P.
PUBLIC CANNOT ASSESS IMPAC'l THOUT R.G.
NO DEFlr11T10N OF 10CFR50.65(B) ACTIVITIES.
C. REGULATORY ANALYSIS ,
O -
NUREG 1212 DATA OBSOLETE.
INDUSTRY HAS IMPROVED FROM 1985 TO 1989.
NRC COST ESTIMATES T00 LOW.
NRC' RISK REDUCTION TOO HIGH.
D. ADEQUATE PROTECTION BACKFIT ANALYSIS
- BOP REQUIREMENTS NEW, THEREFORE BACKFIT NEEDED.
NO DEMONSTRATION PLANTS ARE BELOW ADEQUATE PROTECTION >
LEVEL.
INDUSTRY IS IMPROVING.
VERY FEW BAD / PROBLEM PLANTS (5% NOT 30%).
USE PRESENT REGULATORY AUTHORITY TO IMPROVE FEW BAD PL AllTS .
O
_- - _ _ - - - _ - - -- - - - - J
g y. .
L
SUMMARY
OF COMMENTS AND PROPOSED RESPONSE ~TO NPR QUESTIONS li 15 IT APPROPRIATE FOR THE' INDUSTRY TO DEVELOP A MAINTENANCE STANDARD? WILL THEY?
STANDARD NOT NEEDED INP0 GUIDELINES GOOD.
NO COMMITMENT'.FROM INDUSTRY.
- NRC WILL DEVELOP STANDARD IN REG. GUIDE
- 2. LEVEL OF DETAIL
- GENERAL DESCRIPTION, UTILITY IMPLEMENTATION INPO GUIDELINES
-* NRC WILL PROVIDE IN REG. GUIDE i
- 3. TWO YEARS, EN0 UGH TIME 2 YEARS 0.K. TO DEVELOP STANDARD 3 .5 YEARS NEEDED TO IMPLEMENT
- 2 YEARS FROM FINAL R.G. - INDUSTRY ALREADY HAS GUIDELINES, SELFASSESSMENTS, MART, ETC. j
- 4. THIRD PARTY CERTIFICATION NOT NEEDED COULD USE INP0 l
- MOOT POINT SINCE NO INDUSTRY STANDARD !
- 5. ISSUE R.G. WITH FINAL RULE Q -
NEED TO REVIEW R.G. TO ASSESS SCOPE AND IMPACT OF RULE
- REISSUING RULE WITH DRAFT R.G, FOR PUBLIC COMMENT NOT CONSIDERED NECESSARY, SINCE RULE REMAINS GENERAL.
L c jm 6. BACKFIT ANALYSIS V
NO DEMONSTRATED INADEQUATE PROTECTION INCREASED REQUIREMENTS WITH B0P
- JUSTIFICATION OF RULE ON THE BASIS OF ENHANCED SAFETY BEING PURSUED.
- 7. INCLUSION OF B0P-UNBOUNDED SCOPE GOING BEYOND REGULATORY AUTHORITY WILL DIVERT RESOURCES FROM SAFETY RELATED SSC-
- 8. INDIVIDUAL WORKER ACCOUNTABILITY ;
l LICENSEE ACCOUNTABLE WORKER ACCOUNTABLE 't0 LICENSEES
- NRC ADDRESSES ISSUE ONLY IN A BROAD SENSE IN THr R.G.
i 9, CRITERIA FOR EFFECTIVE MAINTENANCE ,
- SALP, REGULATORY INSPECTIONS EXIST I
- NRC WILL NOT DEFINE SPECIFIC QUANTITATIVE CRITERIA HOWEVER, R.G. IS BASICALLY PERFORMANCE BASED.
4 L
- 10. MAINTENANCE PERFORMANCE INDICATORS ;
NO ONE SET DESCRIBES MAINTENANCE. EFFECTIVENESS PROCESS INDICATORS HAVE SOME USE l
O
- MPIS USED IN CONJUNCTION WITH OTHER FACTORS SHOULD BE USED TO ASSESS MAINTENANCE l
g co ,
rs-
.. J l .1 11 -. NPRDS th.
DOES NOT INCLUDE B0P EQUIPMENT-FAILURE RATE DATA ONLY NOT GENERALLY GOOD FOR MAINTENANCE USE ENCOURAGED IN R.G..
12, COMMISSIONER ROBERTS VIEWS
~~ MOST AGREED TWO DID NOT AGREE
-O O
t 4
S T
N
) .
S E EA CL R
C H NP (
A E T A S D
E N R D R
I E E AE U R UC G U TW GN S NO E A I P F N Y AE R
NA ST 9 O EMR :
O RI OA N 8 9
A T TM 1 T O FE C D V A T OL D
E T AN 0 L N EO 3 C E R U S S U S E NE E H C
G T R NESN P OT T R E I A EM M R ENR EM T
T MEO T ~s I
r F EVF MB MU A RI T OS I S C R UC M Y D QEF A R O
E S F R R EG I
V ,
D O A R
O P l
, l: !l J_
1
^
o E
D I
U G _
Y E S R C _
E I O N T A T A I
V L N
I E T U T _
C G N A E I R A o E M -
C E N H F A T O N F T E O C T J N
I 4 .
I D _
A C N-M O _
N C _
O __
I -
T _
C -
o E S -
!llllll
T R
O I '
P S.
P E T E UT U E SN ,B S E EI N N,M O E T CR OE G N T C I
I T E AN L C AN E
S NO C E R EA N TT O R
I M E NA F D C I R I AH E EO S L MT : N VI S E I
I N E F S SS L EE C D S OE I E EY O CS E O I
US TC DT I I RM R GEC TI CV O ULI UT RP I
O P SO GB T
S R A NR DC AX I
U F S OP NA O
E SL E
EIV
_ I MY I TE C HT AF T CH EI E E T C SS CR C F 4. T E EW NO N C EF SF VO S H AH A S RL NT N UI OL NBE EL EU I
E O S A RN
_ W TA T I VA B I TCA E
I N
I D N I ET CS V AN A RS EEO MA M PE SDT ee R
E V
O O
l
E C
N G A N I
i '
4 N N E I A
T N R I T A
M D N
F A O
N T O S N I T N E A O M C I
T E I A
G F I Z A L I N A N A U A M Q G G R S E N O D N
L E
O
' R I
L E S
S A N T U U C E
_ N R D D N C N O O E E A O E O S P C H N R C
I T R P O C E P N A E U R S T G A Z P S P -
N L N I D I O
N I
E N E E E N A R P T A C C C A M T E N G N N N N E I R A A A G F O K A O N N N N O C D M E E E I N S R T T T T K O F N N N N N E O A I I I A P R C L A A A L Y O E T P M M M P T W R C
U E D ,
N N 1 2 3 4 5 6 7 8 I O 4 4 4 L C 4 4 4 4 4 T
O U O
4
E C G N
_ A N I i ' 4 N N
_ E I A
TN NO R I I T AT Y MA T D
N G Z
I FI L A N ON A I
B I
N A N TG T N O I
A O I NR Y N I T R
A C
T E DRO ML I B G NOC RA F O A AHC TA ON AS I
L A
J N FR E A TU NE I
RC E U H M ENA D N I T ER Q T O MDA FX TU -
D EE OE I
A GN Y T CS NIT O AIFIL WD E NAD NN EI ON ER OTN O MA DB L A C SI I I NF L NE RDA T Y S O HA AT E E A ELN I N TA P R M Z CRO I
TT AO R PU CO NNAP EQ EC O AAES CO ET I G CE C ARS GNLE NMN N AA CD NA OS RECR OT U SI I F N P A M
N M F
A EI CL N
TIA O T T N ee ee A Mee C S I N
L A P M 'e e 1
2 3 1 1 1 1 2 O .
44 4
4 4
'S E
C I
V , ,
R E
S E'
C N .
A _
N E T T S N N
I E
I E
E A E T M C M C I
L P S I N E N I C U A D A Q NE R E A ECU T E N
E F E TNI L C R T D T NNAA A R
U S ND EI N E S OIA NF T S S I AE A S B L E I
E O T ME T TM TF N N P I S EA O A ZFI N O O C DI ET O X E
G N
V I
F N L L BI L D N
I N OAS A I T C ON O
O AA A A T MIS I
DF D L N A T O T MVA G NIL N A N TU N R RF Y L A AE C A E CD S.NS S O OP OO A S UM I
G L P
C UE T ENT I T
PT N R QM O N DS RMOR A T L A NA AN CA R RU R A O D O C O D
N N OB POR F P B OSA PI N D NE I
A EC Y I P N D T FI OD L N A R NNOC LD E A OV L N A 5 I R G N N O A C U AE VDR N I
A E SGRV F EO FS I MSU TE RS F NGL O ND OL N UQ N O EI EO I AN AA I
FCE E E TL C R TV L RE L- OR OOR M AOO FNO L N E NN O TV OGT FF E SCT O A EI I R R C G R NN I T T Q A T EG N N / N N T N oe O O ee A A O NE I C e e C Qee Meoo C A .
M 1 2 3 4 5 6 O
33 3 3 3 3 3 44 4 4 4 4 4
~
o D R E E L N L N i ,
e O A R M T
N Y O L
. C E S M U E D I T
T, T N P
A A A D S O
D N NS T T E V
I OE S I
P O DT EAAU GE R
U P H N TU I
Q E
E P SDO I R T
A LR GPOD K o SA ER DD RU TT EE NMOE R L O CCR I
O .
E ANF F EA ED A C SIL HE D O OY E A CC YN G R LP LR V P A ,
SNITA AL N I
E CYLED D NI DR 4
I N E BE O CNR F I A T LANT NHAI GINIAANIN m
A CER NAVLO NELE _
ETCV T
I N MAPM -
N N I A -
A L m Meee P eeee '
o 4 4
5 4
I1!i l
S E
S SD C ,
4 AA S- G R
I T
S E P I
I D
E R TU I
E V L T I D O
U C T TN D
E A
N C A A R I H
C A
H E SN E
N C O S C O NI C I D I E T AT N
V R Z A NA A E Y C I
EC I S
L F T F D A I NI E N D I D
N O
T A O AO N E M MM D C O A E T E'T E
C L S C N EN N A
T TN P ES I N EE CA N A MA A N
A MS N D LN L L AD L AY RA ENLUB N PUA A I O P GU RP P CELL I
NN E EI T NTAI D EU T QC E
TR V C R E FO F ANC FA E N EE NT UL ORDO NIAT I
T I PN E M AT N A N I AR
. SL N ATL T MN E M RMEI E C O OE N EE EM E OR MO C
I T CWU AVTVU I
V TE EI T NA A NE SE I SEC I ST VN AT RAVR MITYRO NSPD TEE CRD I
T O NA ONE P TERA
T R N OV R D N OI S E R E I CA E R O R e A eee C ee P Me NMe PP I Y -
T 1 2 3 4 5
- O 66 6 6 4
6 4
6 4
44 4
1 -. : ,' i l,
,i ll 1 iI llll1l!b G _
S E
I C
NN G EO I I N D C TA I
T E N
O I FR S E
I A
F EOP N EA C A DTT O N D N AN RIO NSE ODE PTAET T R A RN G TR LCG UIA NEN I EE U OI RIFI NNEDM PSM C TT NN N CT ENUP E E OE A OS LRO KC C Z._
DI O CD I PPP R LDK K O R C -
O E -
Weoee Reee - ._
8 O 7 4 4 l l1'
1
.. 1
.,a i g l
_a IV AND VI DRAF'i MAINTENANCE REGULATORY GUIDE I 1
MONI DEY, TASK MANAGER OFFICE OF NUCLEAR REGULATORY RESEARCH O PRESENTED TO THE ACRS SUBCOMMITTEE ON MAINTENANCE PRACTICES AND PROCEDURES MARCH 30, 1989 4
O i
1 !
4.'
4 e
e REGULATORY GOIDE OUTLINE-REGULATORY GUIDE SECTIONS A. INTRODUCTION B. DISCUSSION i
- C. - REGULATORY POSITION
- 1.
SUMMARY
OF EFFECTIVE MAINTENANCE PROGRAM
- 2. OVERALL MAINTENANCE POLICY
- 3. ESTABLISHMENT OF GOALS AND OBJECTIVES
- 4. CONDUCT 0F MAINTENANCE
- 5. EFFECTIVENESS MONITORING AND ASSESSMENT
. 6. FEEDBACK AND CORRECTIVE ACTION D. IMPLEMENTATION
/
t.
O 1
2
. _ _ _ _ = _ _ _ __ w
.. 1 i
APPROACH / PURPOSE i
0 PROVIDE BASIC GUIDANCE AND CRITERIA FOR MAINTENANCE PROGRAM 0 ADOPT PROCESS' APPROACH
'O ALLOWS FLEXIBILITY FOR SPECIFIC ACTIVITIES 0 MINIMAL IMPACT ON EXISTING GOOD MAINTENANCE PROGRAMS 0 ALLOWS USE OF INDUSTRY GUIDES AND PROGRAMS FOR DETAILS O
O 3
- 1 1
O EFFECTIVE MAINTENANCE PROCESS V
l
) DEFINE OVERALL MAINTENANCE POLICY I
- y
) DEVELOP MAINTENANCE OBJECTIVES V
' DERIVE REQUIRED SSC MAINTENANCE ACTIVITIES 1 r ,
>' CONDUCT MAlf;TENANCE l
EVALUATION, FEEDBACK & 4 MONITOR EFFECTIVENESS CORRECTIVE ACTIONS l
O 4
CRITERIA FOR ACCEPTABILITY
)
- 0. COMPLIANCE WITH RULE VERIFIED BY NRC AUDIT AND INSPECTIONS 0 -F0LLOWING ONGOING STEPS MUST BE ADEQUATELY EXECUTED:-
- PROGRAM' ESTABLISHED, DOCUMENTED, AND IMPLEMENTED BASED ON SSC PERFORMANCE OBJECTIVES;
- PROGRAM EFFECTIVENESS MONITORED AND DOCUMENTED; AND i
- CORRECTIVE ACTIONS EXECUTED AND DOCUMENTED, O. CRITERIA NOT BASED ON THRESHOLD OF EQUIPMENT PERFORMANCE.
l O
5
-- -____________-_-____-_- __ _ 1
9 9
OVERALL MAINTENANCE POLICY.
O REQUIRES SUPPORT / INVOLVEMENT OF ENTIRE ORGANIZATION O PLAN AND ESTABLISH PROACTIVE MAINTENANCE PROGRAM 0 REQUIRES COMMUNICATION TO AND TRAINING OF ALL STAFF O
l O
6
_____m___ _ _ . . - _
SCOPE d
THE MAINTENANCE PROGRAh SHOULD COVER THOSE STRUCTURES, SYSTEMS, AND COMPONENTS (AND THEIR SUPPORTING SYSTEMS):
(A) RELIED UPON FOR THE INTEGRITY OF THE REACTOR COOLANT L PRESSURE BOUNDARY, SAFE SHUTDOWN CAPABILITY, AND ACCIDENT PREVENTION AND MITIGATION; (B) WHOSE FAILURE CAN CAUSE A TRANSIENT OR ACCIDENT THAT CHALLENGES STRUCTURES, SYSTEMS, AND COMPONENTS RELIED UPON FOR THE INTEGRITY OF THE REACTOR COOLANT PRESSURE BOUNDARY, SAFE SHUTDOWN, OR MITIGATIONJ A
(/ (C) AND OTHER SSC'S THAT PROVIDE REASONABLE ASSURANCE THAT THE FACILITY CAN BE OPERATED WITHOUT UNDUE RISK TO PUBLIC HEALTH AND SAFETY AND COMMON DEFENSE AND SECURITY (E.G., FIRE PROTECTION, SECURITY / SAFEGUARDS, EMERGENCY PREPAREDNESS, POSTACCIDENT MONITORING),
'D) WHOSE FAILURE COULD INADVERTENTLY DISTRACT THE ATTENTION OF PLANT OPERAiORS; AND (E) REQUIRED FOR THE PROTECTION OF THE OCCUPATIONAL HEALTH AND SAFETY OF WORKERS AT THE PLANT FROM HAZARDS ASSOCIATED WITH RADI0 ACTIVE MATERIALS.
7
I 1
l O. ESTABLISHMENT OF GOALSLAND OBJECTIVES 0 GENERAL OBJECTIVE IS TO PREVENT THE FAILURE 'RO DEGRADATION OF SSC'S
'O SPECIFIC OBJECTIVES BASED ON DETERMINISTIC EVALUATION OF SSC FUNCTIONS AND SAFETY SIGNIFICANCE <
0 HIERARCHICAL GOALS BASED ON SYSTEMATIC EVALUATION AND IMPACT ON SAFETY 0 ALLOWS CONSIDERATION OF:
O SYSTEM FUNCTION EQUIPMENT REDUNDANCY DIVERSITY OPERATING MODE PLANT CONDITION SAFETY RELATIONSHIP O QUANTITATIVE GOALS CONSISTENT WITH TOP PERFORMERS .
U 8
l
EFFECTIVENESS MONITORING AND ASSESSMENT bs A. MAINTENANCE PERFORMANCE INDICATORS 1
i 0 A MAINTENANCE PROGRAM SHOULD: l 1
l
- ESTABLISH G0ALS, OBJECTIVES, AND USE INDICATORS FOR !
TRACKING PERFORMANCE i INCLUDE DEFINITIONS FOR QUANTITIES USED IN INDICATORS 0 THREE TYPES OF INDICATORS: ,
( -
OVERALL PLANT INDICATORS PROCESS INDICATORS EFFECTIVENESS INDICATORS 0 USE OF NPRDS ENCOURAGED B. MANAGEMENT OVERSIGHT AND SELF-ASSESSMENT 0 SAMPLING MAINTENANCE ACTIVITIES 0 TRENDS EXAMINED I
($) .
9 i
j
s FEEDBACK AND CORRECTIVE ACTIONS FEEDBACK AND-CORRECTIVE' ACTIONS BASED ON EFFECTIVENESS MONITORING.
AND ASSESSMEIT FEEDBACK FEEDBACK TO BE USED TO DETERMINE NEED FOR CORRECTIVE ACTIONS MAINTENANCE PROGRAM SHOULD DEFINE THOSE RESPONSIBLE FOR FEEDBACK INFORMATION AND THE CHANNELS OF COMMUNICATION CORRECTIVE ACTION DIRECTED TO ENSURE DEFICIENCIES CORRECTED CORRECTIVE ACTIONS CORRECTIVE ACTION PROCESS INCLUDES:
DETERMINE CAUSE OF DEFICIENCY PROVIDE TIMELY ACTION DOCUMENTATION OF CORRECTIVE ACTION O
10
_ _ _ _ _ _ _ _ - _ _ _ - _ - _ _ - --- -- 1
g FEEDBACK AND CORRECTIVE ACTIONS (CONTINUED')
(
TIMELINESS
- l ANALYSIS OF DEFICIENCIES SHOULD BE DONE IN TIMELY MANNER PROGRAM-SHOULD DESCRIBE PROCESS FOR TIMELY FEEDBACK AND IMPLEMENTATION MANAGEMENT' INVOLVEMENT
'FOR RECURRENT ISSUES, DEPARTMENT AND CORPORATE MANAGEMENT SHOULD BE INVOLVED l
l l
O 11 L _ . _____ ___ _ _-_ _ _ _ L
.)
, i Oi VI1 a
)
PUBLIC COMMENTS ON REGULATORY AND BACKFIT ANALYSIS T. KING B. RICHTER i i
i o
0FFICE OF NUCLEAR REGULATORY RESEARCH-i PRESENTED TO THE ACRS SUBCOMMITTEE i
ON MAINTENANCE PROCEDUKES AND PRACTICES MARCH 30, 1989 i
O oL--_- _____ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
1 l
SUPPORTING BASIS FOR RULE AND REG. GUIDE ;
O ;
O JUSTIFIED ON THE BASIS OF ENHANCED SAFETY:
QUALITATIVE BASIS QUANTITATIVE BASIS (REGULATORY IMPACT ANALYSIS) 0 MAfiY COMMENTS RECEIVED ON PREVIOUS REGULATORY IMPACT ANALYSIS (RIA): :
1 RIA CURRENTLY BEING REVISED
SUMMARY
OF MAJOR COMMENTS / REVISIONS BEING CONSIDERED l O.
I h
O-
- QUALITATIVE BASIS 0 MAINTENANCE HAS A DIRECT IMPACT ON PLANT SAFETY:
RELIABILITY OF SAFETY SYSTEMS CHALLENGE TO SAFETY SYSTEMS AND OPERATORS (BOP).
THEREFORE, IT IS REASONABLE AND PROPER FOR NRC TO REGULATE IN >
THE MAINTENANCE AREA IN A FASHION THAT WILL RESULT IN AN IMPROVEMENT IN SAFETY.
O THERE IS A VARIATION ACROSS THE INDUSTRY IN MAINTENANCE.
i MAINTENANCE TEAM INSPECTION RESULTS TO DATE (FROM'18 SITES /27 PLANTS) CONFIRM THIS VARI ATION: l 44% - GOOD 50% - FAIR 6% - POOR O THEREFORE, THE RULE WILL FACILITATE THE COMMISSION'S TAKING :
ACTION IN THE MAINTENANCE AREA TO ACHIEVE IMPROVEMENT.
0 OVER THE REMAlHING LIFE OF THE PLANTS, A REGULATION WILL STABILIZE ANb BETTER DEFINE REQUIREMENTS TO:
HELP ENSURE GOOD MAlHTENANCE PRACTICES ARE ACHIEVED AND SUSTAINED ACROSS THE INDUSTRY HELP ENSURE THE EFFECTS OF AGING ARE FACTORED INTO PLANT MAINTENANCE PROGRAMS 0 IT IS COST BENEFICIAL TO HAVE A GOOD MAINTENANCE PROGRAM (1.E. SOME PLANTS ARE ALREADY DOING IT).
O
I f- QUANTITATIVE BASIS d
0 REVISION OF RIA UNDERWAY i.
O GOOD MAINTENANCE EQUATES TO REDUCED RISK:
- WITH RULE AND R.G., P00R AND FAIR PERFORMERS SHOULD IMPROVE.
SHOULD BE COST BENEFICIAL TO ACHIEVE A GOOD PERFORMANCE LEVEL.
O GOOD MAINTENANCE EQUATES TO NET COST SAVINGS:
COST TO IMPROVE - AREAS NEEDING IMPROVEMENT BASED UPON MAINTENANCE TEAM INSPECTIONS.
COST SAVINGS PUE TO IMPROVEMENT - AMOUNT OF
[]) IMPROVEMENT BASED UPON HISTORICAL DATA.
O l
A _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ . _ . _ . _ _ . _ _ _ _ . _ _ _ . . _ . _ _ _ _ _
N I
A H T C E R I B E H C W Y A
N U
, M D 2 D Y S E 1 E N L T V 2 T O H N I 1 A I G E E - M T I M C G I C H M E E T U O R R S D E C : U E E R O F 0
E R
E D
E D
N N
R E
D R
K S
A S
N W U O D E
N U O Y L I R D R S C E T I A T N C A E E T N D B R T P M N I O A M M E A T U M S I U Y M U S M N L O A S I S O R E M T T S C E R E I S A C - B S F E E N R T E R Y V O E Y S N E N I C V A O E V A S 0 M C B O M R - -
N - -
E O T J X A E M O
I 1 1
N O
I T
C E E D P I S U N Y G I R
O Y M T R A A O E IS T T 7 0 36 LI A 5 GS L E S EY U C I RL G N S A E A Y DN R !
L A
EA S T D
T N I F N A V A I E R A M
o Y R D R
O T
A L
U G
E R
Y N R 2 2 I O 1 1 T 2 2 S AS 1 1 0
E LI - -
G US G G 0 0 N GY E E 3 - 1 A EL P R - 0 -
RA U U 0 - 0 W TA N N N 2 9 F
A R
D S
T N
S FE I OM S E A RV B EO 'S S BR R R S
T NP LP E
M E N NI R R E O O M RG F F o E E
L T
S ON FI SI I U SQ R P P R
E R
O R
E R
I D
O O AE O A O C BR P F G
A T
A E D G
N T A N R E C Y D E B A R O S R E T B R N O O A T A M L P
F E G L T N F E A I O S C R T I E S I D D R N I E S D I S B I N N M S E N O U Y L A C N L C O A N
A Y
L I
S N
S T
L E
H T
D O U K A I S N S E T E I I R A R R L G T U E N Y O C H I R N L T R O A R G S C G U E E -
N C T O S E I C A D E V N O C U I I E Q T M E E C I I A R C N N S S X A N A H N E E A N C E U S M E E S L S E R T T - A I G O N K V N F I A 3 C A R A R i F R H E M P R O N C P O
S R
E M
R S O A T F T N S R A E R E D M E P E M 2 L R R 1 E O 2 I
F A 1 E R F -
D E G I P F E U O R
_ G R U N
O b' S Y O O I R P I T S O S S O
Y T R U A E L P
- 0 L A
N A
L U
G E
W E
F N
I C
N O
T R M M D S O O E O W R F
R F
S A
C O L B L G G O N N D F I I E T T S O L L I T U U V S S E D E E R E R R R G U S S N T E E I C S S E U A A B R E E T R R A S C C T E E N A R D I D O'
f ; I
S R
E H
S T
. N O O
I H T T P I W
M U
S N S
I D A A
E M O
F W E O O R R
S R L E
I T
N A
N E
L I
T S
I R A E S T W E R I E S T C N N N O I U I A T T P R M E U C S N S U A
E E G M R O A
_ S L O
1