ML20199B090

Transcript of ACRS Subcommittee on Reliability & PRA Meeting on 971113 in Rockville,Md.Pp 310-422.Viewgraphs Encl

ML20199B090
Person / Time
Issue date:	11/13/1997
From:	Advisory Committee on Reactor Safeguards
To:
References
ACRS-T-3016, NUDOCS 9711180208
Download: ML20199B090 (157)

Text

,

y

' g o;. g..

i e' '

"

• f.

,,,, (. '..,,< 4.,.,.

~

i OV?

.~ '

. gl ^

'SQ

Q:

}Mk %C ' -

• ~,

?.:5, :

l:::. N

,..t Q :.Q,

'f e

W g:

W.

e 2 :..N J !U.W:,%

i,p V Y

.Q" k M,.%((i

'. g -

z=

g.-

n %:Q

'.l*..

l%

1

'y

<. h;;5.h,'.hs.g. d

' i

'I l 4.

. : 'r ' - #

e e

s...

t

~, z.,..

,.. ['.h, g.ag[!ftl.f

fh,. N '. L.,

. ll;. 's ..,t . p-4 y, /, ; S pl. ; /l 6 U.i,* - l g 3 ', },'N . ')"' gy .,.,y"%g 3.; 2 m e . y.p. ?. f e a g,,. ,b. ' - ~ '

p;_ g.

V ~ ,. y_: q. t a,. c f,,, "'* (.p2(* 9:. yg?, .I *.N. .S .s N a.

1. f

'.. h. y h' . : h.:. l . y . u.6 f. ' Msw w'. m ;j ~ ~ w: 6.k 'I

• l',

'O .q. e ,.,,.,. p.. n c.., - '..r < w ^,:,:.D 'X._.

• p.w.4 M..u i- -

h 1:,t 'f k' ' ", - i, ?

• +.

y'. I. ' I, h 5 .:'} 3' ~ % -r q ". &_i. ;;.f g;;;g, ;;g&. R f_.:q ? 3. 3 . QA.f.w:.. l -l b, ;f

w. ~.Y4 '.i c y

• O

, t.1 ;p,y h:f.w'g; 'g $f b I I . ( 4.'. ~ }U g O[' .W):':0 Y ':, y 't ts < , 4.12 - ' i. e 24g$[%,.tm ;; 3 I E . ~ A v. w f;f M: n :.1., .' 'k._.,' D.r .s.... ,.v $ ' Y'. s ~ I - fl.; ~

y shb..w.nl 9~.:,' q;' +.. ).p ' ' '

• m

~. L F. -~.. ':. a.;. W:n y: y 'b M'. A.a. v, ~. . ' $w. .N N$..w 7 4 ':l An .3 pq& a q;- 4

• A '4pf~ :

..r

p" og o

v i.

'. s.n, ? Q W [ J,. s,1. s t \\ ,,. E. ::l,r .: :n QyQ;& a,.,; s [

(. f

~ g .g

6*

_t f',ht.' +.'e t ..g 3 ff p $p .l, ',f ' - . '. f..:.. 4 2 . t.s

i.l_

'

W.g

{, @.

...4.. .,.q..: .'f. .'3j j., V. g.,q., 7 jg? . t. A- .kj a.y.k ' i 5 f .}phj,i '.'R ~ x ... y. .i: d N'. .-w*: .e .%}c[},. :%. . i . ' ' '. 3g 3!

4

., Q, .h'.$ h k ' [, '- t .;. [.I.

h..f:

.s f. ' '. 8.. Y.:.Qg:, :, {} k ? ' ~~ ,E'- A.'.*.- n.Q, .A

a.. s,.

. yy % +. ..s ej s. c .g ~ A ww a v 3, B.,;..EEBWg . yegy&u n n 'O# .'. J

n.

~.

~ x h.g s< a:; s v r .. ny[%%v:i'.[?: l.# Nt Y. ,4 +.: c.- v

+

cx. ~J

c.<;:t.

-'. 'O y .g . y;k. - 7 s '. p v. ...g;; I 'y'W j.U.

• '[..

N 'I4. 't h ,w:. $. d'.: V,. f '.h' I :. ' -.aq_. I 45 '..- -wc y 47. p. +- ,o ..a. s

.. - -a;

.. :. q.,1 - ;.;- l jc g } }.:. /..'.} s g v-n.c i ,.,;,y 9: . ~. .....s.. : -u... a:. y . ~. ' i.{ "ry a,:. [.. L l. l 't' _ ; j;l L, _ ' ' ' ' '... lhf .p.

, W. :::.

. e: e - ~ ., a. ;.,. I,, .y - 1. .m., ;., g..

.g, ;

cg .. ': ;.: L .' : o ' 7 MQ 4 i 'V. Qv.* @p s,- N'-, _q ' ' [ ' [ t._ J Q ~. ,~... - ~ * .+

• -Q. i. '

AS,. n, a = 4p .?

\\ '. _

y, .l ' [ ' .? ;. :t, k.c:.i.;,.C :.'i ? '. e,:.,* ' i : %.'i.?'.'i., ';:;.l%,WQ >? Q's. 'R ..,R _ y:'c.J ;. G. ;t., ' Q;' T.fy.p %. [ ' -R g._. _ ^

q,7,fy.,

. c.

u R,a \\ A _ /CSST-So /6 OFFICIAL TRANSCRIPT OF PROCEEDINGS g, -x}, NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS

Title:

ACRS SUBCOMMITTEE RELIABILITY AND PROBABILISTIC RISK ASSESSMENT

s 4 A ss s,. s A..

F."en.TE y n Docket No.: J - d

x;: s -

Work Order No.: ASB-300-40 a. 9711180209 971113 PDR ACRS T-3016 PDH !! l LOCATION: Rockville,hlaryland DATE: Thursday, November 13,1997 PAGES:310 - 422 ANN RILEY & ASSOCIATES, LTD. ' e\\ 1250 I Street, NW, Suite 300 .( Washington, D.C. 20005 I (202) 842-0034 L'v 2 / Ji k O

O O'

f} .}_1 W1i8L!ieOjo.aCg;gajpgg.

_._..,m.

f"~y

!(/ .t

..u..

^ ~ ~ DISCLAIMER f UNITED-STATES: NUCLEAR REGUIATORY COMMISSION' S-ADVISORY COMMITTEE ON REACTOR SAFEGUARDS P E ' NOVEMBER 13, 1997 i The contents of-this transcript of the proceeding-Eof the United States Nuclear-Regulatory Commission Advisory ~ Committee--on Reactor Safeguards, taken on. November 13, 1997, T as reported herein,'is a record of the discussions recorded at the-meeting held on the above date. This transcript had not been reviewed. corrected and' edited and'it may contain inaccuracies. L 4 .h 4 e ~ ,e ~ ,,.c,--...--,--.,-

~ f 310 UNITED STATES-NUCLEAR' REGULATORY COMMISSION 1 -(f

2-ADVISORY-COMMITTEE ON' REACTOR-SAFEGUARDS.

u i ***-

4 ACRS SUBCOMMITTEE i S RELIABILITY AND PROBABILISTIC RISK ASSESSMENT: 6 :. 7 8 U.S. Nuclear

• Regulatory Commission 9'

Two White Flint North,-Room 2B-3 -10 11545-Rockville Pike 11 Rockville, MD 20852-2738 -12 13 Thursday, November 13, 1997 14-3 - (V-- c15 The subcommittee met pursuant to notice at 8:30 16 a.m. .17 18-MEMBERS PRESENT: 19 GEORGE APOSTOLAKIS, Chairman, Acrts 20-MARIO H. FONTANA, Member, ACRS 21 ROBERT L. SEALE, Member, ACRS 22 THOMAS _S. KRESS, Member, ACRS 23' . RICHARD = SHERRY,. Senior Fellow, ACRS E

24 2 5

ANN RILEY'& ASSOCIATES, LTD.

' urt Reporters 1250-I

eet, N.W.,

Suite 300 Wak :igton, D.C. 20005-(202) 842-0034 7 --w - + - - - -, r --s-

? 311 -i l :.~ , STAFF AND PRESENTERS SEATED'AT THE COMMISSION TABLE: ((

2'

. MICHAEL;T.;MARKLEY,' Staff, ACRS-- -' 3 - , THOMAS-KING, Staff 4

GARY HOLAHAN,-Staff 5'

GARETH PARRY, Staff 6.: . MARK - C'JNNINGHAM, -~ S taif f .7 JOSEPH MURPHY,- RES '8 L JBIFF BRADLEY-10 ill'. i 12; -14 l' 15 16'

17. '

,18 19' 20 .21 .22: 4.. 23-24- ~ .25 ANN RILEY-&. ASSOCIATES,-LTD, . ~- Court-Reporters 1250 I Street,--N.W.,--Suite 300 Washington, D.C. 20005 (202)-842-0034

312 1 PROCEEDINGS h 2 [8:30 a.m.) 3 CHAIRMAN APOSTOLAKIS: Good morning. The meeting 4 will now come to order. 5 This is the second day of the meeting of the ACRS 6 Subcommittee on Reliability and Probabilistic Risk 7 Assessment. I am George Apostolakis, chairman of the 8 subcommittee. 9 ACRS members in attendance are Mario Fontana, Tom 10 Kress, and Robert Seale. 11 ACRS Senior Fellow in attendance is Richard 12 Sherry. 13 The purpose of this meeting is to continue review 14 of the proposed final Standard Review Plan, Chapter 19, and 15 associated Regulatory Guide DG-1061, general guidance for 16 risk-informed performance based regulation and the use of 17 uncertainty versus point values in the PRA-related 18 decision-making process. 19 The subcommittee will gather information, analyze 20 relevant issues and facts, and formulate proposed positions 21 and actions as appropriate for deliberation by the full 22 Committee. 23 Michael T. Markley is t.he cognizant ACRS Staff 24 Engineer for this meeting. 25 The rules for participation in today's meeting h ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

313 1 have been announced as part of the notice of this meeting [ )I 2 previously published in the Federal Register on October v, 3L 31st, 1997. 4 A transcript of the meeting has been kept and will 5 -be made available as stated in the Federal Register notice. 6 It is requested that speakers first identify 7 themselves and speak with sufficient clarity and volume so 8 that they can be readily heard. 9 We have received no written comments or requests 10 for time,to make oral statements from members of the public. 11 We will now proceed with the meeting, and I call 12 upon Mr. Murphy to talk to us about performance-based 13 regulation. 14 MR. MURPHY: Thank you. (.ss) 15 Good morning. 16 CHAIRMAN APOSTOLAKIS: Good morning. I think it 17 is a very thin package here. 18 MR. MURPHY: But I will -- 19 CHAIRMAN APOSTOLAKIS: Take an hour. 20 MR. MURPHY: As much time as you want. 21 CHAIRMAN APOSTOLAKIS: Well, you have until 9:30. 22 MR. MURPHY:

Okay, I remember the day I came down 23 here with two viewgraphs and talked for an hour and a 24 half -- hopefully this won't take that long.

25 What I am talking-about is the. Commission SRM that r^\\ -Q ANN-RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C.-20005 (202) 842-0034

~314 11 came'to theLStaff some time.ago, and_ asked us to consider () -performance-based. initiatives that do'not explicitly 3 reference criteria derived from PRA-insights-and they'said 4 this shall not be excluded from consideration. 5 CHAIRMAN APOSTOLAKIS: Do you have the. actual SRM 6 somewhere? 7 MR. MURPHY: I have the words of it but I don't 8 have the' actual SRM with me. 9 DR.-SEALE: Do you have the number? 10 MR. MURPHY: Anybody have the number? Use the; 11 mike. 12 MR. KADAMBI: This is Prasad Kadambi with the 13-Office of Research. The SRM is the one associated with SECY -14 96-218 dated January 22nd, 1997. 15 CHAIRMAN APOSTOLAKIS: Is there a short paragraph 15 there you can read? 17 MR. KADAMBI: Yes. I'll read the first paragraph 18 from this SRM, which is entitled, "The Role of 19 Performance-Based Regulation in the PRA Implementation 20 Plan" - "The Commission has approved Alternative 1 with 21 respect to the role of performanco-based regulation, but 22 applications of performance-based approaches should not be 23-limited-to risk-informed-initiatives. Thus, the Commission-24 also approves elements-of Alternative 3 as follows. 25' Performance-based: initiatives that do not explicitly 4 ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite--300 Washington, D.C. 20005 (202) 842-00341 .._~

315 1 reference criteria derived from PRA insights'should not be ( J-2 excluded from. consideration.. The Staff should-include in 3-the PRA: Implementation Plan or-in a separate plan how these' 4- -performance-based initiatives will be phased into the 5~ .overall regulatory improvement and oversight program. As 6-part of the PRA Implementation Plan or its separate' plan, 7 the Staff should include its plan to solicit inputifrom 8 industry or develop on its own additional performance-based 9 -objectives which are not amenable to_probabilistic risk' 10 analysis but could be ranked according to, for example, a 11 relative hazards analysis and phase in these initiatives." 12-CHAIRMAN APOSTOLAKIS: So what are you talking 13 about today? Non-PRA? 14 MR, MURPHY: For want of a better way and a I 15 shorthand way of explaining it, the non-risk informed 16 performance-based regulation. 17 CHAIRMAN APOSTOLAKIS: So we have solves the issue 18 of risk-informed performance criteria and now we are talking 19 about non-risk informed? Is that -- 20 MR. MURPHY: It is what do you do when you can't 21 use risk analysis to give you the insights needed, or at =22 least to give you the quantitative measures needed to use, 23 as we have been discussing in your meetings. 24 CHAIRMAN APOSTOLAKIS: But are you also f.ealing -25: with the= issue of how does-one determine performance -.( ) ANN RILEY & ASSOCIATES, LTD. Court Reporters.

1250 I Street, N.W.,

Suite 300 Washington, D.C. 20005 (202) 842-0034

r316 Lit criteria.using risk information? 2" - MR.' MURPHY :: This is: focusing on_-where you don't' 3- --have risk.information.. 4 CHAIRMAN APOSTOLAKIS' Yes,_ILunderstand that,-but 5- =I mean who_is working on criteria 1when you_have risk '6- =-inf o rmation? - Is anybody working on that? '7 MR. MURPHf: _ I p'. resume my colleagues willLbe doing 8 risk-informed performance-based regulation but I'll let them 9" ~ speak for1 themselves, t -10 -CHAIRMAN APOSTOLAKIS: Well -- -11 MR.- KING: Repear the question? We were having a 12. sidebar conversation over here. 13-CHAIRMAN APOSTOLAKIS: Joe is doing what this SRM 14 instructed him to do, namely address the issue of how does l I 15 ~ one determine performance criteria in cases where risk 16 information cannot be used for some reason. 17 So the question is who is looking into the 18 determination of these criteria when risk information can 19 actually be used -- is it you? 20 MR. KING:

Yes, I think that-is what we are doing

21-

.in'DG-1061 where you have risk information and are using.it 12 2 - to make changes to the current licensing basis -- how do you l23 .use performance monitoring to supplement and complement that 12 4 - decision. 25 ICHAIRMAN APOSTOLAKIS: Well, but.I think 1061 ANN RILEY & ASSOCIATES, LTD. Court Reporters. _ 1250'I--Street, N.W., Suite 300 Washington,.D.C. 20005 (202)-842-0034 l

~ ^ 317 c1' 1really says you should have some monitoring strategies but G- -_Q 2. it_doesn't' really get_into-how one defines reasonable-13 criteria, does-it? 4: MR. KING: It leaves it up to the licensee-to i 5 propose what is a reasonable monitoring strategy, what' 6 should be monitored, how frequent, what do you do with the 7 information. 8 MR. MURPHY: So I think the general prir.ciples 9 that you will hear for performance-based will apply both 10-ways, but they.will have to be tailored for the specific 11 application. 12 CHAIRMAN-APOSTOLAKIS: So that is my point. Who 1.' ~is developing those principles? I mean -- 14 MR. MURPHY: Well, you will hear some of them from s/ 15 me. 16 CHAIRMAN APOSTOLAKIS: From you. 17 MR. MURPHY: But it is broader than just what I 18 say, because they have to be tailored for a specific 19 application. 20. DR. SEALE: Joe,_I want to give you an opportunity 21 to disillusion me. You say you are concerned with L22 performance measures which do not follow directly from -23 risk-informed insights. 24 MR MURPHY: Yes. 25 DR. SEALE:- But that doesn't mean -- or does that-() ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

318 l' mean that you are talking about things which because of /~(N) 2 other considerations which give you a qualitative feel for 3 risk levels of concern, that you have identified those as 4 being important and so you are talking about the performance 5 measures for that sort of thing, or are you talking about 6 things that_you'd just like to have performance mera0res on 7 even though their risk status is perhaps uncertain? 8 MR. MURPHY: I think the answer is yes. 9 DR. SEALE: To which one? 10 MR. MURPHY: Both. 11 Let me try to explain. There is a reason this 12 paper is as late as it is in that there's been a lot of 13 philosophical discussions on something that really is fairly 14 simple in concept. 15 I want to start off with a basic premise which I 16 would ask you to note, that our regulations are to a very 17 large extent performance-based today. 18 If you pick up Part 50, Part 50 itself is mostly 19 performance-based. If you pick up Appendix A to Part 50, 20 the general design criteria, probably three-quarters of them 21 are performance-based. 22 As you get further on into the appendices-to Part 23 50, you pick up more prescription, but it is amazing when 24 you sit back and look at it with an unjaundiced eye how much 25 of our regulations are really performance based. [ ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

~. -- -. -. - - _. - - - ~ 319 1 Now I brought a couple other viewgraphs with me L2.

tha't-I.wasn't going to show you, but I am going toiput-them-

-3

on because they are either goingito-cloud this issue more or 54

- they are-going-to help, and-I am_not sure which. The reason 51 I wasn't going-to show them=is because I was afraid of-the-6: first ---VIN diagrams. 7 Basically I have three_ schools of thought. '8 One is that we have a-performance-based space in. 9 .Inside it there is an area that can by risk-informed, and 10 .there's two other spaces -- if.I can get the paper apart. 111 One says performance basis is a subset of risk 12 informed, and there is an argument in favor of this that-13' says at the, time the regulations were written, back in the 14 '60s, we thought'they were-risk informed. Every action we l-(O s/ 15 have taken has been based on no undue risk to the health and -16 safety of the public, and to that extent all our regulations 17 at the' time they were issued were risk informed. We just-18 may not have had a very good idea what the risk was before i 19 we started doing-risk analysis, which was when most of the 20 regulations were-written. 21' My own view of-the life and the way this paper:is, -22 that we have a risk-informed space and we have a. -23: performance-based space, and they happen to intersects-and- '24-so in here I have risk-informed, performance-based '25 regulation. ANN-RILEY & ASSOCIATES, LTD. Court Reporters. 1250 I Street, N.W.,-Suite 300 i Washington, D.C. 20005 l (202) 842-0034 o l?

320 5 1 Now depending-on how-you use that1first () - 2 definition, I would say.wa probably want to be in the-3

position that we evolve to this state, but right now'I think 4 1

-we can have -- and then;there-is a space out here.that is 5 -deterministic. 6' CHAIRMAN APOSTOLAKIS: What is performance?- ' */ MR. MURPHY: What? 8 CHAIRMAN APOSTOLAKIS: What is performance? What 9 is the definition of performance? 10 MR. MURPHY: We'll get to that. 11 The static electricity is good today. The slides 12 won't come apart. 13 Well, what we're trying to do in this paper is-14-first consider those performance-based approaches that do () 15 not explicitly reference criteria from PRA and then plan how 16 they may be phased into the regulatory structure and then 17 solicit industry input. That's the overall goal of the 18 presentation and the overall goal of this paper. 19 The paper, by the way, does not yet have office 20 concurrence. It's just been circulated yesterday. We'd be 21-glad to give the Committee a copy, but right now we, don't 22 have a consensus, and I will get you one as soon as I can, 23 which I hope will be by the end of this-week. 24 Now what's our approach? The approach is first to 25 specify a safety 1 objective -- I'm going to stand up so I can I ' '(d ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

T Jf _ 323 .i 'll see_my own viewgraphs -- and what actions we'll take if the g 2? .objectivefis notimet. i 3 Let_ me see:if - - = in my: briefcase 1I have ' a-hard a' f4'

copylof-these things._ I.can do that sitting-down.

S -Not.* once we. set /this basic safety objective, the j -6 -licensee; determines-how-the objective,will_be met. In- '7 setting the objective we require that there be some margin. 8 In other words, you would not set an objective _that took-you 9 right up to what you believe was the safety _ limit. 10 CHAIRMAN APOSTOLAKIS: Can you give us_an example 11 or two of safety objectives? 1 12-MR. MURPHY: The low-power shutdown, the 13-spent-fuel-pool-may be a good example, although I know the 11 4 rule is having its own share of troubles. The objective may 15 be I don't want the spent-fuel pool to boil. Or the 1 16 objective might be I don't want the level in-the spent fuel 17 to_ lower to the point where I have a shine dose. equal to i 18 some radiation dose, c -19 CHAIRMAN APOSTOLAKIS : And these are areas where a 12 0 you cannot do.a.PRA or -- 21 MR. MURPHY: In some_ places.they are, and in some - 22 places'they're not. 23-CHAIRMAN APOSTOLAKIS: -You.cannot do them for:what 24 reason?- 25 MR MURPHY: Inadequacies in the PRA method 61ogy, t ANN RILEY & ASSOCIATES, LTD. Court Reporters

1250 I Street, N.W., Suite 300 Washington, D'.C.,20005 (202) 842-0034

(- cy e y rW g-y vv .i-. 1%eum

322 -li

I canicome up'with performance-based-rules'for_ security,_ for

() 2 instance. I don't-really; handle sabotage-or-insider _ threat-3- .well in a-PRA. Fitness-for-duty' rules don't really fit-4 'we'll intota-PRA, but I certainly can_have a- -5 performance-based objective associated with them. 6. You know, the basic objective iflI can_put it 7 . bluntly is that I would prefer that, and our safety. 8 objective' is that people who are spaced out on drugs or 9 drunk on alcohol should not be oper?. ting nuclear -10 -powerplants. Now how do you come up with a -11 performance-based objective to show that you meet that -- a 12-performance-based program to show that you meet that -13 objective? In some cases if you did enough PRA you could ?A get there, but you may not have done it yet, and you still 15 _ may want -to go perfor. nance-based. 16 DR. SEALE: I guess external events are another 17 general category where you can have that problem. 18 MR. MURPHY: Yes. 19 CHAIRMAN APOSTOLAKIS: No, but there you do have 20: -PRAs. -21 DR. SEALE: Some do. 22 CHAIRMAN APOSTOLAKIS: Seismic fire. 23 DR. SEALE: Some do. 24 CHAIRMAN APOSTOLAKIS: We're talking about areas 25-where you-cannot do it, not that you haven't done it. ANN!RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

323 1 'MR.tMURPHY:.Well,oI think that I'm talking about. O ( j 2 .both. Onerisethat I haven't done it, let's put it that way.- 31

In - the L PRAs that'have.been~done there are1 sufficient 4

uncertainties or I have eliminated something by' assumption 5 Las,I did the PRA so that I can't use the PRA model today. 6 Now in-some of these I may not be able to.use it in the near 7 f u tu re. _. 'In others I.may be able to start a PRA today. But '8 I can't answer it yet. 9-CHAIRMAN APOSTOLAKIS: It seems to me that if you 10-actually can do a PRA, and you just haven't done it-for some 11 reason, then you should do the PRA. 12 MR. MURPHY: But you could go to performance-based 13 regulation while you're doing it. I mean, the PRA based on 14 past experience may take you two, three, four, five years, ) 15 -- dr.pending on how many you're doing, the level of depth of 16 ~the PRA. To get that kind of insight may take a rather 17 significant expenditure of resources. 18 CHAIRMAN APOSTOLAKIS: That's one side of the 19 -- coin. The other side is that if you do this, you may be 20 -derailing the risk-informed initiative, because people might 21 say well, gee, I'm doing this now, I'm fine,.why do I need a 22: PRA? And we don't-want to do that, do we? 23 'MR. MURPHY:- No, as I say, I think we ought to 24

evolve into where. performance-based is part of 25 risk-informed.

l. ANN'RILEY & ASSOCIATES, LTD. L'- Court Reporters L 1250 I Street,.N.W.,. Suite 300 Washington,.D.C. 20005 (202) 842-0034

.,. m 324 ' l DR.'KRESS:- There_are-likely places where you'll. () 2 never be able to treat well with the PRA and maybe ought'to 3 consider-performance-based--regulations as a-permanent-way to 4 deal with them. For. example,'if you ever really.got into ST organizational factors. 6 CHAIRMAN APOSTOLAKIS: Um-hum. 7-DR. KRESS: That's -- maybe that's never going to 8 be amenable to PRA, and '.t seems like a'likely choice. 9 CHAIRMAN APOSTOLAKIS: I disagree with that 10' DR. KRESS: Well, we'll-just -- . 11 CHAIRMAN APOSTOLAKIS: But there may be -- 12 DR. KRESS: There may be areas. 13 CHAIRMAN APOSTOLAKIS: There may be areas. - 14 MR. MURPHY: -Yes, I hate to say never on almost () 15 anything -- 16 DR. KRESS: Yes, I agree. 17 MR. MURPHY: But practically speaking, there's 18 areas that I just plain can't do very well or I can't do at 19 all yet. Other areas I can do but I know that they have 20" major deficiencies in the method, ao that when I calculate a 21 . PRA answer I have to do it with the full recognition that 22- - there are things that I can't handle with it. 23' _ A.QA is another example of.something that is very 24 difticult tolmodel in a PRA. I have'a concept in'my own 25 Emind of how you might be able to do something, but absent- ^' 'f' ANN RILEY &= ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 T 'W W-' Mr F, v-' t-7 7 e-

~. - 325 1-Ldata-to'proveLwhat's sitting around in the back of my_ head,- 2-that's_a very conjectural-k'ind of thing'. bj; 1 . 3 -DR. KRESS: -You'd have to have a lot of data-forL .4 any model1that relates-QA to actual, improvement in- ~ -5 viability;. 6' CHAIRMAN APOSTOLAKIS: -I think the most ~ 7-challenging problem _is whenEyof have some risk information 8 'but. itis-not-complete and_you do'want to'have performance 9 criteria that, you know, utilize-risk information as much'as .10 you can, but then you have to supplement those by' criteria that do not use risk information. And I think that's the 12 challenge. So you are really doing the extrcme case where 13 there is no risk-information at all -- for PRA, let's put it 14~ that way,_PRA. 15. DR. SEALE: There's no quantitative result. .16 CHAIRMAN APOSTOLAKIS: Quantitative result. 17 No, if you can do'the PRA, though, I don't know, 18 I'm not too sympathetic with that. I mean, I think we've 19 =gone out of our way to accommodate people who haven't done 20 this, haven't done that. Well, I don't know, it can take .21 another 30 years to finally say now you do it. I mean,.this 22 technology didn't start yesterday. 23 DR. KRESS: Well, I-must'say-I'm very sympathetic-24 to:that~ view, but I also believe there are going to be some ^25-things'we have to regulate-that just_are never gcing to be-

l ANN RILEY & ASSOCIATES,~:LTD.

Court _ Reporters 1250 I. Street, N.W., Suite 300 Washington, D.C.'20005 (202)'842-0034 2 + m

~.- 4 {' 1 "r -326-j (1

very well; treated by the_'PRA.,

_j [.2T ~ CHAIRMAN APOSTOLAKIS: If they cannot beLtreated { -: 3 iwell', Uunderstand:-that. 4c . DR. s itRESS : :Yes, and-tliat's-what:I_'m -- I'm'just 5 0"; going-to takeThis -- what he~says and_ apply it-to;that 6-

proportion of.the regulations.

But.:I agree with you, if you 7 can;do_-asPRA,-it juct_hasn't.been done, why -- ' 8 '- DR. FONTANA's. Theres; basically two different -- [ - you're: talking about really-two different things here, '10 because you can_ use a -PRA to detennine prescriptive 11- 'regulatienb if you wanted to. Performance-based doesn't 12 have to be linked with PRA logically. It just makes'a lot-13 of sense to do it that-way.- /N-- '14-CHAIRMAN!APOSTOLAKIS: No, I agree that it does j:V 15- -not have to.. I'm just trying to understhnd under what -16 conditions we develop what. So if.we cannot do it, yes, I 17 agree,-thea,- you know, that's something we have to deal '18 with. 19 .MR. MURPHY: Well,.I'm kind of jumping ahead to 20; .another' slide. Let me:put that.up and-then come back to the 21 one-that was;just_on. This'is what ILconsider the _:22 attributes.of performance-based. And the one is that'I' set 23-in'aLobjective criteria. Now I can set'that criteria using- '24 risk-insights,_=for-instance.

25-L DR' D KRESS : : -Youccan'.make-judgments, say thks is-

' ANN'RILEY &-ASSOCIATES,' LTD. C.urt. Reporters 11250 I Street, N.W., Suite-300

Washington, D.C.:20.005-

(202) 842-0034L

._45 Y e - 4 w~+,44- =,-w-, 4#., ..-,,w -,.---<c-.,--

327 1; likely more risky than_this one, even though I don't have 7 Q: q) _ =21 any numbers. 3 MR. MURPHY:- Yes, o4 -DR. KRESS - Or_PRA.

5 MR. : MURPHY :

Or I may not believe the numbers. I 6 -may:look-in and:I'll take something that may'be- .7

controversial-and I'll~say there are things missing-from 8'

fire-risk analysis' -So I don't really believe the-9 bottom-line number. But I learn a lot from looking_at an. '10 analysis. Doing the analysis is worthwhile, even'though I ,11-may not believe the bottom-line numbers.. It's the logical 12 pattern of doing it and the fact that the integrated look at 13_ the whole system from a fire standpoint gives me a lot of 14 information. It's qualitative information, but I've gained (O ' \\,. I 15 a lot from that kind of thing. So that kind of risk 16 insights'I can use to help set my objective criteria even 17' though I don't believe bottom-line numbers. 18 CHAIRMAN APOSTOLAKIS: But the point, Joe, is that 19 the bottom-line number will never help you set criteria, It's'always the insights that you just mentioned that will. L20-21 So whether you believe the number or not is actually 22 irrelevant. I mean, if I tell you yes, this is the result 23 and.it's a distribution log normal from 5 to 100, how does 24 -that' help?~ It doesn't help you at all, even if you believe 25 it. . k's ( J' -ANN RILEY &1 ASSOCIATES, LTD. f'- Court Reporters 1250 I Street, N.W., Suite 300 (Washington,-D.C. 20005 -(202) 842-0034

~ - ~ - l-i 328 1 DR. KRESS: 1-would-almost call that risk-informed ) ~ 2 if.-- 3 CHAIRMAN APOSTOLAKIS: .Yes. .I think what you.just 4' mentioned,_you know,_the_ insights, doing it, developing the - 5' scenarios and all that, that's'really what will help you do - 6 the -- develop the criteria. So the credibility of the-7 final number.really is an irrelevant'-- 8 MR. MURPHV: I agree with that, and that's why:I 9 said at the beginning that much of what I'm saying, even-10 though it's developed to the nonquantitative PRA kind of 111~ stuff coming in, is' applicable across the board, because you 11 2 can say this risk insights and as I'm using it I'm thinking 13 in terms of qualitative insights, but you get quantitative 14 ~ insights as well. 15 CHAIRMAN APOSTOLAKIS: I would be -- 2 16 MR. MURPHY: Depending on how much you believe the 17 analysis itself. 18 CHAIRMAN APOSTOLAKIS: The thing that bothers me 19 with the four items you have there is that again they can be 20 used by people and say, you know, there are four Z 21' possibilities. I pick.one. If you could prioritize them 12 2 ano show some preference and say looki I really would'like 23 to have the risk-insights. If I can't have those, maybe a a 24L _ hazard analysis would be the.next best thing. Then I-think 25 that would be a much more realistic way to approach the ..A l(j-ANN RILEY & ASSOCIATES, LTD. Court Reporters = 1250 I. Street, N.W., Suite 300-Washington, D'C. 20005 ^ (202) 842-0034 y8ei1 yc- -t 'T 4 ~'*w -+-W** w+% -FT vT ^ '

329 1. -problem._ 2 DR.-SEALE: This~ sounds like the PRA <3 -implementation or let's;say-1061s-integrated assessment or L 4 -- integrated decision making-process where you don't have a . quantitative PRA result as an inputito that integrated : 6 decision making-process. I mean, that's basically what 1 7 'you're talking about here, I guess. I mean, everything else 8 that's there -- 9 CRAIRMAN APOSTOLAKIS: The integrated decision 10 making. ! 11 .OR. SEALE: Yes. 12' CHAIRMAN APOSTOLAKIS: Yes. 13 DR. SEALE: Yes. It's just that process where 14-you've got to know value for the PRA input. 15 CHAIRMAN APOSTOLAKIS: But I still think some-16-cLatement to the effect that certain things are preferred 17 over others would go a long way. We had the question 18 yesterday from Commissioner Diaz, how do you make people who f 19-are skeptical, you know, realite that they have to use this, 20 DR. SEALE: Sure. E .21 CHAIRMAN APOSTOLAKIS: And=I repeat, this is-22 22; years after the reactor safety study was published in final 23 -form. So I don't think we can say, you-know, they need more 24 time-to understand-it, because.I don't think it's so 25 profound. l.; ANN RILEY & ASSOCIATES,-LTD. l Court Reporters

1250 I Street, N.W.,

Suite 300 u Washington, D.C. 20005 (202) 842-0034 = .a. ~.

. ~ ___ l 1 ] 330' 1= 'MR. MURPHY: Well, you:know,-I agree with you,~as, -) '2 you know,-100 percent. I was drafted to' work for four T:f ,' months;on risk analysis in 1972,.and I've been dabbling in 4 it.ever since. The -- . DR. SEALE:- _That's-because you were so good _at it,. 6: IJoe. 7 MR. MURPHY: Well,.either that or it took a long-8' time,_ you-know. 9 CHAIRMAN APOSTOLAKIS: In-1972, you know, the 10 state of the art was not -- l 11 MR. MURPHY: It was kind of nil. 12 CHAIRMAN APOSTOLAKIS: See, that's why -- 13 MR. MURPHY: But, you know, recognize that.when I 14 go' forward with this, I'm starting with a-basic premise, ,- Q(s,/ 15 I'm answering a Commission question.that says what do you do 16' when you're not amenable to PRA? And I'm taking that to 17-mean quantitative PRA, but the -- even, as you know, if I do 18 a PRA and I don't even quantify it, I just get cut sets. 19 I've got a lot of valuable information. 20 CHAIRMAN APOSTOLAKIS: _Unless you are overwhelmed. 21 If I give_you a thousand cut sets, I don't know what you can 22 do with them if I don't prioritize them.using probabilities,

23-right?

If I give you five -- 24-MR.-MURPHY: I have to use some judgment ns how I H25 - look'at them. You_know, I -- in general I can say singles 7 '( ANN-RILEY & ASSOCIATES,'LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 l

I 331 1L 'areiworse than doubles are worse than triples,. And then-I 1 4 2 look at/it in terms of.conmon-cause factors,_cheithings that

3

may/influenceLit.

All:this without ever quantifying <them? j -4; tall. 5. CHAIRMAN APOSTOLAKIS: So you're quantifying in j

6:

J your mind, in casence.- 7-MR. MURPHY: -In-a very-rough way, yes. 8-- CHAIRMAN APOSTOLAKIS: In a rough way 19 MR.- MURPHY: Of course.- 10' CHAIRMAN APLSTOLAKIS: My skepticism has~to do "11 with the fact that I don't -- I. suspect there is nobody who 12 - is thinking about.this thing at a higher level, the highest I -13 possible level, if you were given a task and obviously you 14 have to-respond to that. But'this is not the highest. level. 15 DR. SEALE: Well, let me ask you this, though, or 16 perhaps we could ask the Staff. Are you thinking in terms _17 -of a-hierarchy of inputs in the integrated decision making -18 process? 19 ' CHAIRMAN APOSTOLAKIS: I am thinking.of 20 performance-based regulation. Just as we had Mr. Holahan 21. _and Mr. King-think about risk-informed regulation, I would 22T like somebody to be responsible for that. g 23 Now, Joe, may very well be that man,.but he-was 24 asked to do something very specific when PRA is not s25l available,.and that bothers me, because I would like him to m - [( ANN-RILEY-& ASSOCIATES, LTD.

Court Reporters.

1250 I Street, N.W., Suite 300 -Washington, D.C. 20005 '(202) 842-0034 es ~w v.y- + -.p m. =c-- e v

~_ 4. 332 think about the overall problem and then tell me, naybe in a n()_ 2 hierarchical fashion, for some problems, when you do have 3 this information, this is what you do. For other problems, 4' when you don't have tnything like that, this is what you do. 5 And in between, there is another spectrum of things you can .6 do. 7 And I am not sure that there is somebody senicc 8 enough now at the agency who is doing this kind of thinking. 9 If there is, then my problem goes away. 10 MR. MURPHY: I think the overall development that 11 ne went into in this, which is pretty simple, as I said, it 12

overs both the risk-informed -- you make -- I guess the 13 easiest thing to say-is you make a decision on the basis of 14 all the information that is-available to you.

If you have ( (\\_/ 15 risk information available, cartainly you use it. 16 If you have risk information available and you 17 believe that there are some portions of it that are faulty, 18 but there is still good integrated information there, you 19 use it. l 20 You make a judgement on the validity of every 21-piece of information before you. It may be risk insights, 22 it may come from hazards, or just an-analysis of the 23 hazards. It may come from your performance monitoring. 24 Historically, you'look at the tracking of a system 25 or a train with time and you see it degrading, and that .. p 5-i ANN RILEY & ASSOCIATES, LTD. '~ Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 e.- ,,r--,n

-~. _ _.. _ _ '333 ? 1-tellstyod'someLuseful1information. If:maybe you-have done ~ 'J ( jAT 2' deterministic analysis. n 3L lit ~can be at a-very general level. Remember back 4 in 1976,Eit-is, after-we published WASH-1400, we came to:the 5 conclusion that, gee, we were making a mistake, auxiliary 6 -feedwater systemstreally should be safety graded. They 7 ?weren't up'until then. 8 And so there was a crash effort that Mr. 9 Cunningham in the back of the room,'and~a couple of others 10 .did, to try to quantify the reliability of aux. feedwater L11 systems for the -- every PWR in the country. 12 And that led to new requirements on the plants. 13 Now, they weren't based on a PRA, per se. They were based 14 'on analyzing _one system based on' insights that were given [15 from one PRA. But they_were important insightc and we.did 16 the-right thing. 17 So that you take -- you know, I don't see us 18 discarding any information, but we use what we have. Now, I 19 don't going forward in this manner in any way suggests that 20 we are diminishing our push towards risk-informed, 21 performance-based regulation. 22 -As I_said, you-know, we want to go to the 23 risk-informed. -But right now, in.some areas, we are not 24 there_yet. We'either haven't done the studies, even though

25 theyLcan be done, or we can't -- just plain can't do them

.-b ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250-I Street,- N.W., Suite 300 -Washington, D.C.-20005 (202) 842-0034

334 1-1 lyet. () 2 Anyhow, the attributes that I see for. 3 1 performance-based regulation are that we_have a measurable

4 parameter.

Now, that is either directly measurable or 5 indirectly. And I guess the easiest way to say what. .6 indirectly means is that a relatively simple calculation can

7

_give you, can be used to generate something. 8~ You have objective criteria to assess the 9'

performancs.

You are-measuring something and you have an >10 objective criteria to judge against it. That criteria is 11-chosen'from the information you have available to you. 12 You give the licensee flexibility. Once you have 13 stated the onjectives, the licensee has the flexibility to 14-tell you how he is going to meet that objective. 15. CHAIRMAN APOSTOIJLKIS: So who states the 16 objective, the licensee or you? 17 MR. MUPPHY; No, we state the objective. They 18 state how they are going to meet the objective. 19 CHAIRMAN APOSTOLAKIS: And the objective has to be 20- -fairly high level,.I suppose. 21-MR. MURPHY: The objective has to be at a fairly 22: high-level, but you have to consider a lot of things as to 23 how-you set it. 124 In_other words, one of the considerations, for 25 instance, is defense in-depth. It is very easy to setLa ANN RILEY a ASSOCIATES, LTD. \\ J Court Reporters-1250 I_ Street, N.W., Scite 300 -Washington, D.C. 20005 -(202) 842-0034

335 1 criteria for a system. But even though you haven't done a 2 quantitative analysis of it, you don't want to give a big 3 chunk of probability space away. So you may have to set ~ 4 your objective at a train performance rather than a system l 5 performance in some cases, for instance. Because you are i 6 trying to preserve this cor. cept of defense in-depth. 7 You know, in an auxiliary feedwater system, it is i 8 important that you have both electrical and turbine-driven 9 pumps. You may not want to give that away and set a l 10 performance standard for the aux, feedwater system. You may 11 want to set a performance standard for the electric-driven 12 portion and for the performance,-and another one for the 13 turbine-driven portion. 14 Once the lice.asee chooses, however, at that point, 15 thac becomes fixed. And what I am suggesting here is that 16 be fixed in a licensee -- in a control document. Depending 17 on th9 importance, that could mean something like the FSAR. 18 It could mean a license condition, such that, once chosen, 1 19 it can on./ be changed with care. Perhaps that has to be 20 something like a 50.59 process. Perhaps if we really 21 consider-it.important, it has to meet something like a 22 license amendment. 23 So the concept is basically the NRC sets a safety 24. objective. A reasonable'high level, but recognicing the 25-importance of~ defense in-depth and multiple trains, single g . ANN RILEY & ASSOCIAT3S, LTD. { Court Reporters 1250.I Street,- N.W.,--Suite 300 Washington,-D.C. 20005 (202) 842-0034

i 336 1 failure criterion and that sort of thing. f () 2 Once you set-that general objectiva, the' licensee 3 picks how they are going to meet it. They choose the j 4 parameter they want to meet. Or we may choose it either, if 5 we feel-strong enough. We could choose.the parameter, or we 6 could leave that open so that they could choose it. i 7 Once that parameter is set, however, and they tell 8 you how they are going to' meet it, then, at that point, that 9 becomes a fixed parameter from a licensing standpoint. f 10 CHAIRMAN APOSTOLAKIS: Well, it seems to me that 11 it would be nice to have a set of criteria or guidelines, or 12 principles that will give advice to the people who set these 13 criteria. Because the way you described it, now it is t 14 pretty open-ended. ( 15 MR. MURPHY:

Well, 16 CHAIRMAN APOSTOLAKIS:

I mean I can choose to 17 apply defense in-depth and go to a very low level, or I can 18 choose to do something else. And that -- I mean in the 19 handslof an experienced, rational person, that's fine.

But, 20

- you know, it -- 21 MR. MURPHY: Well, I hope we are. No, what my 22 thought is, you see, as I say, I require a margin, and then 23 to explain. Yes, we are giving the licensee flexibility, - 24 hut we are-also preparing regulatory guides, standard review 25_ plans to identify what that means for a specific () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street,-N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 e ---sm - w n-e--m-- -y--,- w r w,ww -v , m,n w, r-x g-

m. p-qps

-a g-,,p-e.r~wrz ew- ,-~w-- -,----wrm--r--- r r-own,-: e-em p-

337 1 application. What kind of depth, level of depth do yo need-2- to go into? And then that is the purpose of the reg. 3 guides. Is to accomplish just what you said. 4 And then once the licensee determines how they are 5 going to meet it, that determination is approved by the 6 staff. Then it gets locked in to a control document. And 7 tne degree to which it is locked in depends on the document 8 -- on the importance of the issue and how you are going to 9_ grade it. 10 This could be fleshed out more, but I think it ~11 real]y needs to be fleshed out on a case by case basis 12 almost. 13 Now, the implementation of the process -- well, 14 let me talk about the implications side first. As I said, 15 provided we can develop objective criteria, based on any 16 analysis that we have done, which may be PRA-based, it may 17 not be, we may come up with a qualitative safety objective, 18 We also may feel something is strong enough that we will 19 pick a very fixed objective. In other words, the 20 temperature of the spent fuel pool shall not exceed 125 21 degrees fahrenheit. Ncw,_that's a safety objective. So is 22 a safety objective saying it won't boil. 23 We channeled the inspection process, and I think-24 this is. fairly.well addressed in a paper that was sent to 25 the Commission about a month ago by NRR on inspecting for ii ANN RILEY & ASSOCIATES, LTD. Court Reporters 1240 I-Street, N.W,, Suite 300 Washington, D.C. 20005 '(202) 842-0034

338 1 performance and performance-based inspections. I don't know r~s ( () 2 whether the. committee discussed that or not, but it is an 3 interesting paper. 1 4 The inspection focus should be-on the oversight of l t 5 the performance monitoring process and the effectiveness of f 6 the corrective actions that are taken if you start j 7 approaching or exceeding one of those-performance limits. 8 And, as I said, in some cases at least, my guess is that { 9 defense-in-depth considerations may lead us to setting 10 _ performance standards at the train level more than at the 11 system level. 12 Even though I don't have a quantitative analysis, 13 I can.look at, in a number of things, and say, you know, a 14 know a train is worth somewhere between 5 times 10 to the 15 minus and 10 to minus 1, and that element of probability 16 space, I don't want to give up. And so I may want to set my 17 standards at a different level. 18 CHAIRMAN APOSTOLAKIS: I am trying to understand 19 now how one can -- is facing, how can you face the problem 20 of setting performance criteria for trains without a PRA? I 21 mean, can you give me an example of that? 22 MR. MUR"HY: Well, you know, if you look at the 23 general design criteria, they were written in 1969 before '24 ~ anybody developed PRA, and three-quarters of them are . performance-based. () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,-Suite 300 Washington,'.D.C. 20005 (202) 842-0034

h 339 l -1 CHAIRMAN APOSTOLAKIS: Right. But you are doing l () 2 it-now. You are doing it now. So -- 3 MR. MURPHY: Where I have PRA information, I f 4 .certainly use it. Where I haven't done a PRA on 1 specific l 5 plant, or I find there is something significantly wrong with 6 the PRA on the specific plant, with the IPE in th6 specific l 7 plant, that I don't want to believe it, but I have this 8 whole bunch of insights that have come from the IPE program, 9 and from all the other PRA's that have been done. 10 Certainly, I use it. I don't discard information 11-in making safety decisions. I use the whole panoply of l 1 12 information available to me. l 13 CHAIRMAN APOSTOLAKIS: Yeah. But, I mean when I 14 asked you about examples, for examples, you told me fitness 15 for duty and exotic things like that, for which there is no 16 PRA. Now the discussion is on trains. It seems to me that 17 is: Level 1 PRA. That's different. 18 MR. MURPHY: Yeah. Yeah. I am a systems engineer i 19 and I get-down to the level and maybe I should stay higher. l t 20 But the fact is that -- all I am suggesting is, not that in 21 every case will I need to go the train level, but in some 22 cases I might. -23 If I get-to fire protection systems and some 24 things that I am not confident of the PRA, then I may be 25 using more qualitative insights. I () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 q _.. _. ~. - - _ _. - _.,. ,__.....___.._._..2

. _ _ _ _. _. _. _ _. ~ _ _ _. _ _ _ _ _ _ t I 340 i Fitness for duty. I have to start looking at, you i now, their performance data might draw what I eny. I have a 3 sken so many eamples ovJr the years in the nuclear { i ' 'i industry, and I have an idea as to what'my success rate has 1 5 been. 6 Now, I can have some sort of trending information 7 that says I will judge the effectiveness of your fitness for 8 duty program depending on whether you increase detection 9 rate. Now, I have to worry about other things in that, too. 10 - You know, am I getting -- how good is my program? You know, 11 are things slipping through? 12 But I used the entire basis for the program. But + 13 instead of specifying exactly how I may do something, I ccn 14 set an overall target. 15 I'll leave the details of the program _to you, but l 16 once you pick those details, I will lock it into a licensee 17 controlled document so you can't change it without giving 18 serious thought to it. 19 Now that doesn't mean you can't change -- you, the 20 licensee, can't change it. You can change it under a 50.59 21 type process is what I would imagine, so the licensee still 22 has the flexibility. The licensee still can make the 23 changes but they are reasoned changes. 24 Finally, the question that the Commission asked us 25 is-how do we implement it?' [ ANN RILEY & ASSOCIATES, LTD. Court Reporters s 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202)-842-0034 ..-.___;__m..

341 1 Well, what we are trying-to do to the extent () 2 possible is to piogyback onto DSI-13. I believe the 3 committee was briefed on DSI-13 at the last meeting by John 4 Craig. ] 5 This is the direction-setting issue on-6 interactions with industry. As part of that process, as i 7 part of the public law just passed a couple years ago, we 8 are required to raise questions with our.stakeholders if our 9 regulations can be better represented by consensus standards 10 that have been developed by presumably one of the 11 profecsional societies. 12 This will require that we will have frequent 13 interactions with stakeholders to solicit from them 14 questions of what issues are amenable to the use of 15 consensus standards. 16 What I had hoped to do with this is to minimize 17 resource commitments. It would be to use those kinds of 18 neetings that will be going on under the DSI-13 rubric to 19 ask the additional question as to are there any regulations 20 out there that should be made performance-based and solicit 21 the input from the industry, from the public in one case. 22 Now if for some reason the plans for DSI-13 don't 23 fit the kind of schedule we want to make'here, obviously we 24 -can separate those two but the real thing is to solicit 25 ' industry suggestions-for candidate regulations that might be ANN RILEY & ASSOCIATES, LTD. Court: Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034' ~ . ~

342 1 . converted to a more performance-based scheme. j l 2' We would encourage petitions. We published a Reg f 3 LGuide_-- I guess it is about a year and a half ago now -- j 4 our Reg Guide 10.12 that provided information as to what 5 information should be submitted with a petition under 10 CFR 6 2.802 to get us to change the regulations and identify the 7 kind of information that would help speed the process t 8 through. 9 We would continue as we did for the last several i 20 years through our marginal to safety program to encourage 11 petitions on that type case, in that kind of a situation. 12 We'd evaluate the need for pilot studies. Perhaps 13 they are needed, perhaps they aren't -- perhaps the pilots 14 that are already going on in many different areas would be 15 helpful, some PRA-based, some weakly PRA-based. You know, 16 do you need something or can you gain? 17 18 As you say, when you are looking for 4 19 performance-based standards, a performance-based standard 20 that11s based on risk information is better than one that [ 21 isn't, in my view, but the performance-based aspect of it 22 you may be able to gain insights from looking at the other. 23 Finally, we are committing to report to the 24 Commission on what we are doing by the end of fiscal '98. 25 CHAIRMAN APOSTOLAKIS: Is the ACRS getting () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 b ,,- + ~ .-wr = ,9 -m%., y v.,,w 3-r,- ,.---,,+_-,p ,w,, --r--

343 1 involved at some point in this? Are we writing a letter? 2 MR. MURPHY: I don't know. 3 CHAIRMAN APOSTOLAKIS: Do you have any plans to l 4 solicit'a letter or -- 5 MR. MURPHY: I think it probably would be wise for 6-us to request your review. 7 CHAIRMAN APOSTOLAKIS: When? 8 MR. MURPHY: I would hope to be able to have the 9 letter to you -- 10 CHAIRMAN APOSTOLAKIS: The letter? -11 MR. MURPHY: What? I 12 CHAIRMAN APOSTOLAKIS: The letter -- are you i 13 writing a letter as well -- or the report? 14 MR. MURPHY: No, no, we are writing a report to 15 the Commission in response to the SRM. 16 I hope we have that Commission paper available -- 17 I think we can make it available to the committee if things -18 go well in the internal review process early next week. 19 Now at that point I guess that would mean ---I 20 don't know when your next meeting is after December but -- 21 CHAIRMAN APOSTOLAKIS: So we will have a chance to 22 comment on what you are doing well before you finali::e 23' anything? 24 MR. MURPHY: Well, I hope to send the letter 25 forward soon. It is due to the EDO within a week. ANN RILEY'& ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,-. Suite 300 Washington, D.C. 20005 (202) 842-0034 -v--imrv,,,es, -.w-, ,.--we w g- ~, - - ---,r-- e, y -,-r- -r., -,. -,,n,- ,r---r

_ - _. =. l 344 1 CRAIRMAN APOSTOLAKIS: How about this report to ) () 2 Commission by end of 1998? That's different. 3 MR. MURPHY Well, certainly you would have full 4 opportunity to comment on that. 5 DR.:SEALE: Joe, have you -- r 6 MR. MURPHY: But it is something that says 7 basically what I said to you today was what our current 8 Commission paper says. 9 I would plan to send that forward to the EDO -10 almost immediately and we could use the committee conments 11 to-influence how we develop the report that is due at the 12 end of '98. 13 CHAIRMAN APOSTOLAKIS: Well, could we put it on 14 the agenda for Decenter? That is two weeks. 15 DR. SEALE: No. The agenda for December looks 16 like -- it's full. 17 MR. MARKLEY: February would be the soonest next 18 date. 19 CHAIRMAN APOSTOLAKIS: That doesn't help him very 20 much. 21 DR. SEALE: Joe, could I ask ytu a question? ~22 MR MURPHY: Sure. 23 DR. SEALE: You have here I'll say a sketch of a 24 plan to use'-- a technique that can be used to identify I 25: guess performance-based regulation candidates in absent a

( 3

^f ) ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 x--.

i 345 1 PRA. i 2 It is a sketch of-the use of something you could 3 use to do that, f 4 What do you'put in here t'o protect-against abuse?- l 5 This is almost kind of a hunting license-for things that -- 6 you know, I could see where if a person has a hobby-horse in 7 the current regulatory process.and he wants to legitimatize 8-that interest,.and there is no support for that necessarily [ 9 in a present PRA, I-could see you trying to go through this

10 process'in order to cover that particular item with a 11 performance-based approach where there may not be the 12 full -- a real legitimate reason to include it in that part 13 of your VIN diagram.

j 14 MR. MURPHY: Well, I have the same concern. 15 DR. SEALE: Have you thought about how you would 16 go about it? 17 MR. MURPHY: Yes. I think in the Reg Guide that I '18 mentioned it asked for a complete technical analysis of the 19 -issue. Clearly this would have to be reviewed by the Staff 20 in some depth. 21 -It would have to have a_ safety evaluation to '22 approve it. 23 The process for changing a regulation is an-24: involved process. You know, it involves seeking public 25 comments and putting things out for comment, considerable ANN RILEY &. ASSOCIATES, LTD. Court Reporters -1250 I Street, N.W., Suite:300 Washington, D.C. 20005 .(202) 842-0034 .c u

346 .1 deliberation with you guys, and to be honest with you, I () 2 guess the answer to your question is that that is what the 3 NRC Staff is for, is to keep something like that from coming I f 4 through the system -- to look for those-that make sense, to 5 try to implement them, and.to stop those that don't. 6 DR. SEALE: Well, I think a few words that 7 specifically address that might be -- l l l 8 MR. MURPHY: That is a good suggestion -- 9 DR. SEALE: But I would recommend again that you 10 look at this integrated decision-making process and ask 11 yourself how close is that to what you are talking about t 12 absent a PRA. 13. CHAIRMAN APOSTOLAKIS: Well, I would like to have 14 an opportunity to comment in writing as soon as I can 15 because frankly I am cool to the whole project. There is a 16 good chance-it will derail the risk-informed initiatives so 17 I don't like that. 18 I may be wrong but -- so I don't know when we can .19 have an opportunity to write something to somebody. 20 MR. MARKLEY: We can't schedule any briefings for 21 the December meeting. However, if we have the draft paper, 22 you could do a review and possibly recommend a future 23-meeting at the December meeting, something like that. 24 CHAIRMAN APOSTOLAKIS: Recommend to the full-25 committee? () ANN RILEY & ASSOCIATES, LTD. Court Reporters L 1250'I Street, N.W., Suite 300 Washington,-D.C. 20005 (202) 842-0034 . -. =...

f 347 1 MR MARKLEY: Yes. 2 CHAIRMAN APOSTOLAKIS: W.11, I guess the earliest 3 is Februury. 4 MR. MARKLEY: Yes. 5 DR. SEALE: Yes. 6 CHAIRMAN'APOSTOLAKIS' Now you say there is a good 7 chance your letter will go up in a couple of weeks? Is 8 that -- 9 MR. MURPHY: I hope it's a lot faster than that l 10 but it depends on -- it is_very basic principles but they 11 are very' philosophical in nature, which means a lot of '12 people want to comment en it, so it's taken a long time to 13 write it and even though it is only a few pages long, but my 14 goal is -- I'll put it this way. The due date to the EDO is 15 next Monday. Whether I am going to make that o not, I i 16 don't know. 17 DR. FONTANA: But then you have got a whole year 18 practically after that. That is mostly taken up with 19 reviews with industry and all that kind of stuff. 20 MR. MURPHY: Yes. It is sitting down with i 21 industry, soliciting their-ideas, encouraging them that if 22 they want to do something -- I still think they are the ones 23 that know where this may do the most benefit for them. 24-I would encourage them to submit a petition -- and 25- 'we have tried to make that as crystal clear as to how to do ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I: Street, N.W.,-Suite 300' Washington, D.C.-20005 -(202) 842-0034

..._..__.-.___.__.._.____e - 348 l i i that as we can, and then we would review the petition as we j 2 do any rulemaking petition and try to apply these general () 3 principles to it. 4 DR. FONTANA: But that occurs after the report j t 5 goes to Commission at the end of fiscal '98. [ 3 6 MR. MURPHY: No, it could happen parallel to that. [ 7 DR. FONTANA: Okay. 8 MR. MURPHY: In other words, what the report at [ t 9 the end of fiscal '98 is to say we met with the industry, [ i 10 this is what they suggested, this is what they have done or j 11-this is what they haven't dene. 12-, If the answer is they have shown no interest [ 13 whatsoever in this, then it may die on the vine if the 14 answer is that they=have proposed a whole bunch of things l ( 15 and we are now characterizing them and prioritizing them and 16 basically it is a status report of where we are at the end 17 of the fiscal year. 18 This gives us nine months to gather information to 19 try to put it together to flesh this out more. 20 DR. FONTANA: Well, I respectfully don't agree 21 with George on this. I think there's a lot of benefits to 22. this that shouldn't be held hostage to requiring everybody - 23 to do a PRA. Of course long-term is that.it. should be 24 -risk-informed as quantitatively as possible. Of course, 25 there's'always something you.can't do.. () ANN RILEY & ASSOCIATES,.LTD. Court Reporters 1250 I Street,_N.W., Suite 300 Washington, D.C. 20005-(202) 842-0034 -.,m.,. y y,-- wv.-

-,--,y-.

y,-- ,p- --. w... m ww.~c..,,--, --v,-w,,-- yy,,e .,- r ,+,m, .y ,.e= w r- -e*- w e v,* w .r---,n--

. -.. - -.=-. i f 349 I 1-CHAIRMAN APOSTOLAKIS: I didn't say that we have j () 2' .to have a PRA. I would like the approach to be different. 3 That's all I'm saying. But maybe the emphasis, f i 4 DR. FONTANA: Yes.. 5 DR. SEALE: There are a lot of things like the 6-fitness-for-duty thing and those kinds of things where there i 7 .has been criticism that there's not anything being done in 8 those areas, and this is one way to begin to address some of 9 those -- I shouldn't call them' side issues, but things that 10 are not pretty,high on the screen right now, j 11 CHAIRMAN APOSTOLAKIS: Would the ccmmittee know l 12 enough about what you are doing if we had say a presentation, an hour an a half from you and your people in 14 February, or do we need a subcommittee meeting? Because 15 this is an extremely important subject, and I really want to i 16 understand =where the Agency's going with that. What do you 17 think?- 18 MR. MURPHY: I suspect in its conceptual stages 19 I'm not going to be able to give you-much more in a month 20 than I could give you -- than I gave you today, 21 CHAIRMAN APOSTOLAKIS: Okay. So maybe an hour, 22 .and hour-and-a-half with the full committee. 23 MR. MURPHY:. Yes. 24' CHAIRMAN APOSTOLAKIS: In February, 25_ DR..SEALE: Yes, that sounds reasonable. i ~ ()'- ANN RILEY & ASSOCIATES, LTD. Court Reportern 1250 I Street, N.W., Suite 300. .Washingtonc D.C. 20005 7 (202) 842-0034 d'

350 1 CHAIRMAN APOSTOLAKIS: Would be sufficient. [ 2; DR. SEALE: Yes. 3' MR, MURPHY: We'll be a little further along in 4 our thought. 5 CHAIRMAN APOSTOLAKIS: Yes. 6 -- MR. MURPHY: I value your input, and I certainly 7 don't want to do anything to derail the risk-in' work ~ 8 that's going on, and if -- 9 CRAIRMAN APOSTOLAKIS: I think, you know -- 10 MR. MURPHY: Your suggestions are very valuable. 11 CHAIRMAN APOSTOLAKIS: You know, I have 12 reservations. I think it's best. to air them earlier rather 13 than later, so you have that input. 14 MR. MURPHY: Yes. Mk 15 CHAIRMAN APOSTOLAKIS: Okay. Well, any other 16 questions? 17 Thank you, Joe. We'll take a break now. We'll 18 come back at 9:45. 19 [ Recess.) 20 CHAIRMAN APOSTOLAKIS: Now we're going back to 21 1061, right? And the policy issues. And we're back to Mr. 22 King, Cunningham, and Parry. Where is Holahan? He is not 23 coming? 24: MR. KINO: He's probably still getting a cup of 25 -coffee..LHe'll be here. i . y ~~ ANN RILEY & ASSOCIATES,'LTD, '( Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005' j L(202)'842-0034: 4 ~.. ~

i [ 351 l CHAIRMAN APOSTOLAY,IS: Oh, okay. f 2. MR. KING: We're going to pick up where we-left 3 off yesterday in going through the changes to 1061, and we 4 had gotten up to discussing PRA quality and-scope, and 5 that's where we'll pick it up. F ~ 6 MR. CUNNINGHAM: As we were discussing yesterday, 4 7 we've added.a section into the document that talks about the 8 quality and scope of the PRA that would be needed. Before- [ 9 we had more of a kind of a vague reference to NUREG 1602 and 10 - to talk about this, and-given-that we've kind of removed I 11 1602 from the process right now, we wanted to go back and 12 put something in. 13 I think that there are a couple of key points in 14 terms of the scope and the quality of the PRA, and our first 15 two are on these bullets on the slide, the first two bullets 16 on this slide. One is that we want the plant -- the PRA to 17 realistically reflect the as-built and as-operated practices 18 in the plant. So it gets at this issue of a need for a 19 living PRA that'we talked about a little bit yesterday. 20 The second key piece is that we try and reinforce. 21 the point that the scope and quality of the PRA required of 22 the PRA depends on.the application, and that there's not a - 23 single standard for the PRA=for all applications. There may 24 be kind of.a base standard above which -- minimum standard, - 25 ' if you will,-but there's no single standard for any ANN RILEY'& ASSOCIATES, LTD. Court Reporters - 1250 I Street,-N.W., Suite 300 z L Washington, D.C. 20005 (202) 842-0034 w y v -w y wr-e = t-e, -c,-m i, :. M _ m,v3-ye-,. -mw,r-- M.n - e-r s~ - y -r- . < ~ -,

l 352 I application. l () 2 Another point we made in this section is that the 3 -acceptance guidelines deal with all operating modes and all j 4 initiating events, but it's not necessary that we have a PRA 5 for all of these modes, again reinforcing the point that l 6 this can be done quantitatively and -- or qualitatively as J 7 well. 8 CHAIRMAN APOSTOLAKIS: Which section is this, 9 Mark,-in the actual report?- MR. CUNNINGHAM: It's section 2.4.2.1. 11 CHAIRMAN APOSTOLAKIS: 2.4.2.1. i 12 MR. CUNNINGHAM: Page 12. 13 CRAIRMAN APOSTOLAKIS: Yes. So again this is 14 wheta Dr. Powers may raise questions regarding the 15 credibility of the models and the availability of all the 16 information they might need. 17 MR. CUNNINGHAM: Yes. 18 CHAIRMAN APOSTOLAKIS: Do we have anything else 19 after that? It-says here the.t-all plant operating modes and 20 initiating events should be addressed. However, it is not 21 necessary to have a PRA that treats all of these modes. For 22 every application I guess is what you mean, but if you don't 23 need that. 24 MR. CUNNINGHAM: Yes,.that's right. - So one say to 25 deal-with-the concerns about-shutdown might be that proposed I ANN RILEY & ASSOCIATES, LTD. Court Reporters

1250 I Street, N.W.,

Suite 300 l Washington, D.C. -20005 -(202) 842-0034 [ r -7's-w pt 87 9'ygae r y -w -m g me t ? -ey t-p- -pt-'r. 7WW '1T1+r r M-W e 's**-Ww"'rT' w'we-:---rcr--" w '--V'*l'v-- Wv--

• w e'B"

'--h-

• Eru +

sw

353 1 CLB changes that deal with parts of the plant that are not ,(,j 2 related to shutdown operations in one way or another might 3 be a wiser course of action, if you will. 4 CHAIRMAN APOSTOLAKIS: You want to change that 5 word "can" to "may"? May be svtficient? What do you think? 6 It's the second sub-bullet, qualitative treatment of missing 7 modes and -- yes. 8 MR. CUNNINGHAM: I guess so; yes. 9 CHAIRMAN APOSTOLAKIS: It's okay? 10 MR. CUNNINGHAM: Okay. 11 CHAIRMAN APOSTOLAKIS: It's also in the text, page 12 13. 13 MR. CUNNINGHAM: Okay. 14 CHAIRMAN APOSTOLAKIS: There must be, though, I 15 don't know if it's a good idea, but the message should be 16 clear that if one chooses to give you qualitative arguments 17 he will not have as easy a time as, you know, trying to 18 quantify things. We don't want people to start waving their 19 arms and say, gee, you say here I can do qua41tative, so I 20 will do qualitative here. I don't think that this is so, 21 and I think this is so. Somehow we have to discourage that. 22 Is it going to be a practical matter when they're going to 23 have a hell of a time getting anything out of you if thef 24 try to do that? 25 MR. CUNNINGHAM: I'm not sure I want to discourage r ( ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

354 1 them. l () 2 CRAIRMAN APOSTOLAKIS: If it can be done f 3-quantitatively, yes, I think we should discourage them. 4 MR. CUNNINGHAM: But remember our objective is not l 5 moro quantitative analysis. Our objective is good safety 6 ' decisions. j 7 CHAIRMAN APOSTOLAKIS: Yes. 8. MR. CUNNINGHAM: And if you can make a good safety decision with qualitative information, and that's more 9 1 10 officient, why require more? ~ 11 r CHAIRMAN APOSTOLAKIS: That's true. That's true. - 12 If you can do that. 13 MR. CUNNINGHAM: So I would think that perhaps you 14 ought to apply the same standard, not encouraging more or () 15 less quantitative analysis but the demonstration that 16 whate er analysis, quantitative or qualitative, is i 17 appropriate to the decision that's being made it seems to me 18 is independent of how the analysis was done. 19 CHA7RMAN APOSTOLAKIS: I think, yes, this w!.ll 20 avolve from actual practice. That's okay. 21 MR. CUNNINGHAM: The last podnt that we have in 2" this section is the issue of peer review and certification 23 -processes and things Jike that. Again we'rc encouraging the 24 use of' peer revieu of the PRA to help give the staff more 20-Econfidence of the quality of it, and that certification ( ANN RILEY & ASSOCIATES, LTD. . Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C.' 20005 (202) 842-0034 t .--m. ,~n.,,. w, y .h[,, -,..,,.,..y,. _#,,,,,'.m... .,,m-7

355 1 programs and some cross-comparison studies could be a () 2 support to this overall review. It doesn't replace 3 necessarily a peer review or that type of thing, but it [ t 4 could be of banefit to acknowledge the types of efforts that-5 have been going on in the industry. [ 6 CRAIRMAN APOSTOLAKIS: Is t.nybody going to tell us l 7 at some point what the inadequacies of modeling are? I [ 8 mean, we have a list of those somewhere? 9 MR. CUNNINGHAM: I'm not sure I *tnderstand what 10 the inadequacies -- 11 1 CHAIRMAN APOSTOLAKIS: We have a number of people 12 now who are very experienced doing PRAs and arguing with 13 their peers and so on, and some of them of course are at i 14 national'laborntories who work for NUMARK. Others have been 15 in private industry. There is a whole body of knowledge 16 there regarding models, how good they are, what questions 17 frequently arise. 18 Do we have a report somewhere where the insights 19 these people have gained are there? For example, you know, 20 I ment,ioned once that we really don't have many model 21 uncertainties in Level 1 PRA and I got the answer no, that's 22 not-true, it's because we're not asking the right questions. 23 And there are a lot of questions.about success criteria, but 24-it seems that we just accept them. These kinds of insights, 25: you know, it would be nice to have a document, a NUREG or O' ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,, Suite-300 . ashington, D.C. 20005. W t (202) _842-0034 -a-,

__ -._ _= - -- t 356 1 1 something where something like that would be listed, in an l [) 2 appropriate way, of course, so it -- an evaluation of the v 3 methodology. 4 Because it seems to me that these standards and 5 these peer reviews,-they probably will have to address 6 minimum ^ requirements. I don't think we're ready to say yes, 7 this is what a PRA should look like, because then you are 8 inhibiting progress in some sense. But if you say yes, it's 9 unacceptable not to do common-cause failure analysis, much 10 to my surprise several IP's did crazy things with that, as 11 you guys know. 12 I mean, we choose to do it this way for these 13 components or will only do it for these components. What is 14 that? I mean, you have to do it. Now choosing the actual \\ms - 15 model may be a different story, but you have to do it. 16 Choosing not to do an uncertainty analysis, you know, why? 17 I mean -- so on. 18 So minimum requirements make sense to me, but 19 setting up standards is, I don't know, I'll have to 20 underrtand the subject more. But I think it would be useful i 21 to everyone, including the peer review panels, to have a 22 document from well-recognized experts where the limitations 23~ of models -- in fact there was an interesting paper 24 published a few years ago by three people from-PLG -- 25 MR. CUNNINGHAMi Yes. \\ ANN RILEY & ASSOCIATES, LTD. .^'# Court' Reporters 1250 I Street, N.W.,-Suite'300 , Washington, D.C. 20005 (202) 842-0034 ye yrwwa- -r,ry-p- -q 9y v-ve'9 4-g 'r"& 'ryr e-m- --~v*-- e'W+" 9 eew-

r F 357 1-CHAIm004 APOSTOLAKIS: And what was it, strengths ( 2 and limitations of PSA. [ 3 MR. CUNNINGHAM:

Yes, s

4 CHAIRMAN APOSTOLAKIS: - So, you know, maybe we need 5 an update about it, or maybe we need a broader group of l 6 people doing something like that. Because obviously they t 7 represented a certain point of view. 8 MR. CUNNINGHAM: Yes. 9 CHAIRMAN APOSTOLAKIS: You are familiar with it, i ~10 Gary? 11-MR. CUNNINGHAM: Yes. It's one of those things - 12 where the PLG people have touched on this. To some degree 13 it was touched on in the IP insights report. 14 CHAIRMAN APOSTOLAKIS: Um-hum, o k/ 15 MR. CUNNINGHAM: Where they talked about at least s 16 some of the issues. 17 CHAIRMAN APOSTOLAKISt-Yes. Yes. 18 MR. CUNNINGHAM: Some of the work that's being 19 done by CSNI touches on pieces of issues, if you will. 20 CHAIRMAN APOSTOLAKIS: Pieces of issues, yes. 21 MR. CUNNINGHAM: CSNI has a group now looking at 22 the issue of where are we in the state of the art i n fire 23 -PRA, for example, 24 CHAIRMAN APOSTOLAKIS: Yes. 25 -MR. CUNNINGHAM: But it -- I'll have to think 37 ^-)~ Court Reporters i ANN RILEY & ASSOCIATES, LTD. 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 .(202)-842-0034-7- -+-r + tr y .w h r w e.w, e-c r-.-,. -,-+ wow--, ,www v. -* I v a w

358 1 about something. () 2 CRAIRMAN APOSTOLAKIS: That would be a useful 3 thing-to have. We also have to think about the.best way of l 4 doing it. Maybe it's not worth having a NUREG. Maybe just t 5 having somebody write a paper. 6 MR. CUNNINGHAM: Yes. 7 CHAIRMAM APOSTOLAKIS: That will be reviewed 8 ertensively by the.poers. 9 MR. HOLAHAN: I don't think it lends itself to 10 being in this reg guide. 11 CHAIRMAN APOSTOLAKIS: Oh, no. No. No, but peer [ 32 reviews, you know, I've participated in a lot of peer 13 reviews, and I know there are peer reviews and peer reviews. 14 So just by having a peer review doesn't mean much to me. ( 15-There are some peers that are better than other peers. Or 16 they're given different charge than other groups, you know. 17-MR. CUNNINGHAM: As I said when I started this 18 slide, I said that we put this into the document because we 19 had had, in a sense, a reference to new Reg. 1602 was our 20 method of dealing with this issue of quality and scope. 21 1602 was-probably far beyond what we needed to talk about. -22 This is a cut, at least, of what we see here at 23-some of the issues. We are kind of, you know, hoping that 241 we can get some feedback from the committee, tnat are there 25' other issues with respect to quality and scope that we ought () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,_ Suite 300

Washington, D.C. 20005 (202) 842-0034

-__..-.____.m_ I 359 j t 1 to touch upon in 1061? I think, - this niay not be -- I don' t j () 2 think we are completely comfortable yet that this is a i 3 sufficient set of issues to talk about in 1061. 4 CHAIRMAN APOSTOLAKISt-I find that the potential f 5 problem here is, as I said yesterday, you are using the 6 baseline CDF and'LERF,.plus the changes, to make a decision. 7 So if people raise a question regarding our ability as a 8 community to estimate the baseline numbers, what do we do? 9 The criteria, the QHO's themselves, CDP and so on, 10 are supposed to be numbers that are applicable to all modes, l 1 11 all considerable failure modes, except sabotage and so on. 12 So -- you wera about to say something. I 13. MR. PARRY:

Yeah, I think it is an overstatement l

' I4 'on Dr. Powers' part to say that we don't have the techniques s_/ 15 to do this assessment of the total core damage frequency, 16 for example, I think we do, and they have been done. You 17 can argue about whether they have been done very well, or 18 whether some parts of them, conservatively, you know, is compare to the others. But I think we are not totally in 20 ignorance of these other areas. 21 So -- but the real problem is that, I think, to do 22 -- well, he had a two part comment, actually. One of them 23 was that we don't have the techniques, and the second part 24 was that the. staff doesn't-have tl'e experience to ~ review - 25' those, even if they.were presented with them. ? () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,. Suite 300 Washington, D.C 20005 (202) 842-00.4-y, -n .-w -~,.,,:.,- e,;., ...v.y, y.. ..m.,,,.. y ....,,_+ry..- ..-y ,--,.,,,,,-.,.,,w- ,wm,.e y-,, - - -,,, _

t f 360 1 And I think the experience in the industry with i '() 2 the missing parts of the ar,alysis is not that great. That 3

is certainly true.

But it is wrong to say.that the { 4 techniques don't exist. [ 5 ' CHAIRMAN APOSTOLAKIS:. Well, speaking of shutdown I 6 and inw power, I mean we have had the two studies that 7 Sandia and Brookhaven have done. 8 MR. PARRY: And there have been several industry 9 one that have been done, too. 10 CHAIRMAN APOSTOLAKIS: Yeah. Those were, though, lL1 limited in scope. Is that correct? i 12-MR. PARRY: Yes. t * 'R. CUNNINGHAM: No. l 14 CRAIRMAN APOSTOLAKIS: Your studies. 15-MR. CUNNIrJGHAM The two PRA's we did for shutdown 16 operations, there was a screening analysis of all the 17-operating, plant operating states that are associated with 18 shutdown operations, coming down in power and going back 19 up. So at a screening level, it was covered, the -20 waterfront, if you will. In terms of initiators, it 21: included fire and seismic and things like that as well. So 22 it-wasn't -- I wouldn't call those limited scope analyses. [ 23 CHAIRMAN APOSTOLAKIS: But they are detailed 24-

analyses.

25l 'MR.-CUNNINGHAM _ Detailed analysis for-one plant ' ANN RILEY &~ ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,- Suite 300 WLshington, D.C.-20005

(202).842-0034

. ~, - ~,---~~-n,-- - -. - -,...,,,, -, - m-~..

361 1 operating state that just -- for each one that seemed to () 2 jump out as the most significant. So if there is a 3-limitation, and'certainly Dr. Powers has made this point 4 Lefore, there was certainly a limitation that we didn't do 5 ev*1nsive PRA studies on each of the operating states. 6 DR. SEALE: And I believe 't our last meeting we 7-found that there was come -- or we heard that there was some 8 reason to question whether or r. 't the screening had beer.

9 fully successful in identifying;the reported sequences, or 10 failing to identify significant sequences.

11 MR CUNNINGHAM: It must have been a meeting I 12-missed, so I am not sure. But, clearly, it is a screening 13 and that is what it was. 14 DR. SEALE: Yeah. 15 CHAIRMAN APOSTOLAKIS: Now, the industry has done i 16 a more complete job, right? I don't know. 17 MR. PARRY: Not as a whole, certainly. 18 CHAIRMAN APOSTOLnKIS: Seabrook, the Seabrook 19 folks have done low power and PRA. 20 MR. PARRY: Right. 21 CHAIRMAN APOSTOLAKIS: Is that, have they done 22 more than Brookhaven and Sandia?- 23 MR. CUNNINGHAM: I kould guess it would not be 24 more. It's has been a while since I have thought about the 25 Seabrook study, but I wouldn't characterize it as j ANN RILEY & ASSOCIATES, LTD. '/ Court Reporters 1250 I. Street, N,W., Suite 300. i Washington, D.C. 20005 (202) 842-0034

i 362 1 . substantially more or substantially less than what was done () 2 by the staff studies, i 3. MR. HOLAHAN: You also can recall that the three 4 NEI pilot plants, San Onofre, Arkansas and South Texas, are j 5 committed to doing shutdown, full scope PRA's. -Okay. [ 6 External, internal, external and including shutdown. [ 7 CHAIRMAN APOSTOLAKIS: Is this a new development T8 now? 9 MR. HOLAHAN: Yes. 10 MR. KING: Well, no, we talked about it. It's the ull NEI initiative where they are going to do the full scope 12 PRA.- 13 CHAIRMAN APOSTOLAKIS: But that will be 14 everything, all balls and everything. 15 MR.-KING: That's almost -- now, they are all, all 16 of those are PWR's. We have been hoping they would throw a 17 BWR in there but 18 CHAIRMAN APOSTOLAKIS: In fact, speaking of that, 19 are we going to discuss this at some time, you know, what 20 the staff is expecting to get out of this initiative? I 21 mean I understand they expect you to bless it in some sense. 22 They are not going to go ahead unless they have some sort of 23 blessing. 24 MR. KING: They are concerned that it going to 25 take quite, you know, a significant effort on their part ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202)1842-0034 ,..,4 .,.,,...~.,.-_,,...,n.,,,.. ~ , ~,,. .,,._.m.,~,-,,nn~., ,...Jw

363 -1 CHAIRMAN APOSTOLAKIS's 'Sure. to do theselthree fril scope ~PRA's. 2 MR. KING: 31 They would like some assurance that what comes out the'other 4-end has~a reasonable chance ~of leading to some turden___ 5- ' reduction. So-what they have proposed is three pilot' f 6 proposals where they are taking what they consider 4 7-Sr_elatively simple changes as examples of the kinds of things '8 - that are going to come out of this study, and they want to 19 -submit them_as pilots to us. And they want to use'that to 10' get.a warm feeling that we are willing to process those kind -.1 11 of ventures in a' reasonable time-frame with a reasonable-12 effort on their part. l 13- -And they say, given that, then they are willing to ~ -14 .go ahead and invest the rest of the money for the full scope l 15 PRA. And we are negotiating now the' schedules and scope of 16' those pilots, as well as the criteria they are going to use 17 for the full scope activity. 18 But, yeah, if you would li:ce to be briefed on 19_ _this,-we could arrange'a briefing. 20 CHAIRMAN APOSTOLAKIS: What do you think about-21- .--we are' talking about'the Bob Christie, the so-called Bob 22- -Christie initiative. They are negotiating what the staff 23. would be willing to accept.or see. '24-MR. KING: :Ou'r next meeting with them-is on 25 November 24th. It is a public meeting. If you can ' t wai t c . ANN-RILEY & ASSOCIATES, LTD. Court-Reporters-1250_I Street,-N.W., Suite 300 Washington,-_D;C. 20005 (202) 842-0034 g m ,,_,_m. _ - a, .1.~..

e .,JRr .A ---h4

• =--.4-.-

,i,A .- N A 5--.A"_A.--m.'tA JKa-J .a 4

W4

.a --A4' b-. 364 1- - for youriown subcommittee,1you ure welcome to attend that. -2 CHAIRMAN APOSTOLAKIS: No, I won't here though. 2 MR. KING: You won't be here.. - 4: CHAIRMAN APOSTOLAKIS: Maybe,.I don't know,_ Mike, 5-can you go? 6 MR. MARKLEY: Yeah. What date was it again? 7 MR. KING: November 24th. It's a Monday,-it is -8 going to be in the afternoon, starting at 1:00. It's 9 - somewhere -- it's here at headquarters, I don't remember the 10 room number. 11 MR. CUNNINGHAM: So, if you we like, we will just 12 ' inform Mike. 13 CHAIRMAN APOSTOLAKIS: Yeah. Yeah. 14 MR. PARRY: George,'to get back though to this 15 full scope issue,' remember that part of the -- with the 16 modification to the acceptance guidelines, that, in a sense, 17 removes the need to at least assess the baseline on the full 18 scope. 19 . CHAIRMAN APOSTOI AKIS : What modification is that? 20 MR. PARRY: The modification of having that very 21 i smr.11 region below 10 to the minus 6. .22 CHAIRMAN APOSTOLAKIS: Yeah. 23 MR. PARRY: Well, a lot of applications could be = 24 - in that region. 25 CHAIRMAN APOSTOLAKIS: But even there, you need to f ANN RILEY & ASSOCIATES, LTD. i '7: Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

365 l i know that they_are below 10Eto the minus! 3, I hope. You () L2 usedito;have it-open-ended,Lbut you really didn't mean that. 3~ 14R. PARRY: That's true. But whether -- whether 4-one needs to do a formal assessment to do that. 5 MR.-HOLAHAN: I don't think that is an issue for 6 individual license amendments, any more than it is -- I mean 7 we have process 1,000. license amendments a year, okay. We 8-think we have an idea that the plants we are dealing with 9 are not 10 to the minus 3 on a day to day basis. We don't 10 stop at-every license amendment and ask them, by the way, 11-are you still below 10 to the minus 3? 12 And'I don't think necessarily that we are going to 13 do that, you know,1 for these appilcations either. 14 CHAIRMAN APOSTOLAKIS: To recapitulate. The / 15 answer is, first of all, that you don't agree with him that 16 the situation is so bleak, right, that we do Pave 17 information? 18 MR. PARRY: We do have information, right. 19 CHAIRMAN APOSTOLAKIS: And the second, is there a -20 second part to the answer? 21 MR. HOLAHAN: I have a second part of the answer. 22-I don't agree-with him on the bleakness, which I think, 23 really, you could assess in terms, not of impossibility, but 24 as-a. level of uncertainty you have left after you do the 25_ analysis. ANN RILEY & ASSOCIATES, LTD, D;f- . Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 m ,-,r n -+ g"

366 -1_ 'And it seems to me that I -- even recognizing that p. Jt c-2. there are: things that were not1 analyzed and there can be %/

3 large: uncertainties in some areas, I don't think that-4-

changes what you want-to do. 5 It seems to me that what you want to do is still 6 make, you know, an integrated decision with the best risk 7 insights you can bring to that decision. Whether it is very 8 limited information or,-you know, complete quantification 9 with very small' uncertainties. It seems to be me it doesn't

10

-- in some ways Dr. Powers is suggesting that his concerns 11 about what we know and what we don't know will derail the 12-process, and I don't see that at all. 13 It.seems to me that our objectives are not -- not, 14 - you know, the gold-plated PRA. Our objectives are to make 15 the best decisions you can, with the.best information you R16. have got. 17 And, you know, to the extent that there is limited 18 information-in some areas, well, that influences your 19 decisions, but it.doesn't, to me, it doesn't stop you from 20 using risk-analysis or_ risk insights. 21 DR. KRESS: It may be you even identify places -22 where you'need to do more work. 23 MR. HOLAHAN: Yeah. -24. DR. KRESS: More research to advance this thing. 25 MR. HOLAHAN: I-mean it seems to me the ( ' ANN.RILEYl& ASSOCIATES, LTD. Court Reporters 1250 I-Street -N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

~ '367 11-alternative 11s worse. -~ 2 DR. KRESS: Yeah, not doing.any. -3 DR. SEALE: -This is a candidate -- or-these are-4- candidates # for the-inevitable upgrade of-the quality ast 5. _ experience--is accumulated.- That'isreally what he has done 6 here.. I 7 MR. KING: And, if you recall,_the-Commission has 8 charged the Office of Research and given us resources to do -9 more detailed risk studies on low ~ power and shutdown. 10 DR. SEALE: Sure. Sure. 11 MR. 10353 : They recognize the fact that this is an 12 area that has got less information than we would like to 13 -have. 14 MR. HOLAHAN: It can't be worse than what we have 15 been doing before. All right. And so I think his note is, 16

in.my. view, too pessimistic, in that, sure, there tre lots 17

'of limitations to what we-know, but the objective is to.take .18 - what you know, recognize what you don't know, and make the 19 .best decisions you can. And I think we are putting in place L 20 a-process to do that. 21-DR. SEALE: Clearly, you don't want to be too 222 ' timid in this-process. 23 MR.-HOLAHAN: No, If.you are too -- well, you 24 know, whether you act or don't act, whether you make a 25 conservative decision or use the numbers, every decision has ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250:I-Street,' N.W., Suite 300

Washington, D.C. 20005 (202) 842-0034

368 ~ 1-consequences. Even the decision not'to act has .( NJ_ 2 consequences. r-v 3 DR. SEALE: That's is correct. 4 CHAIRMAN APOSTOLAKIS: That is a statement that 5-can be debated for a long time. But let's not do that. 6 [ Laughter.) 7 MR. CUNNINGHAM: If I could come back a moment to 8 the point that we were talking about a little bit earlier on 9 kind of where are we in terms of strengths, weaknesses and 10 PRA's and shutdown being one issue there. .11 I should mention that a couple of weeks ago, the 12 Office -- another thing that the Office of Research has been 13 asked to do is to expand our international cooperation in 14 -PRA research. And a couple of weeks ago, there was a (_,/ 15 meeting where, kind of a kick-off meeting. We had 15 or 16 16 countries represented. One of the things we did there was 17 discuss what people perceived to be the big issues in terms 18 of future PRA research. In a sense, what are the weaknesses 19 of current PRA? 20 And the topics that came up, I am trying to recall 21 now, shutdown was one of them. Fire, risk analysis, human 22 reliability, including -- 23 MR. HOLAHAN: Digital I&C. 24 MR. CUNNINGHAM: Yeah. HRA, including management 25 and organization influences. O- ) ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

369 l' CHAIRMAN-APOSTOLAKIS: They have been saying that j )l 2 foril5l years. 'es. _And' digital'and -3 MR. CUNNINGHAM: Yes. s -4 software reliability and-risk issues ~were-the biggies. 5: -As a next step in.this program, we have committed 6 that.we would trylto get the researchers-together involved 7-in-shutdown and digital systems risk and fire risk. No, I 8 am sorry, not-fire -- management and organizational factors-9 risk. To try to understand better at the researcher level, 10 what are the details of the issues and how can be: 11 collaborate more, internationally, in help to quit just 12 talking _about it and do something about some of these 13 things. So for what that is worth. 14 The issues that came up'in-this meeting, from-a - 15 variety of" count a, seemed to be the same issues that we 16 have been talking about here, in terms of HRA, and shutdown 17_ and fire,-and that sort of thing. For what this is worth. 18 CHAIRMAN APOSTOLAKIS: Okay. 19 MR. CUNNINGHAM: And -- 20 CHAIRMAN APOSTOLAKIS: Yeah,'let's move on. 21 MR. CUNNINGHAM: Okay. We have tried to go back 22 and look at what -- and better define what we meant by 23_ = management attention, cn: increased management attention in 24' the Reg'. Guide. 25: .We had kind of a long list of items in the first ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 -(202) -842-0034-

4 370 11-draft,Lin'the' draft for comment. We tried.tx) consolidate ( ~2 uthem'and.to make it a little more clear in the present 3 ' version that-you have, and, basically,, it comes down to the L4. bullets that'are on-Slide 21. ~ l5 .We are-looking at,.first of all, the issue of 6 cumulative impact. And what, in risk management, what -- we 7 are evaluating -- in evaluating certain CLB changes now, 8 'what has the licensee been doing ja terms of risk-management first of the proposed CLB 9 in the past? Is the first 10- ' changes? Is it one of many involving -- involving 11-increases? Is it a -- it is a more balance type of thing. .12 That sort of thing would be brought to consideration of 13' . management. Again, it is -- that would apply for CDF and 14 for LERF. %,) 15 The bullet that we had before that remains is the 16 impact of proposed changes on the complexity of the issue, 17 the burden -- operational complexity, I'm sorry. The burden 18 on the operating staff of the plant, and overall practices- '19 in the. plant. We have discussed before that we didn't want 20-to have. programmatic, tradeoffs between hardware and 21 cprogrammatic types of parts of the plant and practices. 22 Then,' finally, other plant-specific factors, 23 including the siting. Recent inspection findings, 24 performance indicators and LER's from the plant.

25-This is the place where we had talked yesterday y)

ANN RILEY & ASSOCIATES, LTD. f( Court Reporters 1250 I Street, N.W., Suite 300 ' Washington, D.C. 20005 _(202) 842-0034

~ 371. 1: ~about_ including in here information froc a Level 3 PRA. (]3 2 That it could be part_of siting factors, if you will, or 1 31 something -_-That this could be-a. place.to consider, if it is 4 .a~high population' site versus a low population site, or the-5- _ impact of population and' weather'together in terms-of 6 looking at Level 3 PRA. So this could be a place where we .7 could add something here too deal with that issue. 8-DR. KRESS: Now, when we go deeper into the 9 darkness of this gray-area of increased management

10 attention, which of those is it that varies?

Is it the last 11 two' bullets? 12 MR. CUNNINGHAM: Yeah. Perhaps the one that 13 varies the most is the last one. 14 DR. KRESS: The last one. -O D 15 MR. CUNNINGHAM: Yes. As they get closer to the 16 bright lines, and the dark areas, then -- 17 MR. KING: And there is another one that is not 18 shown here that we are still kicking around, and that has to 19 do_with the closer you get to the decision guidelines, maybe 20-the more you ought to think about what is the benefit that 21 is being accrued by this change. Sort of.a cost benefit or 22 regulatory analysis guideline kind of consideration. 23~ It is not stated in here at this point, but we are -24 kicking.that.one around, as well as something that would-25 .come:into play as you get into the grayer and grayer areas. l ( ANN RILEY_& ASSOCIATES, LTD. Court Reporters 1250 :I Street, N.W.,- Suite 300 Washington, D.C.-20005 -(202) 842-0034-

.. - ~ _ ~. ~.. _ -.. -... 372-i 1 CHAIRMAN APOSTOLAKIS:- Incidentally, we are done= l '. 2 - Lwith the bright lines?

Yeah, I~ don't see any.

3' Coming back to you. What is negligible -- what,is 4 small and_so on?_-It is the numbers that you-had in that 5 - previous.viewgraph that-we surveyed? Can we spend five 6 minutes to make sure-that everybody understands that? L7= Because I am not sure that we actually-addressed that. 8-MR CUNNINGHAM: Yes'. 9 CHAIRMAN APOSTOLAKIS:.For example,.I got the comment yesterday from Dr. Miller that-he would like the

- 11 very small, the line that defines the very small region to 12

>be three or four times 10 to the minus 6. So -- 13 MR. CUNNINGHAM: I am trying to go back and find 4 14 it. 15 CHAIRMAN APOSTOLAKIS: Yeah, you have the 16 transparencies here? 17 MR. CUNNINGHAM: Yes. 18-CHAIRMAN APOSTOLAKIS: Okay. 19 MR. CUNNINGHAM: I just have to find them. 20. MR. HOLAHAN: It was number 13. 21 'MR. CUNNINGHAM: It-is'over here. -22 Number 15. -23 CHAIRMAN APOSTOLAKIS: Yes. Now you are talking _24 -in the text someplace about ---I think, first of:all, the 25: shading perhaps should not be!--

O g;

ANN'RILEY & ASSOCIATES, LTD. Court Reporters-.. 1250 I Street, N.W.,-Suite 300 Washington, D.C. 20005'- (202) 842-0034 .=

373 1 DR.-SEALE:- The deltas don't line up. (f~ 2 CHAIRMAN APOSTOLAKIS: The other_way -- 1 3 MR. CUNNINGHAM: Trying to do it backwards. ~4 CHAIRMAN APOSTOLAKIS: That's it. There you go. -5 Move it to the right a little bit.- That's it -- that's it. 6 'DR. KRESS: So Region II is all -- ] -7 CHAIRMAN APOSTOLAKIS: What I call negligible -- 8 MR. HOLAHAN: It obviously takes a coordinated 9 team to make these decisions. 10 (Laughter ] 111 DR. SEALE: Integrated decision-making. 12 CHAIRMAN APOSTOLAKIS: With advice -- 13-DR. KRESS: Did Dr. Miller have any technical 14 basis wanting to move the thing to 3 times -- ' O \\/ 15-CHAIRMAN APOSTOLAKIS: No, that is a policy issue. 16-He thinks it is too small. 17-DR. KRESS: I know. Certainly could be a 18 technical basis having to do with being able to predict it 19 or something like that. There could be a technical basis. 20 CHAIRMAN-APOSTOLAKIS: Well, one argument that can 21 be advanced is-that you can have a CDF,-say, of 10 to the 22 minus 5, and by changing it to one point -- two 10 to-the 23 minus 5, you don't really change much depending on the . 24' l return you have. J25 DR. KRESS: I don't-call that a good technical EO / t,^j ANN'RILEY & ASSOCIATES, LTD. Court Reporters 1250 I. Street, N.W.,_ Suite 300 Washington,~D.C.-20005-(202) 842-0034

374 1 basis myself, but I think he, as best I recall, was e) 2 concerned about if you apply this region, say, to certain i,x 3 outage times that you get very short times to he consistent. 4. CHAIRMAN APOSTOLAKIS: I don't know. 5 DR. KRESS: For things that are normally 6 granted -- you get such a short time it is not consistent with what we do now, or something like that, but I don't 8 recall -- do you remember why he wanted that -- 9 CHAIRMAN APOSTOLAKIS: No. As he was leaving he 10 told me he thinks it's too small, 10 to the minus 6, that's 11 all. That's all he told me, so I assurt.ed -- that is why I 12 gave you this argument -- that if you are already 10 to the 13 murls 5, you know, 1.2 versus 1 really -- I mean it's not 14 the technical argument but it is a numerical argument. (~)/ l i, 15 The technical argument is that you really don't m 16 have models that allow you to make a distinction between 1.2 17 and -- 18 MR. HOLAHAN: That is the point I agree with. 19 DR. KRESS: That's the point I agree with. 20 CHAIRMAN APOSTOLAKIS: On the other hand though, 21 let me finish the argument, it seems when we put this 22 together with gray areas, it says there's NRC Staff 23 scrutiny, but perhaps we meant more than that. I still 24 don't like bright lines even if they are buried in a gray 25

area, rx

( ) ANN RILEY & ASSOCIATES, LTD. / Court Reporters 1250 I Stree; N.W., Suite 300 Washington. D.C. 20005 (202) 842-0034 1

. ~ - 375 ~ 1 :- I think;the grayfarea-should indicate also that 10 i ,~ -2 to:the minus 6 really we don't mean 10 to the minus 6. We-

3 -

mean1something;:somewhere there.

L4-Now you guys-had affigure some time ago where you 5 actuallygmade those lines fuzzy. -i 6 MR.- KING: -Yes.

-7 CHAIRMAN APOSTOLAKIS

And that will probably be. 8 more accurate'than this. 9 MR. PARRY: It's pretty difficult to see-it.- 10 CHAIRMAN APOSTOLAKIS: I think people though '-11 should realize that these are really not bright lines,fso 12 the gray includes that. It's not just a scrutiny. 13 MR. PARRY: _Right. 14 MR. MARKLEY: Actually, I liked Gary's focus on it () -15 being regions for-decision-making rather than focusing on 16 the lines. 17 CHAIRMAN APOSTOLAKIS: Yes, but what does that 18 mean? I mean if the lines define the regions -- 19 -DR. KRESS: That is not a very big conceptual 20 step, it seems to me, like to look at a line and say it 21~ represents a fuzzy. thing. . CHAIRMAN-APOSTOLAKIS: It is not a conceptual 23 -step, Tom, until you-realize that there are pecple are

24 scared and when they see two times 10 to the minus 6 they 25

'say that's above the--line, I reject it. ANN RILEY & ASSOCIATES,'LTD. -Court Reporters 1250 I--Street, N.W., Suite 300 Washington, D.C. 20005 (202)'842-0034

~. 376 g. -l' HDR. KRESS - Tell them not to. ~g 2' CHAIRMAN APOSTOLAKIS:- If I find them-in the-(s-)- 3-cafeteria, I will. 4' (Laughter.] 5 MR. HOLAMAN: I don't think.it matters'how broad a e '6 brush we use. You'll. notice that the-first time we showed 7 -- the 10 to the minus 4-line, at least one owners' group came 8 back in to show us that all their plants had been just 9 recently reanalyzed and rethought and they were all below,10 10 to the minus 4. 3 11 DR. KRESS: Gee whiz. 12-DR. SEALE: Could I ask a question about another 13- . caveat ~that you laid on us here the other day, and that was 14 something to the effect that you didn't want to introduce 15 another dominant sequence. 16 MR. HOLAHAN: Yes. 17. DR. SEALE: Now lee's suppose you had a CDF.of 10 18 to the minus 5, and someone came in with a request for a 19 Region II change with all that Region II implies -- 20 MR. HOLAHAN: Oh, you ought not to imply that we 21-can't function with a fuzzy curve. 22 DR. SEALE: No, no. My point is that in the limit 23 there was a Region II_ request for a 10 to the minus 5 24. addition. 25: Now clearly that is a major sequence that has been

i. f% -

p ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street,_ N.W., Suite 300 L Washington, D.C. 20005 (q; (202) 842-0034' l' L-

uv "i 377 1 'added; a dominant' sequence. g-.- 4-. MR.;HOLAHAN:- Yes.; %/ 3-lDR. SEALEs-And you wouldn't do~that? 4 MR, - HOLAHAN: Well -- it is at the limitlof what-i 5-

we might do.

6 DR.'SEALE: Okay. -Okay.-- so that really is what 7 defines the fuzziness of the fuzz is that 8 MR. HOLAMAN: Yes. 9 CHAIRMAN APOSTOLAKIS: That is why they have 10

scrutiny, 1

'll DR. SEALE: I just wanted to reconfirm that you 12 were reading from the same page. 13-MR. HOLAHAN: I think so. 14 CRAIRMAN APOSTOLAKIS: Can you point to me where 15 in-the text you make it clear that-10 to-the-minus 5 is not 16 10 to the'minus-5? I remember seeing it someplace but I 17 want to read it again. 18-MR, HOLAHAN: While they're looking can I go back 19. and try to answer the question about how we picked the 10 to 20 the minus 5 and.10 to the minus 6? 21-DR. KRESS:

Yes, a

22 MR. HOLAHAth I don't-think of them as -23. percentages. To me if the absolute steps are meaningful, I 24 .think for the current generation.of nuclear power plants, 25- 'another 10 to the minus 5-is-an-important change, so whether 7 7- ~~J(,3) _ ANN RILEY & ASSOCIATES, LTD. -f Court Reporters-1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

~ .. - -. ~ ~. -. -.... _ - -. -.. -.. 378 11 the plant is'at 10 to'the minus-:7 or whether-it is at five j h [ 2-times 10 to1the minus 41--- 3 - CHAIRMAN-APOSTOLAKIS: So Mr. Holahan does not 4: subscribe to the-view that-the PRA numbers can only be used-5 in a. relative senseJand I think that is a great thing. 6-You think that absolute numbers mean something~ I 7-fully. agree with you. 8 MR. HOLAHAN: Yes. When they become a certain 9 size, I think they do. l10 CHAIRMAN APOSTOLAKIS: Yes. ~ 11 - MR. HOLARAN: And likewise I think 10 to the minus 12 6 doesn't rise to-that level -- 10 to the minus 6 you are 13 talking about they are sufficiently small that they are in 14-most. cases either not dominant sequences or you are talking ) His 'about a relatively small. change to a dominant sequence. 16' MR. PARRY: Of changes spread over several ~ 17 sequences even, so -- 18 CHAIRMAN APOSTOLAKIS: I. fully agree with what you 19 are saying. I think this is the way these. numbers should 20 be--- 21 DR. FONTANA: I agree-with the approach and I .22 agree with what you arc saying, but your identification of 23 10 to the minus 5 is a major change. -24 It doesn't necessarily mean that-you have full 25 ~ confidence?in the bot' tom number. () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250.I Street, N.W., Suite 300 Washington, D.C. 20005 -(202) 842-0034 = e =., s- .w m r i-m t- .. ~--

t 379 1-You know that 10-to.the minus 5 is similar to ) 2 other 10 to the minus 5s which are major changes. 1 .3 MR HOLkHAN: Yes.

- 4 DR. FONTANA

But remember we didn't call them,-we -5 don't say 10 to-the minus 5 7.ra major changes. 'We-said that 6 they are -- in fact, we cal'..ed them small, but they.are just 7 at the upper limit of smalt. 8' DR. - SEALE: They are at the level of becoming a-L 9 dominant sequence in a 10 to the minus 4 10 DR. FONTANA: In any plant. 11 DR. KRESS: Now if you 10 to the minus 6 changes, 12 you suddenly have a 10 to the minus 5 cnange, are we scing 13 to require them to track the changes in Region III? 14 MR. HOLAHAN: Yes. Well, I think we have some -- 15 they will be tracked. Whether we will require licensees to-JU5 track them or whether we will simply keep our own database I 17 think is a discussion we are having. 18 DR. KRESS: You will be notified on any changes. 19 MR. HOLAHAN: Well, remember -- no, what we are 20 talking about it if it has 10 to the minus 5 changes, we are 21 talking about things that we have approved, right? You 22 know, the current regulation and the 50.59 and all that 23 don't allow licensees effectively to make risk increases on 24 their'own, so all Tna have *.o do is keep track of things that o25 we have approved.. Not so hard to do. ANN RILEY-& ASSOCIATES,-LTD. " '\\+ Court. Reporters 1250 I Street, N.W.,' Suite 300 Washington, D.C. 20005 (202):842-0034

380 11 Now you could argue that ) 2 DR. KRESS: Now we are_ going to change 50.59, 3/ right? 4 MR. HOLAHAN: Well, if we change 50.59, and we say 5 that the level that we pick-is sufficiently small, then I am 6 not all that worried about accumulating a hundred 10 to the 7

minus 7s.

8 DR. KRESS: I wouldn't worry about that either. 9 MR. PARRY: George, the section you are looking 10 -for, if you are interested, is on page 16. 11 CHAIRMAN APOSTOLAKIS: Okay. 12 MR. PARRY: It's the first paragraph under 13 comparisons with acceptance guidelines. 14 CHAIRMAN APOSTOLAKIS: Okay. 16? () 15 MR. PARRY-Page 16, the last' paragraph on page 16 16. 17 MR.-KING: Last paragraph -- 18 CHAIRMAN APOSTOLAKIS: Oh. 19 So,.where exactly do'you have it? What? 20 MR. PARRY: Well, read the whole of the first 21-paragraph there. 22. CHAIRMAN APOSTOLAKIS: Oh, an indication? 23 MR. PARRY: Um-hum. '24 CHAIRMAN APOSTOLAKIS: Yes. Approximate values. 25-Okay. So if I go back to page 14 then, the last bullet,- '(- ANN RILEY & ASSOCIATES, LTD. _ Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 r

381 -1 ! applications which result in increases to CDF above 10 to (3( ) 2 the-minus 5 per reactor year would not normally be 3 considered. Was that too absolute? 4 MR. HOLAMAN: -Well, I don't think so. 5 DR. SEALE: It's not a "may," it's a "would." 6 MR. HOLAHAN: Yes. In terms of regulatory -7 language I think it's more flexible than one normally sees-8 CHAIRMAN APOSTOLAKIS: And the 10 to the minus 5 9 again is to be interpreted according to page 16. 10 MR. HOLAHAN: Right. 11 CHAIRMAN APOSTOLAKIS: An indication. 12 MR. PARRY: And again, stressing that it's not 13 just numerical, the results that you're using. but 14 CHAIRMAN APOSTOLAKIS: As long as people ( 15 understand these thingt I don't think there's any problem. 16 Yes, would not know. So when you train the staff, you 17 should spend some time on this, that these numbers are 18 really fuzzy numbers. But don't use fu y set theory now. -19 MR. PARRY: We try not to. 20 CHAIRMAN APOSTOLAKIS: So are we happy with the 21 figure? Everybody's happy? 22 DR. FONTAMA: Yes. Sure. 23 CHAIRMAN APOSTOLAKIS: The combined figure, I 24 hope. 25 DR. FONTANA: As long as it has the lines on it. 1 I ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

-382 CHAIRMAN'APOSTOLAKIS:: Huh? 12- .DR. FONTANA: As long as.it'has the lines on it. 3. [ Laughter.) 54 CHAIRMAN APOSTOLAKIS4.All right. You come from~ -51 thefold school, too; i I 6; DR. FONTANA:- I would use fuzzy dots, but that's a -7: different thing. 8 DR. KRESSt It'd be a lot better if you didn't 9-have that gray stuff on there. 110 CHAIRMAN APOSTOLAKIS: Now there is a cle.ar change 11 -there in the-gray area when we cross 10 to the minus 3. Is ~12 that something that you guys will leave there? I mean, 13 that's a criterion, really. We're talking about a 14 criterion. s 15 MR. HOLAHAN:- Well, I have a different way of 3 16 drawing it. You just stop drawing the curve. 17 CHAIRMAN APOSTOLAKIS: Stop drawing. 18 MR. HOLAHAN: Yes. 19 DR. KRESS: -Yes,. the curve -- it was incomplete. 20 MR. HOLAHAN: Yes, it becomes undefined territory. 12 1 DR. SEALE: Terra incognita. 22 -CHAIRMAN APOSTOLAKIS: So it's not unacceptable, 23 it's-undefined? ' MR. HOLAHAN: Both. -25 CHAIRMAN APOSTOLAKIS: So you would not have that

f-1(

-ANN RILEY-& ASSOCIATES, LTD. $~ - - ' Court Reporters 1250 I: Street, N.W., Suite 300 Washington, D.C. 20005- .(202) 842-0034

383 =1L ' dark region at all?-- .{e % /- 2

MR. HOLAMAN - Right.

3 CHAIRMAN APOSTOLAKIS: The very dark. 4 MR. HOLAHAN: Right. Well, not this part of it. 5 I don'c are any-reason to -- 6 CHAIRMAN APOSTOLAKIS: I think there.is_a message 7 there. Look at how it bect,mes very dark. 8-MR. HOLAHAN: I don't.think the industry has any. 9 misconceptions about running _their plants above 10 to the 10 -minus 3. 11 CHAIRMAN APOSTOLAKIS: That's correct. In fact, I 12 would make it n little darker to the left. 13 [Laus er. ] _ 14 DR. KPESS: How can you make it absolutely black? (m,/ 15 That's what they need over there, t 16 CHAIRMAN APOSTOLAKIS: There is a command that 17 says 100 percent. .18-DR._KRESS: 100 percent reliable. No light-19 reflected whatsoever.- 20 CHAIRMAN APOS'f0LAKIS: The greatest difficultly '21 here was making sure that the transition was smooth. That 22 was the problem. 23 Okay -- but you.wouldn't do the same thing at 10 24~ to the.minus 5,-I hope. 25-MR..HOLAHAN: No, that's different. I think L (7 sJ -' ANN RILEY & ASSOCIATES, LTD. . Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005' (202) 842-0034 _ ~~

384 1 that'i different.- . r-s\\ { 2' CHAIRMAN APOSTOLAKIS:- Okay. So the only thing x_/- 3: you're droppingfis what we call:there unacceptable, the very- .4-dark'part that-says.-- 5 MR. HOLOHAN: -This'part. - Well, I think this 6 region-stilll exists. 7 CHAIRMAN APOSTOLAKIS: The region will be there? 8 DR. SEALE: Yes, it's the stuff to the right of 10 E9 to the minus 3.. 10 CHAIRMAN APOSTOLAKIS: Oh,;okay, to the right of 11 10 to the;minus 3. Okay. 12-MR. HOLAHAN: Just draw the curve. 13- -CHAIRMAN APOSTOLAKIS: Okay. 14 MR. HOLAHAN: -- about like so. V(~%(-l15 - CHAIRMAN APOSTOLAKIS: Any other changes you would' 16 make?- You will drop the region below 10 to the minus 7, I 17 suppose? You don't need that. 18 MR. KING: Yeah, we don't need to show that. It 19 might-be wortli putting a footnote on the. table saying that 20 these -- the numerical values are approximate values. 21 CHAIRMAN 1APOSTOLAKIS: I like that. That would go '22 'a-.long way toward making me happy. 23 MR. KING: So it's right there all-in one place. -24 CHAIRMAN.APOSTOLAKIS: I really'like that. Yes. 25 _That's a_ great idea. And'I'm not sure you need this bar a /j ANN ^RILEY & ASSOCIATES,.-LTD. ( Court Reporters. 1250;I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

v.- m l 'T 5 385 J1l 'thatTsays1NRC? staff scrutiny. -I mean,--darker and= lighter. ' l r

MR.JHOLAHAN:7?Otherwise you? haven't. defined what:

1 L2; i 3: the gray:is.: I P You can~~1 eave it there if. i !4: ! CHAIRMAN APOSTOLAKIS: 5 -::you want, but tryito reverse the order.- The less should be 6 on the left.

. 7.

MR. HOLOHAN: That's a' good idea. X i 8= CHAIRMAN APOSTOLAKIS. We were' unable to do that. -9 DR.'FONTANA: Just cut it and. turn it-around-and

-iOL

-glue _it on. ill-(Laughter.) 12: CHAIRMAN APOSTOLAKIS: That's a'-- 13 DR._FONTANA -- They told me I was -- 14-CHAIRMAN APOSTOLAKIS: That's why they-want an 15 ' advisory committee. We really give good advice. L-DR. KRESS: Now if you had a plant that would come 17 in and had a CDF of 10 to the.minus 4, and'you automated a 18 delta CDF change of 10 to the minus 6, he's right on that 4 19 point corner there. L 20 MR._ HOLAHAN: -Yes, j x21 DR. KRESS:- It's_either Region II, Region III, or 22 is unacceptable.

23 MR. KING:

We have a lot of flexibility.

24 '

.- [ Laughter. ). 2 5 -- CHAIRMAN APOSTOLAKIS: He's. going to get in a. lot-f- [ ANN RILEYr&-ASSOCIATES, LTD. a _ Court Reporters 1250~I. Street, N.W.,. Suite 300 Wachington, D.C. 20005 '(202) -842-0034-y .. ~,..,, - + -, -c

b 386 l' of. trouble.. 2' MR. HOLAHAN: It has an emissivity of_.9. 3 CHAIRMAN APOSTOLAKIS: _So, going once -- you have 4 to realize, gentlemen, this is a historic momen*, It we say 5 this_is-it, this is it. Is this it? 6 MR. HOLAHAN: This rev zero. 7 DR. KRESS: I don't think. 8 CHAIRMAN ~APOSTOLAKIS: You're not going to .9 surprise us in December with a different figure, so tell us

-_10 -

what changes you're going to make. I want to start drafting fil the letter before that. 12 DR. KRESS:- You know, except for minor adjustments 13 of the lines, which I can see no technical basis for it, I 14 can't see any way the -- 15 CHAIRMAN APOSTOLAKIS: I don't see -- I mean, if 16 you make it clear, especially if you put -- follow Tom's 17 suggestion and have a little note there, which copies-18 essentially what you have in the text, that the numerical 19 guidelines are approximate values and give an_ indication, 20 something to that effect, I-don't think we should be talking 21 about'whether it should be 2 10 to the minus 6 or 10 to the 22-minus 6.- And that's the. idea of the gray actually, also. 23 But,-you know, there is a certain continaity there, and life 24 gets harder as you-move up or-to the right. That's really 25 the message. f f A ~ ANN RILEY & ASSOCIATES, LTD._ Court Reporters 1250.I Street, N.W., Suite 300 Washington, D.C. 20005 _(202) 842-0034

_. _. _ -. ~. _ 387 -1~ MR. KING: RecognizeLthe-numbers.are in the policy. () 12. paper going _up_to the Commission,_that_10 to the_minus-6ois 3 partiof the policy recommendation that's going up,_ and-we -4: have the words in:as touwhy we chose that. 5- - CHAIRMAN APOSTOLAKIS:' And 10 to the minus-6 is a-6 sharp number there?.Is a crisp number? 7 MR. KING: I'd check the words in the paper to L 8 make sure it says -- 9 CnAIRMAN APOSTOLAKIS: To make sure it isn't. '10 DR. SEALE: Is it 1.0-times 10 to the minus 6 or 11 10 to the minus 6. That's -- 12 MR.' KING: I think it's 10 to the minus-6. -13 MR. HOLAHAN: We removed the 1.0's a long time 14 ago. 15 DR. SEALE:. Good. 16 MR. HOLAHAN: Right. In the context of not 17 qurprising you, just. recognize this is-our current thinking. L. 18= ewe have four office directors, four levels of management, 19 and one other committee to consult with 20 CHAIRMAN APOSTOLAKIS: Well, if there is any ^ 'significant change, maybe you should let us know quickly, 21. 22: .before the December meeting. 23. MR. KING: Well, we should know. At the December 24 meeting. The policy paper should be up there, which will havei he concurrence right on_up-_throughlthe EDO. J25 t ) -ANN RILEY & ASSOCIATES,'LTD. Court ~ Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 .(202)1842-0034 c w t-+ eer -et-Tr w-wVDen -

• • T-T-T m'-"

-~ ~. _. - 388

1

- MR. HOLAHAN: But it won't have the delicacies of .(x 1(> f' : '2:

shades of gray of-what the figure'looks like.

3 MR.1 KING: _No, no,.the: figure's not in the policy 4 _ paper. 5 CHAIRMAN APOSTOLAKIS: So 10 to 'the minus _3 is the 6l absolute upper bound? 7 DR. SEALE: Yep. 8 DR. KRESS:- I'm happy with that. 9' CHAIRMAN APOSTOLAKIS: You would shut down a plant 10 that shows higher than 10 to the minus 3? Il DR. KRESS: No. . 12 CHAIRMAN APOSTOLAKIS: No? ~ 1. DR. KRESS: Not under the context of what we're doing now, but I -- maybe later on I would. 15-CHAIRMAN APOSTOLAKIS: No, no, no. You get.an IPE 16- -that says -- 17 MR. HOLARAN: Wait, wait. Let's answer the 18 different questions in a different context. What this says L is~that the staff wouldn't entertain license amendments. 19 20 DR. KRESS: .That's right. 21 MR. HOLAHAN: Okay. And one of the reasons we 22 wouldn't-entertain license amendments is because I think we 23 have other business to do with that licensee. Now whether E2 4 - that business was shutting them down immediately or seeing 25' .whether they're meeting the regula.tions or whatever else is 4-G- ANN RILEY &_ ASSOCIATES, LTD. Court Reporters 1250 I~ Street, N.W., Suite 300 Washington, D.C. 20005- '(202) 842-0034

389 1 going on I think is not answered by the darkness _of this y 3' CHAIRMAN APOSTOLAKIS: No, it's_not, it's not. I-4 know. But that was a separate question. 5 MR. MARKLEY: I think this jumps into the category 6-of speed-limit stuff. You start defining it that way, then 7 they'll start looking at the other lines as limits the same 8-way. 9 CHAIRMAN APOSTOLAKIS: But that's my point. I 10 mean, at some point you switch to speed limit and you draw 11 the line and you say enough is enough. No more fuzziness. _12 I really don't want to see any plans with core damage 13 frequency greater than 10 to the minus 3. 14 MP. HOLAHAN: _ Recognize that that is -- in my mind (D ( _,/ 15-that's a one-siced limit. It doesn't say'that 9 times 10 to 16 the minus 4 is perfectly acceptable. 17 CHAIRMAN APOSTOLAKIS: It is not. It is not. 18 -MR. PARRY: It is a limit of tolerance rather than 19 the speed limit. 20 CHAIRMAN APOSTOLAKIS: Right. Okay. So as long 21 as we all understand this. Okay. 22 MR. KING: Thank you. 23 CHAIRMAN APOSTOLAKIS: So we're back to 22 now? 24 MR. KING: Twenty-two. 25-CHAIRMAN APOSTOLAKIS: No, we are not there. We ,m ). ANN.RILEY & ASSOCIATES, LTD. Court Reporters 1250 I' Street, N.W., Suite 300 Washington,-D.C._20005 (202) 842-0034

b g ' 390 - 17 - are - ' U,f f2{ MR'. CUNNINGHAM 4 Here wetare.-- -31 CHAIRMAN APOSTOLAKIS: ; I'misorry. : I thought _ i,t'-

- 4 :

was=-- 4 -5. MR.;CUNNINGHAM: We are on'22. 16' CHAIRMAN-APOSTOLAKIS: Huh?-- We are on 22. 7' MR. CUNNINGHAM:. _._ l '8' CHAIRMAN APOSTOLAKIS: Yeah, we will be there; 91 too.: Given enough tirl.e. --10s .MR; CUNNINGHAM: Wejtried to go back in the 411: = document and! clarify what we:were talking about in-the '12-context of performance monitoring. We'had comments about 13 - what-is the relationship of this performance monitoring 14 1 process with the: maintenance rule performance monitoring? 3-I 15 'What are we Frying to accomplish here? This type of: thing. 16 So the text-in the document as.it is-now, 17 -basically, is as-laid out in this Slide 22. The goal of the 18 performance monitoring here is that,Eto ensure that no 119- - adverse safety degradation-occurs because of the change that-20-is beings--Etrat in approved, if you will. 21 DR. KRESS: What do you mean by adverse safety 22 -- degradation?" c.23 MR.-CUNNINGHAM: I'm sorry, I-didn't 4 24 _DR.. KRESS: I'm sorry. I am just not sure I know !25' what youimean by; adverse safety-degradation. A safety risk? e. s ,-- [f AuNLRILEYz & ASSOCIkTES, IirD. + ~ 5- . Court Reporters -1250-I Street,EN.W.,-Suite:300 Washington, D C.' -20005= (202)1842-0034~ gy C l g g N w-dw-p w = w

. _ _ _ __ _ _ _ _ _ _ _ _. _.. ~.. _ 391 1 MR, CUNNINGHAM: In a sense, it goes on, and if l 2 -you look in the sub-bullet there that-we are talking about, -3 that we have -- get to the point of having an unacceptable 4 number of, an increase in the number of failures of pieces i 5 of equipment, is what we were thinking about in terms of 6 that, numbers. [ 7 DR. KRESS: Okay. You have further defined in 8 that sub-bullet. i 9 MR. CUNNINGHAM: Yes, that's right. And this 10 performance monitoring is particularly of concern in cases t 11 where we are allowing changes, i'or example, in grade and QA, 12 whero a number, a large number of components in the plant f 13 are having a change made to them, to relax, in this case, 14 CA. 15 The concern is there is something that, while no 16 -- you know, an increase in the failure rate, or the number 17 of failures of an individual component in there is probably-18 not going to make much difference to risk, the concern is 19-that, collectively, we may be doing somathing that is going 20 to have a significantly larger number of failures across the 21 board. And that is what we are trying to protect again, the 22 potential for common cause ad that sort of thin *J. 23 And, again, talking about it in terms of the 24 implementation of it, that if we are more certain about the 2 5 -- types of changes that are occurring and the impact on these, i. ANN RILEY & ASSOCIATES, LTD. Court-Reporters 1250 I Street, N.W., Suite 300 ' Washington, D.C. 20005 (202) 842-0034

i l 392 l 1 that we -- we me.y have a broader implementation. Dut where () 2 we are not so sure of what the impact of the changes are j 3 going to be, it may be appropriate to have a slower, more 4 slow implementation, or have an initial smaller set of 5 components that would be permitted to be changed, and then 6 allow that to expand over time or something like that. 7 Go on and go on to Slide 23. Continue on i 8 performance monitoring. You expect the program, the 9 performance monitoring program, should be monitored 10 commensurate with the safety significance, or safety 11 importance. That we would expect that the monitoring for 12 low LSSC's would be less, could be less rigorous. I am not p 13_ saying that right. Monitoring for low LSSC's would be less 14 rigorous or less intensive than for the those of the HSSC's. 15 And, again, we would expect that you would want to t 16 have timely feedback and that your performance measures 17 would be set up so that you were detecting unacceptable 18 performance before you can really comprising plant safety. 19 DR. FONTANA: As I remember, there were at least 20 three approachea for identifying the safety significance. [ 21 Do you endorse any particular one? You know, like the T 22 fossil vessel and the risk importance and some other one. 23 MR. CUNNINGHAM: I don't know that we have a 24 specific. There is not a universal endorsement, if you 25-will, of one or the other. They all come into play and are () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,_ Suite 300 Washington,_D.C.-20005 (202) 842-0034 .=.

.. ~ _. - _ __. _.._. _. _. _ _ i 393 l 1 used-in different ways. Risk achievement and risk reduction l 2 are complementing each other as opposed to being alternative 3 measures.- i 4 CHAIRMAN APOSTOLAKIS Don't they usually use two 5 of them? 6 MR. CUNNINGHAM:

Yeah, Usually, use --

7 CHAIRMAN APOSTOLAKIS: Fossil vessel plus risk 8 achievement. I 9 MR. CUNNINGHAM: Yes, some sort of a risk 10 achievement. f 11 CHAIRMAN APOSTOLAKIS: Because they measure i 12 different things. l 13 MR. CUNNINGHAM: They measure, right. l 14 DR. FONTANA: In other words, they do it, and if 15 it looks good to you, it is okay. That's fine. 16 CHAIRMAN APOS70LAKIS: Now, all this information 17 is Section 2.5, page 19? 18 MR. CUNNINGHAMt Yes. 19 CHAIRMAN APOSTOLAKIS: There are bullets here on 20 1, 2, 3, 4, 5. Can you tell me where these bullets are? i 21 Okay. They are not really in a bullet form. They 22 are 1,'2,, 3, 4 in the text. 23: DR. SEALE: Page 20, 24' CHAIRMAN APOSTOLAKIS: Page 20, very top. 25 MR. CUNNINGHAM: Yes. ANN RILEY & ASSOCIATES, LTD. - N Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 '(202) 842-0034 - _ L._x,,.-

.~ . ~... -_=__ 394 1 CRAIRMAN APOSTOLAKIS: Okay. So this is as a ~ 2 result of public comments? 3 MR. CUNNINGHAM: Yes. 4 CRAIRMAN APOSTOLAKIS: That you are doing this? l 5 MR. CUNNINGHAM: Yes, that's right. Just 6 CHAIRMAN APOSTOLAKIS: Are there any public t 7 comments that you decided not to respond to, in the sense i 8 -that you are not changing the guide, that have some 9 significance? 10 MR. CUNNINGHAM: There was a comment that we not 11 use 10 to the,minus 4, have an acceptance guideline of 10 to 12 minus 4 and not differentiate. e 13 CRAIRMAN APOSTOLAKIS: For core damage frequency? 14 MR. CUNNINGHAM: For core damage frequency. i 15 CHAIRMAN APOSTOLAKIS: Right. 16 MR. CUNNINGHAM: But we just let it kind of be 17 independent as Gary was saying yesterday, we have kind of 18 compromised on that, and some very small increases would be 19 permitted. 20 Are there others that are -- 21 MR. KING: Changing, that they don't like the CLB 22 definition. We didn't accept that comment. 23 CHAIRMAN APOSTOLAKIS: Change what? f 24 MR. KING: The definition of current licensing 25 basis. People-said they didn't like it, but we haven't ANN RILEY & ASSOCIATES, LTD. Court' Reporters 1250 I Street,-N.W., Suite 300-Washington, D.C-. 20005 (202) 842-0034

~ _. - _. _.. _ - 395 1 accepted that comment or made any change. () 2 You could probably go through, find a number where 3 there was a comment, but we chose not to make a change. 4 MR. CUNNINGHAM: Slide 24 goes, discusses 5 performance monitoring in a little more detail, or continues ) 6 it. Again, we want to make the point explicit in the 7 document that it is monitoring that is being performed as 8 part of the maintenance rule implementation, can be used' 9 under some circumstances. 10 That was a point of contention in the draft. They 11 were saying, you know, do we have to have a monitoring i 12 program here and a maintenance rule implementation? I said, 13 no, under certain circumstances, you can use the maintenance 14 rule implementation, performance monitoring program to O( s/ 15 handle this as well. 16 And, again, related to that is that you want to 17 h ve provisions for specific cause determination and 18 trending of failures and that sort of thing, and to have 19 corrective action. 20 DR. SEALE: I understand that there is an effort 21 underway to perhaps modify the maintenance rule to bring it 22 up, bring it back, or to -- to take care of discrepancies, 23 if you will. 24 I assume that you put a notch of the people that 25 are doing that, to suggest to them that they also talk about () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202)-842-0034

396 1 reciprocity, if you will, on the monitoring process. f () 2 It seems to me the other way to go is to say that-I 3 the maintenance rule monitoring should be -- well, I am 4 saying compatibility between the two processes. 5 MR. HOLAHAN: Yes. And if you look at the options 6 offered to'the Commission, there is a spectrum of I 7 possibilities. I think there were three. One of them was 8 do nothing. But both of the other two options would more 9= closely link the maintenance rule to our risk-informed 10 framework, one further than the other. 11 And I think it is_ fair to say, from Commission or i 12-Commissioner _ questions, so far, the Commission is thinking ~ -13 about perhaps something in between those options. But i 14 something definitely in the direction of more tightly 15 coupling with risk-informed initiatives here. 16 CHAIRMAN APOSTOLAKIS: How about a break? 17 DR. SEALE: All right. 18 [ Recess.) 19 CHAIRMAN APOSTOLAKIS: Okay. Back in session. So 20 we are on 25? 211 MR. CUNNINGHAM: Yes. The rest of the slides that 22 we.have in this package discuss documentation, and even 23 though it covers a number of slides what we have done in the 24 document-is reduce the amount of documentation that would be 25' submitted as-part of the proposed CLB change. ) ANN RILEY'&' ASSOCIATES, LTD. Court Reporters 1250,I Street,_N.W., Suite 300 Washington, D.C.-20005 -(202) 842-0034

i 397 l 1 CHAIRMAN APOSTOLAKIS: Thet's fine, f () 2 MR. CUNNINGHAM: Okay -- 1 3 DR. SEALE: I have a question. Yesterday, we were 4 earlier talking about-what constitutes the current licensing i 5 basis, and I found this footnote that is on page 3 of 1061, 6 and there are some differences from some earlier drafts of l 7 1062 and the other things that we saw. 8 Is this footnote going to be common to everything? 9 MR. CUNNINGHAM: Yes.

Yes, t

10 DR. SEALE: I have one other question. It says 11 here it includes the regulations contained in, among other . 12. things, Part 50 and Part 54 in that listing, and then in the 13 rest of the footnote on the next page it say it also 14 -includes the plant-specific design basis information in 10 15 CFR 50.2 as documented in the most recent safety analysis 16 report, as required by 1050.71. t 17 That is kind of redundant, isn't it? I mean if it 18 has 50 -- if it is 50 then it's 50, I would assume, in its 19 entirety. 20 MR. KING No, I understand your point. We just 21 took this right out of Part 54, 22 MR. HOLAMAN: No, I think in fact it is a 23 restriction. What it says is the FSAR, the total FSAR, has 24 lots of stuff in it, okay?: The portion of the FSAR which - 25 . relates to the design ~ basis of the plant are those things l ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street,-N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 i ,,x.- _ym....,-.m_,,-, ._...m. - _., -.. ~...,,..... - ,., _, ~,,,, _,, _... ,,_,_..._,_.,.-.m..

398 1 defined by 10 CFR 50.2. 1 (g i 4 2 I think it is a subset of the FSAR. i L./ 3 DR. SEALE: And 50.71. 4 MR. HOLAHAN: 71 is just -- or 71(e) I think is 5 what it means is just the update requirement. 6 MR. MARKLEY: Dut isn't the point here also that 7 those continuing words in that footnote are parts of things 8 that are considered current licensing basis under Part 54 9 but are not specifically defined in Part 50.2? 10 MR. CUNNINGHAM: That is correct. There is no 11 similar definition in Part 50. 12 MR. MARKLEY: Right. 13 MR. CUNNINGHAM: So we are using the Part 54 14 definition. 15 DR. SEALE: No wonder we need lawyers. 16 MR. CUNNINGHAM: The remainder of the slides were 17 just what we would have and it's smaller than what we 18 requested before. No questions on that? 19 CHAIRMAN APOSTOLAKIS: One of the things *. hat we 20 have not done is we haven't looked at the SR?, so -- 21 DR. FONTANA: I did. 22 CHAIRMAN APOSTOLAKIS: We just got it. 23 DR. FONTANA: Must have read an old one then. 24 CHAIRMAN APOSTOLAKIS: What, you read 1602? You 25 finished early, () ANN RILEY & ASSOCIATES, LTD. i Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

.~. 399 1 DR. FONTANA: So I can go home. () 2 [ Laughter. ) l 3 CHAIRMAN APOSTOLAKIS: Well, is there anything in 4 the SRP that is inconsistent with what the Guide has? l 5 DR. KRESS: Of course not. 6 CHAIRMAN APOSTOLAKIS: Okay. Well, one of tho l 7 things we have to do with you is what you will do at the 8 December meeting, right? 1 9 How much time do we have for this? .10 MR. MARKLEY: About an hour and a-half -- it's 11 8:35 -- to 10 o' clock. 12 CHAIRMAN APOSTOLAKIS: Do we need an hour and a i 13 half? Do the members present feel that we need an hour and 14 a half? 15 Do you think that we need an hour and a half? 16 DR KRESS: Yes -- 17 CHAIRMAN APOSTOLAKIS: You don't care. Yes? 18 DR. KRESS - Dana'will be there. He hasn't heard 19 this. 20 MR, HOLAHAN: His note may be the only 21 controversial element. 22 CHAIRMAN APOSTOLAKIS: That's probably true. =23 MR. HOLAHAN: At the December meeting. 24 CHAIRMAN APOSTOLAKIS: Is it possible that we get .25-a responsetbefore the meeting? ~ ANN RILEY & ASSOCIATES,.LTD. I \\ Court Reporters 1250:I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 i

-~ i 400 l' MR. HOLAMAN: A response to his note? l 2 CHAIRMAN APOSTOLAKIS: Yes. 3 DR. KRESS: I personally don't think we ought to 4 as); for a response to the note other than verbal. 5-CHAIRMAN APOSTOLAKIS: Okay. Well, there is a 6 transcript of course that he can read. l 7 DR. KRESS: I don't think these guys ought.to 8

write down --

9' CHAIRMAN APOSTOLAKIS: No - - L10 DR. KRESS:~ Every time we get an internal note il like that, I don't think -- i 12-CHAIRMAN APOSTOLAKIS: Well, the thing is that I l 13 would like to start on our way to resolution before the 14 . actual meeting. ( 15 DR. KRESS: I think this may be a case of there is 16 no resolution. They just explain their position and 17 response and then Dana votes the way he wants to vote on -18 letters or whatever we write. 19 CHAIRMAN APOSTOLAKIS: Certainly, well -- and as I 20-say, there is a transcript of today's meeting where the 21 gentlemen already have expressed a reaction. 22-DR. KRESS: I think it would be good if they are - 23 prepared to talk directly to Dana and let him interactLand question and their answers will be useful. . 25L CHAIRMAN APOSTOLAKIS: Directly? You mean what? ANN RILEY & ASSOCIATES, LTD. -Court Reporters 1250 I Street, N.W., Suite 300 02 84 I b3 ,...e ,,.~.,..,,g-

1 l 401 t i Before the meeting? l () 2 DR. KRESS: No, during the meeting. 3 . CHAIRMAN APOSTOLAKIS: Oh, during the meeting. t 4 DR. KRESS: At the meeting. That's why I am -5 saying we will probably need an hour and a half. i t 6 CHAIRMAN APOSTOLAKIS: No, but I would like him 7 also to have an idea what their thoughts are, and he will 8 because-he can look at the transcript, so there is no need 9 for-a written response. I 10 Okay, period -- so we have an hour and a half. I 11 guess you can-go over a reduced version of this? ~ 12 MR. HOLAMAN: Yes. 13 CHAIRMAN APOSTOLAKIS: With whatever changes you { 14 make. -15 Are you going to change the name to combined 16 change requests or you haven't decided yet? 17-MR. KING: I think that sounded pretty good, 18 We'll talk about it, t 19 CRAIRMAN APOSTOLAKIS: CCRs. Maybe it would be a 20 good idea to have a viewgraph with comments to which you 21 decided not to respond, like you gave me-a few examples. 22 DR, SEALE: Yes, okay. 23 CHAIRMAN APOSTOLAKIS: I think we definitely need 24 to have something on the Standard Review Plan because we 25_

haven't_ covered it at all,_so whatever you give us to ANN RILEY'& ASSOCIATES, LTD.

t ' Court Reporters-1250.I Street, N.W.,~ Suite 300 Washington, D.C. 20005 (202) 842-0034 ~,..-..m.,

( 402 1 enlighten us on that, that will help, l () 2 I'm sure there will be a discussion on the figure

that you come up with if it survives.

4 MR. KING: We will present the combined figure. 5-CHAIRMAN APOSTOLAKIS: Yea, but you have several ( 6 reviews to go through, as Gary said,-so I don't know if -- 7 is it a serious kind of situation if any ons of those 8 reviewers says no? 9 MR. HOLAHAN: Yes. It fas last time. 10 DR. KRESS: Is it likely that Region III will 11 disappear -- i 12 CHAIRMAN APOSTOLAKISt Last time? Who derailed it 13 last time? 14 MR. HOLARAN: Well, it was the shape of the curve -15 and the acceptance guidelines I think were influenced by 16 both CRGR, the committee, and by the office directors. 17 Well, in my opinion, some for better and some for 18 not quite as good, but I mean they had influence. 19 DR. KRESS: Let me ask it another way. Is there a 20 lot of expressed concern about Region III? 21 MR HOLAHAN: Tom King and I had diucussed their 22 being, conceptually being the Region III with both at the 23 EDO level and at the Chairman level, but we haven't covered 24-everyone. 25 MR KING: No,1but_at those levels I didn't hear f^h g ANN RILEY & ASSOCIATES, LTD, Court Reporters 1250 I' Street, N.W., Suite 300-Washington,.D.C. 20005 (202) 842-0034-

403 i I any serious reservations. t .2 CHAIRMAN APOSTOLAKIS: About? L 3 MR. KING. About Region-III, the concept of Region 4 III. 5 Yes.- 6 CHAIRMAN APOSTOLAKIS: Oh, you need something-like 7 that. I mean you really need something like that. i 8 DR. KRESS: Yes, we think so. 9 CHAImt4N APOSTOLAKIS: Okay, so you will sunnarize 10 today's presentation. When you say agreement, do you mean 11 with the comments or with us, because we agreed on 12 everything and disagreement with the comments like we 13 already said that. We discussed the figure. We'll5 discuss 14 Dana's points and Chapter 19. That's it as far as I can 15-tell. 1G DR. KRBDS: One reason I think they need to 17 discuss Dana's points is when the committee, if and when '18 they write a report, that will be-a big debating area 19 between us. 20 CHAIRMAN APOSTOLAKIS: Which one? 21 DR. XRESS: Dana's concerns. 22 CHAIRMAN APOSTOLAKIS: Oh, yes. 23 DR.-KRESS: So we will need to be able to have a 24 good understanding among the full committee of what your -25 response is.. i [ ANN RILEY & ASSOCIATES, LTD. ~' Court Reporters. 1250 I Strect, N.W.,-Suite 300 Washington, D.C. 20005 (202) 842-0034

404 _1 CHAIRMAN APOSTOLAKIS: Now we are supposed to () -2 write, I think the Planning Committee decided, two letters 3 to respond separately to the SRM on uncertainty. Is that l 4 what you decided? 5 DR. SEALE: Well, I think that was the thought we l 6 had during our meeting and I think our discussions yesterday i 7 sort of reinforced the idea that we keep the focus on our 8 letters. 9 CHAIRMAN A?OSTOLAKIS: But what they are doing is 10 combining the two, right? You are not planning to do 11 something separate on the use of point values, are you? 12 MR. KING: That is correct. We're not. 13 CHAIRMAN APOSTOLAKIS: So we are using 1061 then 14 as a response to that question from the Commission? ) 15 DR. KRESS: I think that would be an 16 appropriate - 17 MR. KING: And that is what we are doing. 18 CHAIRMAN APOSTOLAKIS: Okay, so in the hour and a 19 half you gentlemen could address that question too and maybe 20 point to the committee where in 1061 you feel you have been 21 responsive to that particular request, particular SRM, 22 right? 23 DR KRESS: They did this pretty well in our t 24' meeting that you missed. They might want to dig out those 25 old slides because they actually addressed _it pretty well ANN RILEY &. ASSOCIATES,-LTD. i2\\ Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005-4 (202) 842-0034 -,,-.-._cm

u,._.

.2-

_ -. _ - - -. ~ _. 405 1 then. () 2 CRAIRMAN APOSTOLAKIS: Do I have those slides? I 3 think I do. Okay. That's it. I can't think of anything 4 else. l 5 What I propose is that we go around the table and i i 6 you give me your input, points to be considered when a draft f i 7 letter is put together. 8 DR. SEALE: I have one typo to mention, okay. In 9 your standard review plan, which we are not going to go into 10 in a lot of-detail here, but-in it, the footnote that 11 defines the licensing basis is not the same as it 11 in the 12- -Reg. - -draft Reg. Guide, and they are within three days of 13 each other in terms -- ru), one day of each other in terms of 14 the date on the front. O(_/ So I am not going to ask you how you are going to l 15-16 reconcile it. I am just to say that is something that would 17 it would probably we a good-idea to straighten out, j 18 We don't need more definitions of licensing basis, i 19 CHAIRMAN APOSTOLAKIS: Now, during the discussion, 20 so we will kave these gentlemen present, okay. 21 DR. KRESS: Sure. If they want to stay. 22 CHAIRMAN APOSTOLAKIS: So clarification needed. 23 Any problem? 24 Do you want to stay? Wold you like to stay? l 25 -DR. SEALE: Sure. O ANN-RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 -(202) 842-0034

-- -~ i 406 1-CHAIRMAN APOSTOLAKIS: You don't have to stay. /; ( s) 2 MR. HOLAHAN: In a risk-averse environment, I 3 think we ought to stay.. l 4 [ Laughter.) S' CRAIRMAN APOSTOLAKIS: This is more than 10 to the f 6. minus 5. -7 Okay. Who wants to go first? Rick, do you want i 8 to go first? This is not for the letter. Just give me your 9 opinion, your judgments. 10 If you don't want to go -- 11 MR. SHERRY: No. Just a coupl' of things. One is 12 .I think.that it is probably a good idea to combins your-13 response in one letter since the response on the uncertainty 14 is so integrated with what is in 1061, I think. 15 CHAIRMAN APOSTOLAXIS: Well, we have two members 16 of the Planning and Procedures Committee here. 17 DR. SEALEt We could be argued with. l 18-DR. KRESS: I think that we would probably come 19_ 'down on two separate responses. Because with two separate 20 -- we have got a specific SRM. 21 CRAIRMAN APOSTOLAKIS: Two specific SRM's,.okay. 22 So that is one point. 23 MR. SHERRY: And I guess my only other comments 24 are with regard to something that is not specific to what is 25: .being done now for the regulatory guides. But the future ((q ANN RILEY &. ASSOCIATES, LTD. 'j Court Reporters-1250 I-Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

i 407 i i 1 work to be done on a re-look at the safety goals. Because I () 2 . think there are some significant considerations which should -3 - be acknowledged with regard to derivation of the criteria, j ~ I 4 You know, we have been talking about it. 5 CRAIRMAN APOSTOLAKIS: Wait a minute now. That is 6 a separate issue. Isn't it a separate SRM on elevating CDF? { 7 MR. SHERRY: But I am talking more generally, f 8 CHAIRMAN APOSTOLAKIS: Yeah. And they said that ~ 9 that will be more general than just elevating the CDF. But ~ 10 that is not part of the two letters we are writing now. 11 DR. SEALE: No. 12 MR. SHERRY: No, it is not. 13 DR. SEALE: But it is something we discussed. 14 CHAIRMAN APOSTOLAKIS: Go ahead. [ 15-MR. SHERRY: No, I was just saying that it must be 16 recognized that there is a possibility that the acceptance 17 guidelines for LERP may possibly be impacted by that 18 activity. Okay. And comewhere that, it might be a good 19 idea to acknowledge that in 1061. 20 For example, if a decision is made to include 21 consideration of population density, societal risk, land 22 contamination, or whatever, within the safety goals, that 23 might impact the acceptance guidelines for LERF. 24 DR. KRESS: I have thought about that some, too, i 25 Rick. In fact, that was one of my expressed concerns. And 1 () ANN RILEY-& ASSOCIATES, LTD. l Court Reporters 1250 I Street, N.W., Suite 300 1 Washington, D.C. 20005 + (202) 842-0034 P w w-- y-ym-- ,e --ey. - - < .-.-m3-.,.-w g ,ig.,.--y ,p- .,y cyy, w --.-=, ,,-,-,,,+----ew.~,-,-e-,% .----.------------,.-~-,-.vwer--,r.

408 t 1 it seems to me like the best way to treat that societal risk t () 2 type of thing would be to incorporate it in some sort of 3 - siting criteria, and separato this from the siting - Take i 4 care of that with.your siting criteria and keep this stuff ) .5 the way you got it. l t 6 - Lecause you are really going to foul this thing up j 7 . when you try to get a LERP that incorporates societal risk j 8 more than it does now. You are going to get LERF's that [ 9 vary all over the place. Or you are going to have one that u 10 10 so badly -- - 11 DR. SEALE: Skewed. 12 DR. KRESS: Skewed to be bounding that it is not 13 very useful to a lot of them. 14 So my recommendation is look to see if you can't 15 hide that other part in the siting criteria and finesse the i 16 issue. But that is just one thought. 17 MR HOLAHAN: That is an interesting thought. i 18 Because I have been somewhat concerned, when we talk about 19 . drawing QHO's or other similar measures into the regulatory 20 process, that you are drawing in information that, in many 21 cases, neither the NRC, nor the licensee, has control over. 22 DR. KRESS: Has any control over, or any way to 23-change or do anything about. That'is my concern also. 24 CHAIRMAN APOSTOLAKIS: Based on what I know now, I 25. would be very reluctant to. start-this revision with the high () ANN RILEY & ASSOCIATES, LTD. Court Reporters-1250 I Street, N.W., Suite-300 Washington, D.C. 20005 .(202) 842-0034 j 7 'i ,irwwt-vv-i v v et,--"-*r-tv er 'u . ww,ww w +vvy re t-- mye'-e v,wgi-g o m,' r e. t w.-wr s.== -w-,ver--m-wu-r =- w+ w,, v w-Je--r--inw-w e eswm ev- -m- +- -- -a, 5

i 409 f 1 . level goals and think about land contamination and societal 2 risk. I don't think-that is a pressing need for the agency. 3 DR. KRESS: I think -- I think those'are 4 reasonable things to deal with in high level goals, but I j 5 wou3d deal with them in my siting. { 6 DR. SEALE: Yeah. I 7-CHAIRMAN APOSTOLAKIS: You will deal with what? 8 DR. KRESS: Within my siting criteria. 9 DR. SEALE: Siting criteria. 10 DR. KRESS: And I would-still separate design from l 11 siting. 12 CHAIRMAN APOSTOLAKIS: Okay.- Rick. 13 MR. HOLAHAN: Just, can I -- i 14 CHAIRMAN APOSTOLAKIS: Yeah. 15 MR.-HOLAHAN: Just complete that thought a little 16 bit. That is, there had always been some controversy over ^ 17 what is the role of policy statements anyway. Our legal 18 staff has never really liked them. They like either, if you 19 vant something done, you put it in the regular. ion, and if 20-you don't care whether it is done, you don't say anything 21 about it. 22 Policy statements are sort of -- 2? DR. SEALE: In between. 24 MR. HOLAHAN: In between philosophical things. 25 Something has happened-between the time of most, between all ANN RILEY & ASSOCIATES, LTD. j Coart Reporters i 1250 I Street, N.W., Suite 300-Washington,--D.C. 20005 (202) 842-0034

410 1 the policy-statements' writing and the current, and that is () ~ -that Congress wrote a law which basically said agencies have 2 3 to have things like strategic plans. And the strategic plan 4 is a sort of philosophical, non-regulation, sort of like a 5 policy statement. And when you read it, it has, at least in 6 shorthand form, a lot of things that look like policy 7 statements. 8 And you might ask, in the long run, whether your 9 policy statements are not just explanations and elaborations 10 of your strategic plan. And they are not, you know, meant 11 to last a lifetime, but they are, you know, they are Volume 12 3, 4, 5 and 6 explaining what your strategic plan is. 13 So, you know, I think there are a lot of things to 14 think about in the policy statement area. 15 CHAIRMAN APOSTOLAKIS: Okay. Rick, any more, 16 anything else? 17 MR. SHERRY: No. 18 CHAIRMAN APOSTOLAKIS: No. 19 DR. SEALE: Well, I didn't -- I hadn't thought so 20 eloqueAtly, or perceptively, I guess, as some of the other 21-people had on this issue. But I had written down the words 22 ballast versus baggage. In trying to characterize that wist 23 list or -- I don't know whether it was a wish list or a bad 24 dream that-you guys had put up there that had things like 25 =landLeontamination-and-so forth on it. And I think we have () ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W.,_ Suite 300 Washington,.D.C. 20005 (202). 842-0034 i

. = 411 1 to be very careful about what we would integrate into any i 7x ) 2 kind of revisitatica on the safety goals. ( 3 And, as I say, I had ballast versus baggage, and 4 that kind of characterizes the way I felt about it. 5 The only other comment I would make is that I 6 endorre the idea of having a very open discussion of Dana's 7 issues at the meeting. My own feeling though is that he has 8 articulated goals that need to be -- that we would hopefully r 9 meet in '.he maturation of riak-informed regulation. But I 10 don't think he has made the case for a call for inaction on 11 the process. I think he is just trying to -- I hope he is 12 trying to stear us in the direction we ought to be going. 13 Other than chat, I think what you have done so far 14 is a tour de force. It is an extraordinary effort. It is 15 clear that everybody hus been very thoughtful, and at the 16 same time, wiling to think large rather than small, and I 17 congratulate vou. 18-CHAIRMAN APOSTOLAKIS: Mike, do you want to say 19 anything? 20 MR. NARKLEY: I just want to mention, looking at 21 two lett ers here, it seems to me that the letter on the SRP 22 and Reg Guide could be a fairly simple letter in terms of is 23 it okay to go fo ward or not go forward and then any 24 comments you might have on the policy issues so then 25 considering that you might want to look a little more [) ANN RILEY & ASSOCIATES, LTD. \\~# Court Reporters 1.'70 I Street, N.W, S.ite 300 Washington, D.C. 20005 (202) 842-0034

t 412 l 1 closely at that draft policy paper before the meeting if you 2 are going to separate out an uncertainty, which is one of -3 those policy issues that might be the foundation for -4 spinning it off to the other letter. 5 CHAIRMAN APOSTOLAKIS: Okay. Mario. 6 DR. FONTANA: Again, like Bob I think you guys 7 have done a tremendous job on this thing. 8 There are soma things that, some comments, i 9 I'm still a little -- not real happy about 10 Appendix B to do Level 2 so that don't forget about issuing '11 the NUREG report that drires'what that is all about. i 12 The question of raising CDP to a fundamental goal, t 13 my opinion and I think I am in the minority, is not to do 14 it. 15 I thin). the way you are using it is just the right 16 way of doing it becauce a fundamental goal should be to 17 protect the health and safety of the public and the CDP is 18 one way of demonstrating you have got defense-in-depth, so 19 it is extremely useful, but I don't think you have to go 20 through this other stuff -- but that is an opinion. 21 I don't really understand why plant design is 22 separated from siting. Does that also apply to advanced 23 plants?- 24. MR. KING: Of course. It was prompted by advanced 25' plants.-- ANN RILEY & ASSOCIATES, LTD. Court Reporters' 1250 I Street, N.W.. Suite 300 Washington, D.C. 20005: 1 (202) 842-0034-

413 r 1 DR.=FONTANA: Because one would think if you'd J 2 come up with a bulletproof plant you ought to be able to put 3 'it on a different site. 4 MR. KING: No -- t 5 DR. FONTANA: Just logically. 6 MR. KING: -- the idea was'we didn't want urban l 7 siting tegardless of the plant design. [ 8 DR. SEALE: Ravenswood is a no-no -- -9 DR. FONT 1.NA: Well, I am not going to go that far. 10 MR HOLAMAN: Well, it is an element of 11 defense-in-depth. 12 DR. FONTANA: It is. 13 MR. HOLAHAN: No matter how good the plant is. 14 DR. FONTANA: Principally to put more of it into 15 the design -- 16 MR. HOLAHAN: Yes. I recently visited the 17 Ravenswood site, just to see what it was like, and you 18 wouldn't want to put a plant there. I wouldn't want to put 19 a plant there. 20 DR. FONTANA: Is there a steam plant on there now? 21 MR. HOLAHAN: No, actually, it is a maintenance 22' yard'about large enough to put a power plant on -- 23-DR. FONTANA: It's good for it. 24 In reading-the Standard Review Plan it looks like 25 a:real good' job and it givesla lot of guidance, but on the ,s '_( ) ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250.I Street,'N.W., Suite 300 Washington, D.C. 20005 -(202) 842-0034

~. ~_.. - ~. 414 1 other hand, it leaves -- as it has to -- it leaves a lot of .2-judgment, it leaves a lot of leeway to the reviewer, which-3 leads to the next question. 4 What kind of training are these reviewers-going to 5-get in PRA? I know you guys are not going to do it 6 directly. You don' t have to answer that., but it is 7 something to be concerned with. 8 MR. HOLAHAN: Well, I can give you that, the 30 9 second version. 10 There are three major courses going on. One is 11-sort of a two-hour introductory lecture that we have been 12 doing and-I think one was done-yesterday so probably about 13 400 of the NRR Staff have been lectured on unct is i 14 risk-informed regulation and what are your responsibilities. 15 Then there is a two and a half day PRA for 16 ' technical managers course, which all NRR managers should 17 take. .I think all of them are scheduled for this year or it 18 might run a little bit beyond that. That is ongoing. 19 Then there is a course for technical reviewers, 20 which is a revised version of what used to be called 21 fundamentals of PRA or a title like that. 22 There's basically a commitment over the next two 23 years, something like that, to have all of NRR's technical i 24 staff take that,-which is about a four or four and a half 25 day course, ' ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street,.N.W., Suite 300 Washington, D.C. 20005 .l (202) 842-0034 .=

415 1 Then there's an additional cource specifically set s ( ) 2 up for inspectorn -- resident inspectors, regional. 3 inspectors, and the few headquarters inspectors -- which is j 4 actually a two-week course. That is a little bit further 5 behind, but the. intent is to get at least one resident i 6 . inspector at each site covered within about the next year, 7 and then in a year after that pick up the second inspector 8-on each site and the regional office inspectors. 9 Sn I would say within two years most all the 10 technical staff in the reactor area will have been touched 11 by~this one way or ar.other. 12 DR. FONTANA: Okay. Scunds good. 13 Have you given much thought on how this would 14 spill over into license renewal yet? 15 MR. HOLAMAN: I cadn't but like all other i 16 technical aspects of the licensing basis, I think it just 17 carries on. 18 DR. FONTANA: It will occur.. 19 MR. HOLAHAN: Yes. 20 DR. FONTANA: That's all I have. 21 CHAIRMAN APOSTOLAKIS: Ton? 22 DR. KRESS: First, I want to say I am real pleased r 23' with this effort. I view this as probably one of the most 24. -important things that the agency has done in a long time-and 25 I am_very pleased with it. [) ANN RILEY & ASSOCIATES, LTD. \\- Court Reporter 7 11250LI Street, N.W.,: Suite 300 Wachington, D.C. 20005 (202) 842-0034 i

i 416 i 1 I think you guys have done a great job and I see () 2 it as a standard for how to do risk-informed regulation, not 3 just this specific application but when we get around to l 4_ doing risk-informed throughout the whole body of what we do, 5 it's_the standard of how to do it, so it's real important i 6 stuff to me. 7 Early on I had three concerns. 8 - One of them was the societal risk concern, which I 9 think-is an important one, but I already expressed how I 10-think that_should be dealt with. 11 The other two were -- I was concerned early-on 12 that a plant with a very low CDF automatically meets the 13 LERF and this could compromise things having to do with 14 containment and mitigation. 15 I think you dealt with that very well with your 16 integrat(d decision process and defense-in-depth 17 requirements, so I no longer have a concern there. 18 The other one that I had a concern with was 19 performance monitoring. It seemed to me like there was a 20 disconnect between the way that performance monitoring was 21. established and the risk-informed process itself. 22 It seemed to me like one needs to fold-that back 23' in and say let's 1cok -- let's make a risk-informed or I 24. risk-based performance monitoring process. 25' I didn't~really see the' risk basis for it. We t ANN RILEY & J.9SOCIATES, LTD. Court Reporters 1250 I. Street, N.W., Suite 300 Washincton, D.C. 20005 (202) 842-0034 -[ m.m.,,- ,_,,.,y 6_ .,e-- y ,m.3 -~,w .c, __-.,,1 ,rm a+. ~., m --.g...,9 v.e,_.,,,-<c.-

'417 1 1 fell back on the old process of how to do performance () 2 monitoring. 3 CHAIRMAN APOSTOLAKIS: Which is why I objected to i 4 Murphy's presentation this morning. g 5 DR. KRESS: Yes, I had a-little bit of problem 6 with that that still exists, but it is not enough of a 7 problem that I want to mak? any issue of it. 8 CHAIRMAN APOSTOLAKIS: So 1061 as it is now has t 9 that problem but -- 10 DR. KRESS: Yes, but I am not going to be that 11 concerned with it. 12 CHAIRMAN APOSTOLAKIS: Okay. 13 DR. KRESS: Because I dnn't want to do anything to 14 derail the process. 15 CHAIRMAN APOSTOLAKIS: I'11 check it. Anything i 16 else? 17 DR. KRESS: No, that was it. 18 CHAIRMAN APOSTOLAKIS: Well, so if we were to 19 draft a series of conclusions and recommendations I guess 20 the first conclusion would be this Guide, assuming that the 21 guys do everything that we discussed, is the Regulatory 22 Guide and the associated Standard Review Plan chapter are 23 ready-for adoption by the agency. Right? 24 1DR. KRESS: That would be right, yes. 25 CHAIRMAN APOSTOLAKIS: Do we need to repeat the f ANN RILEY & ASSOCIATES, LTD. 1 Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

i 418 i 1 excellent effort and all that? Maybe in passing but not in () 2 the conclusions -- in the discussion. I 3 DR. KRESS: Your suggesting word changes earlier -4 on yesterday were pretty good al.so. I think they are going S to incorporate them. 6-CHAIRMAN APOSTOLAKIS: So 7 can't think of any [ i 7-other conclusion or recommendation. We're just blessing it. 8-DR. SEALE: Yes. l 9 DR. KRESS: I pretty mucn agree with that. j 10 CHAIRMAN APOSTOLAKIS: And I-don't see intensive 11 discussion and we can say the responses have been 12 reasonabls.- 2 13 DR. KRESS: To the public -- 14 CHAIRMAN APOSTOLAKIS ' I didn't expect ycu guys to ) 15 como up with Region III. I was very impressed by that. 16 DR. KRESS: Yes, that was a good move. 17 CHAIRMAN APOSTOLAKIS: But maybe we should put 18 that in the letter? 19 [ Laughter.) 20 DR. KRESS: I was real happy to see-that Region 21 III. 22 CHAIRMAN APOSTOLAKIS: I was happy too. 23 DR. SEALE:- We may want to take some credit by way 24~ of just--noting that the way in which-this thing evolved'with 25 the1 discussions-between Staff and ourselves and so forth F ANN RILEY & ASSOCIATES, LTD. . Court Reporters. 1' 1250 1 Street, N.W., Suite 300. Washington, D.C. 20005 (202) 842-0034 ---e-re--,--tm,- y r-- e p %w - n eHey7 -1w--= >,r-[.- ww y w w w

• ----w e

---ve---e, 2-w-e-*--s+e- +-i"rw--

419 i 1 .seemed!to be.--- 2L iMAIRMAN APOSTOLAKIS: -I'can't hear you. 3 - DR. SEALE:- -I'm sorry --Ese?med to be a ~ 4: particularly efficient way or at_least effective way. It 5 may'not have been efficient with your time, and we-certainly 6 . appreciate your tolerance in putting up with us, but I think 7 in general _it was an effective way of getting from where we 8 . were to where we are, and we may want to make that comment, 9-CHAIRMAN APOSTOLAKIS: Okay, so this is going to 10 be a very short letter. 11 MR. HOLAHAN: Can I make a suggestion? 12 CHAIRMAN APOSTOLAKIS - Yes. 13 MR. HOLAHAN: This is a suggestion of what I think 14 you should do, not necessarily because I think it is good .l 15 for'me. 16 CHAIRMAN APOSTOLAKIS: Yes. 17 MR. HOLAHAN: Like a lot of good efforts, somebody 18 ought to be monitoring whether they are really achieving 19 -what was intended and I think maybe it is good for us to 20 have someone watching, because I think the subcommittee 21 ought to take some role as to seeing in practice whether all 22 .these princtsics and good ideas are really working out, and '23 you might wn:<. to take on some sort of_ role of looking at 24 even if-it is not.the pilots ---just in the normal process. 25 CHAIRMAN APOSTOLAKIS: Oh, we will. ( ANN RILEY4 & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 .(202) 842-0034

~ _ _ _ _. _ 420

1 -;

MR. HOLAHAN: -- within a year cr so,-are we f - 2: _really achieving all these good-ideas we have laid out._ 13 DR. SEALE --Volunteering is so dangerous. 4 CHAIRMAN APOSTOLAKIS: Well, this is not for the 5 letter. This is just-something to do here. 6-MR. HOLAHAN: It could be. I', coul d be. ve will -- that 7 CHAIRMAN APOSTOLAKIS: Oh, yes 8 is how we finish the letters. Do we have the benefit of aay 9 documents in this case? . DR. FONTANA: I would like to get rid of that. -- 11 CHAIRMAN APOSTOLAKIS: What? 12 DR. FONTANA: Anyway, the last sentence should be j 13 more than the boilerplate is I think what he is saying. 14 Normally we have a boilerplate sentence we want to. keep up () 15 with this. We may want to elaborate something. 16 CHAIRMAN APOSTOLAKIS: Yes, we',ll do it. Oh, yes. 17 In fact, I will go back to our first letter where we had ~ 18 several suggestions and say that maybe this -- like urging 5 the Commission to encourage the industry to come back with 29 N' major studies -- 13 21 DR. KRESS: Oh, yes. and maybe this thing 22 CHAIRMAM APOSTOLAKIS: 23 that you guys are negotiating now is one of them, right? -- e 24~ -and see whether we want to say that we'd like to monitor .25-progress on these fronts. ( --ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

421 1 By the way, what did you decide to do about 6) bibliography, or you haven't? 2 i 3 MR. CUNNINGHAM: I think we will look at having a 4 more complete set of references or -- in a bibliography or 5 something like that. 6 CHAIRMAN APOSTOLAKIS: That 's a good idea. 7 Well, how about a second letter? The second 8 letter is not going to be so easy. I'm not sure we have an 9 answer that will satisfy the Chairman, frankly. 10 DR. KRESS: On values versus uncertainty? 11 CHAIRMAN APOSTOLAKIS: 'le s. 12 DR. KRESS: Well, these guys had a pretty good 13 story. 14 CHAIRMAN APOSTOLAKIS: What story's that? \\_s/ 15 DR. KRESS: You didn't hear it? 16 CHAIRMAN APOSTOLAKIS: Huh? 17 DR. KRESS: You didn't hear it. You've got to go 18 back to the transcripts and -- 19 CHAIRMAN APOSTOLAKIS: I heard it. I heard it. 20 And I still ask myself what story is that. If I were the 21 Chairman would I find thac satisfactory? I don't know. 22 Maybe we ought to spend a little more time at the 23 presentation next time. 24 DR. KRESS: On that particular issue. 25 CHAIRMAN APOSTOLAKIS: That really bothers me. if ) ANN RILEY & ASSOCIATES, LTD. Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034

..... ~.. 422 - 1' DR. _KRESS - You_ guys may-' ant to_ repeat some--Lof -f - 2: what you-_said=at -- .3 CHAIRMAN APOSTOLAKIS !And we-had -- remember at ~ i '4 one'of the subcommittee meetings in June._we had~a list of- -- 5 1 questions and-issues. I don't know that we have addressed 6 all of-them. 7-Okay. Anything else?. .8 (No. response.]' 9-This is then the closure-of this subcommittee '10 meeting. Thank you gentlemen very much. -11' .[Whereupon, at 11:36 a.m., the meeting was 12 concluded.) -13 14 , O 1 Aj 15 16-17 18- '19 20 21 22 23 I .24: ~ 25-t ( ANN RILEY--&-ASSOCIATES, LTD. Court Reporters ~ 1250 I Street, N.W.,--Suite-300 E Washington, D.C. 20005 (202) 8421-0034 -.u.. m .2

REPORTER'S CERTIFICATE. () - This is; to, certify that the ~ attaciled proceedings before-theLUnited States Nuclear Regulatory Commission in ~ the: matter of: NAME40F PROCEEDING: ACRS SUBCOMMITTEE RELIABILITY AND PROBABILISTIC RISK ASSESSMENT 4 - DOCKET. NUMBER: 4 PLACE OF PROCEEDING: ROCKVILLE, MD t 4 were held as herein appears, and that this is the original O) - transcript thereof for the file of--the United States Nuclear (_- r Regulatory' Commission taken by me and therea'fter reduced to typewriting by me or under the-direction of the court-reporting company, and that the transcript is a true and-accurate record of the foregoing proceedings, i .%Si'YW J(o)n Hundley Official Reporter Ann Riley & Associates, Ltd. o 4 ...-......e__.._...,...-.J.-.,,-.,s- --..c..-..-.-

O O O \\ i ) United States \\..../ Nuclear Regulatory Commission l Proposed Modifications to DG-1061 and SRP Chapter 19 in Response to Public Comment and Pilot Plant Experiences Thomas King, Mark Cunningham Office of Nuclear Regulatory Research Gary Holahan, Gareth Parry, Michael Cheok l Office of Nuclear Reactor Regulation Presentation to Advisory Committee on Reactor Safeguards Subcommittee on Reliability and Probabilistic Risk Assessment November 12,1997 i

O O O Overview Summary of public comments on DG-1961 and SRP Chapter 19 Summary of planned changes Discussion of specific changes Policy and other issues ACRS issues of June 12,1997 I 2

t Summary of Public Comments on DG-1061 and SRP Chapter 19 PRA Standards 3 Acceptance Guidelines i Integrated Decision Making l Licensing Issues Licensee Burden 1 Staff Review Process Implementation Issues i 3

O O O. Summary of Public Conunents (Cont.) Acceptance Guidelines Use of the 10~'/RY benchmark on CDF L Process for treatment of very small risk increases Allowance for very small increases in risk Treatment of uncertainties t Guidelines for temporary changes Specific guidelines for shutdown operations i i Guidelines for use oflevel 3 analysis 4 4

O O O Summary of Public Comments (Cont.) 1 Integrated Decision Making 1 i Reconsider use of absolute quantitative criteria a core damage frequency large early release frequency conditional core damage probability (TS) Provide better definition of the roles of defense-in-depth and safety ~ margins a Provide better definition ofincreased management attention Provide guidance on bundling of changes Provide more guidance on use of qualitative and quantitative evaluations 5 i

b O .O O Summary of Public Comments (Cont.) Licensing Issues j Definition of CLB too broad; limit scope to regulations, orders, license i conditions, exemptions and Technical Specifications Guidance for conducting evaluations per 10CFR 50.59,(i.e., does NRC have to review all risk-informed changes to CLB?) l I I i I I l l 6 1 1 l

Summary of Public Comments (Cont.) J Licensee Burden PRA quality; guidance implies only state-of-the-art PRA is acceptable I level of detail 4 scope QA Peer review (particularly in comparison with traditional analyses) Overlap with maintenance requirements (categorizing SSCs, performance monitoring, configuration risk management) Monitoring and corrective action f more focused guidance needed; too much expected monitoring of SSCs oflow safety significance j 1 1 Excessive documentation requirements 1 l 7

---

PT

O O O Summary of Public Comments (Cont.)' Staff Review Process Inconsistency among reviewers l Common interpretations of guidance by NRR and Regional offices - Bringing complex issues to closure i Implementation Issues I Tracking of cumulative changes to risk - What is purpose? 1 additional guidance needed t Limited scope (of application) submittals e.g., only address IST requirements for selected set of pumps and valves Need for a defacto "living PSA" i 8. i i t

0 O O Summary of Planned Changes I Principles Acceptance Guidelines Bundling PRA Quality and Scope r Management Attention Performance Monitoring Documentation Appendix B - NUREG/CR report 9

O O O Specific Changes - Principles 1. The proposed change meets the current regulations. This principle applies unless the proposed change is explicitly related to a requested exemption or rule change (i.e., a 50.12 " specific exemption" or a 2.802 " petition for rulemaking"). 2. Defense-in-depth is maintained. i j 3. Sufficient safety margins are maintained. 4. Proposed increases in core damage frequency and risk are small and are consistent with the intent of the Commission's Safety Goal Policy Statement. 5. Performance-based implementation and monitoring strategies are proposed that address uncertainties in analysis models and data and provide for timely feedback and corrective action. 10 l {

O O O f Specific Changes - Acceptance Guidelines l 4 ACCEPTANCE GUIDELINES FOR CDF If the application can be shown to result in a decrease in CDF, or is CDF-neutral, the l change will be considered to have satisfied the relevant principle of risk-informed regulation with respect to CDF. When the calculated increase in CDF is very small, which is taken as being less than 1E-06/ reactor year, the change will be considered, regardless of whether there is an i assessment of the total CDF. The technical review will address the scope, quality, and i robustness of the analysis of the change, including consideration and quantification of uncertainties. When the calculated increase in CDF is in the range of IE-06 to 1E-05/ reactor year, applications will be considered only if it can be reasonably shown that the total CDF is less than IE-04/ reactor year, subject to an NRC technical and management review. The technical review will address the scope, quality, and robustness of the analysis of both the change and the baseline CDF, includmg consideration and quantification of uncertainties. Applications which result in increases to CDF above 1E-05/ reactor year would not normally be considered. ) i 11 ) l

1 O O o Specific Changes - Acceptance Guidelines ACCEPTANCE GUIDELINES FOR LERF If the application can be shown to result in a decrease in LERF, or is LERF-neutral, the change will be considered to have satisfied the relevant pr.'nciple of risk-informed regulation with respect to LERF. When the calculated increase in LERF is very small, which is taken as being less than i 1E-07/ reactor year, the change will be considered, regardless of whether there is an assessment of the total LERF. The technical review will address the scope, quality, and robustness of the analysis of the change, including consideration and quantification of uncertainties. When the calculated increase in LERF is in the range of IE-07 to 1E-06/ reactor year, applications will be considered ce if it can be reasonably shown that the total CDF is i less than 1E-05/ reactor year, sdject to an NRC technical and management review. The i technical review will address thu scope, quality, and robustness of the analysis of both the change and the baseline LERF, including consideration and quantification of uncertainties. l Applications which result in increases to LERF above IE-06/ reactor year would not j normally be considered. l l 12 u

O O O i l Specific Channes - Acceptance Guidelines Concept of Very Small Chanaes no changes Region I anowed Reenagemesd Region 11 assention full tv-vz:..1-i an=8rses track canuieuwe knpoets l u. O very small Region Ill g <1 Region i

• M.

basekne CDF unewt amesi. l only on a CDF to track cumulative i kapacts i l Region ll s' Region 111 i p' t$ 10 CDF 15

O O O Specific Changes - Acceptance Guidelines COMPARISON OF PRA RESULTS WITH ACCEPTANCE GUIDELINES l The focus of the comparison is to assess whether principle 4 has been met, namely that increases in core damage frequency and risk are small and are consistent with the intent of the Commision's Safety Goal Policy Statement". It is not sufficient to simply compare the values calculated from a PRA with the guidelines; both the-contributors and how the results could be impacted by uncertainties in the analysis should be understood, Uncertainties to be addressed: l parameter uncertainties model uncertainties completeness uncertainties Scope of uncertainty analysis required is a function of the role the quantitative results play in the decision, and on the significance of the calculated change. i 14 - - +

0 0 0 Specific Changes - Acceptance Guidelines L COMPARISON WITH ACCEPTANCE GUIDELINES (Cont'd) When comparing quantitative PRA results with the guidelines, mean values should be used. Mean values capture the uncertainty in the parameter values to some extent, are-compatible with traditional decision-making practices, and are appropriate given the origin of the values used in the acceptance guidelines. The requirement to use mean values does not imply that a detailed propagation of uncertainties is always necessary; in many cases it will be possible to show that a point j estimate is an acceptable approximation to the mean value, using qualitative arguments about the contributors to the assessment. p Unquantified uncertainties such as those arising from model uncertainties and questions of completeness must be addressed, even, and perhaps especially when, the changes in j risk metrics are in the region of the acceptance guidelinas where only the change is i requires to be evaluated. i l 15

O. O O i J Specific Chenges - Acceptance Guidelines ) i i i COMPARISON WITE ACCEPTANCE GUIDELINES (Cend) In addressing model uncertainties, the focus should be 'on those that most strongly I impact the application. For small increase in risk, and relatively minor changes, the number of issue!to address will be small. For cases for which increases lie in the l intermediate range, such that the baseline risk metrics are to be evaluated, the number of issues will be correspondingly larger. l Model uncertainties may be addressed by appropriate sensitivity studies to assess the j impact of alternate assumptions or approximations, by demonstration that the assumptions adopted in the analysis are bounding, or by qualitative arguments. Alternate assumptions or models for key issues should be reasonable in that there is some precedent for their use, and that they have a reasonable basis given the state-of-knowledge in the industry. l l 16

O O O-Specific Changes - Acceptance Guidelines COMPARISON WITH ACCEPTANCE GUIDELINES (Cont'd) When the analysis is not full scope, it is necessary to address the impact of those risk contributors (initiating events, modes of operation) not modeled. This may be done by bounding analyses, by a qualitative argument that the contribution i from the missing analyses does not impact the decision, or, if necessary, by l supplementing the analyses with detailed analyses. t One acceptable alternative is to design the change to the CLB such that the missing risk contributors are not impacted by the change, or that the assessment of the change would not require a particular modeling issue to be addressed. i k i 1 i 17 u

r O O .O Specific Changes - Bundling of Changes. i Changes;that make up a MCR will normally be related to one t another, for example by affecting i i -l the same single system or activity, the same safety function or the same accident sequence or j group of sequences, or the same type (e.g., changes in TS allowed outage time). Does not preclude unrelated changes being accepted i l l l 18

O O o i Specific Changes - Bundling of Changes 1 Relationship among mdividual changes and how it has been modeled in the risk assessment should be addressed. i Licensees should evaluate the overallimpact of the changes m a MCR against the safety principles and qualitative acceptance guidelines in Section 2.1 and the quantitative acceptance guidelines in Section 2.4.2.2 ) l Staff will-consider the acceptability of the individual changes in its revie.w of the MCR; but will focus primarily on the overall j impact of the MCR on safety at the plant. 4 t 19 j i

O O

O.

Specific Changes - PRA Quality and Scope l The PRA performed should realistically reflect the actual design, construction, and operational practices. The scope and quality required of the PRA is commensurate with the application for which it is inter.ded and on the role the PRA results play in the integrated decision process. Acceptance' guidelines require that all plant operating medes and initiating events be addressed. Not necessary to have a PRA that treats all these modes and initiating events. j i Qualitative treatment of missing modes and initiators can be sufficient in many cases. Adequacy of modeling could be assessed by a ver review of the PRA. Industry PRA certification programs and PRA cross-comparison studies could support this review \\ process. i 20 l

~ O O O a Specific Changes - Management Attention t i l Issues addressed by management will include: The cumulative impact of previous changes and the trend in CDF (the licensee's risk management spproach); The cumulative impact of previous changes and the trend in LERF (the licensee's risk management approach); o The impact of the proposed change on operational complexity, burden on the operating staff, and overall safety practices; and t Plant-specific performance and other factors, including, for example, siting factors, inspection findings, performance indicators, and operational events. i 21 1

O O O' Specific Changes - Performance Monitoring Primary goal is to ensure that no adverse safety degradation occurs because of the changes to the CLB. i principal concern - possibility that the aggregate impact of changes which affect a large class of SSCs could lead to an unacceptable increase in the number of failures j due to unanticipated deg.adation, including possible increases in common cause j mechanisms Decisions concerning implementation of changes should be made in light of the uncertainty associated with the results of the traditional and probabilistic engineering i evaluations. Broad implementation within a limited time period may be justified when uncertainty is shown to be low 1 i slower, phased approach to implementation when uncertainty in evaluation findings is higher and where programmatic changes are being inade which potentially impact ] SSCs across a wide spectrum of the plant 22 e m - - - - ~

O O O i Specific Changes - Performance Monitoring i Program should be structured such that: SSCs are monitored commensurate with their safety importance, j i.e., monitoring for SSCs categorized as low safety significant may be less rigorous than that for SSCs of high safety significance; feedback of information and corrective actions are accomplished in a l a timely manner; and degradation in SSC performance is detected and corrected before plant safety can be compromised. j i i 23

l-O o o 1 i l l Specific Changes - Performance Monitoring 1 ? Integrate or coordination of monitoring for risk-informed changes with existing programs for monitoring equipment performance and other operatmg experience on their site and throughout the industry. j F monitoring performed as part of Maintenance Rule implementation can be used in cases where SSCs affected by the application are also covered under the Maintenance Rule and if the Maintenance Rule criteria are compatible with the application of interest j t e I i Important that provisions for specific cause determination, trending of ( l I degradation and failures and corrective actions be included. i Monitoring program should identify any corrective actions to preclude j l recurrence of unacceptable failures or degraded performance below expectations. j \\ i 24 1 1 I -~

^ O O O Specific Changes - Submittal Documentation Information expected to be submitted: A description of how the proposed change will impad the CLB A description of the components and systems affected by the change, the types of changes proposed, the reason for the changes, and results and insights from an analysis of available data on equipment performance A reevaluation of tLe licensing basis accident analysis and the provisions of 10 CFR Parts 20 and 100, if appropriate An evaluation of the impact of the change in licensing bases on the breadth or depth of defense-in-depth attributes of the plant IdentificaGan of how and where the proposed change will be documented as part of the plaats licensing basis (e.g., FSAR, TS, licensing conditions). This should include proposed changes and/or enhancements to the regulatory controls for high risk-significant SSCs which an not subject to any requirements; or where the requirements are not commensurate with the SSCs risk-significance. 25 l

O O O Specific Changes - Submittal Documentation Licensee should also identify: Those key assumptions in the PRA that impact the application and commitments made to support the application SSCs for which requirements should be increased Submitted information summarizing the risk assessment methods used: A description of risk assessment methods used The key modeling assumptions necessary to support the analysis or that impact the application The event trees and fault trees as necessary to support the analysis of the CLB change l A list of operator actions modeled in the PRA that impact the application and their j error probabilities 26 l

O O O Specific Changes - Submittal Documentation Submitted information summarizing the results of the risk assessment should include: 1 The effects of the change on the dominant sequences (sequences that contribute more than 5 percent to the risk) in order to show that the CLB change does not create risk outlicrs and does not exacerbate existing risk outliers. An estimate of total plant CDF (including a qualitative or quantitative assessment of uncertainty) before and after implementing the proposed CLB change An estimate of the total plant LERF (including a qualitative or quantitative assessment of the uncertainty) before and after implementing the proposed CLB change, and a summary description of the methodology used to calculate this LERF Analyses that show that the conclusions regarding the impact of the CLB change on plant risk will not vary significantly under a different set of plausible assumptions. A description of the licensee process to ensure PRA quality and a discussion as to why i the PRA is of sufficient quality to support the current application h I 27

0 O O Specific Changes - Submittal Documentation Cumulative risk documentation should include: the calculated change in risk for t.ach application (CDF and LERF) and the plant elements (SSCs, procedures, etc) affected by each change qualitative arguments were used to justify the change (if any) and the plant elements affected by these arguments compensatory measures or other commitments used to help justify the change (if any) and the plant elements affected a summary of the results from the monitoring programs (where applicable) and a discussion on how these results have been factored into the PRA or into the current 2 application 4 28

O O O Specific Changes - Submittal Documentation f Performance Monitoring Documentation i Description and rationale for the implementation and performance monitoring strategy for the proposed CLB change. t i l i l l 29 i

i i Mean Core Damage Frequency Acceptance Guidelines 10-5 m.. .t ,L^~y,n a w.m,a ,r s 10-6 a- '- W y.w.g_ u H s'- y-4* a~'$39 ' w :,.p a pgap ~,ny ~w" 2: ji 1%nbi

,y

k

.n; ' m. s. 5!Nyn;g,ipc u. a .:i. '..yy, z.gr,.

=

..,.- eg O mW ,p; - .- rm . c.7 3 m 4 ww; m e y. o

• l e 'r " i Q' Wic s

,e r3 '* LA =, T' i, ,6 J,t '. mf ~3 .i 10-7----------,r-------------r,-------------,s---~~~~-- , m., r aw e e

• h Negligible Calculated Change di s e ;l i

i i t .I ,e n,, 10 8 10 7 10-6 10 5 10-4 10 3 Mean CDF ?_' More NRC Staff Scrutiny Less

4 I Jcnusry 22, 1997 i I g4EMORANDUMTO: Hugh L. Thompson, Jr. Acting Executive Director for Operations P I Karen D. Cyr General Counsel FROM: John C. Hoyle, Secretary /s/ SUBJECT - STAFF REQUIREMENTS - SECY-96-218 - QUARTERLY STATUS UPDATE FOR THE PROBABILISTIC RISK ASSESSMENT (PRA) IMPLEMENTATION PLAN, INCLUDING A DISCUSSION OF FOUR EMERGING POLICY ISSUES ASSOCIATED WITH RISK-INFORMED PERFORMANCE-BASED REGULATION The Role of Performance Based Regulation in the PRA Implementation Plan The Commission has cpproved Alternative 1 with respect to the role of performance-based regulation but applications of g arformance-based approaches should not be limited to risk-W informed initiatives. Thus, the Commission also approves olements of Alternative 3 as follows: Performance-based c initiatives that do not explicitly reference criteria derived from PRA insights should not be excluded from consideration. The ctaff should include in the PRA implementation plan, or in a esparate plan, how these performance-based initiatives will be phased into-the overall regulatory improvement and oversight program. As part of the PRA implementation plan, or its separate plan, the staff should include its plan to solicit input from industry on (or develop on its own) additional performance-based objectives which are not amenable to probabilistic risk analysis, but could be ranked according to, for example, a relative hazards analysis, and phase in these initiatives. (EDO) (SECY Suspense 8/29/97) 1 p Tu r ee wcy v3-wt-W- tt'- P g-- 'w? -,e--e r-

Tha stoff chould provida tha Commiccion a cummary diccusoion on how performance monitoring is being addressed in current PRA Pilot Applications and, where appropriate, other planned .erformance-based approaches. The staff should address the gtechnicalquestionconcerninghowtheimplementationand monitorAng aspects of performance based regulations (Attachment 3, Item IV) are considered in these planned performance-based cpproaches. For the maintenance rule implementation activities, cddress how these issues are considered within the context of the inspection process and inopection program. These items should be cddressed in the March 1997 quarterly update and in the next commission briefing on the PRA implementation plan. (EDO) (SECY Suspense: 3/31/97) The words " intolerable outcome" in the fourth key element are too v gue and require further definition. For example, the words could be revised to read " failure to meet a performance criterion will not result in violation of a Safety Limit" or some other cpecific terminology. (EDO) (SECY Suspense 8/29/97) Plant-Specific Application of Safety Goals The Commission has tentatively approved Alternative 1 with - rcspect to plant-specific application of safety goals and/or 7 (,;ubsidiary objectives, but prior to issuance of the final guidance, the staff should explore the legal ramifications of the uce of numerical guidelines for plant specific regulatory d cisions and prepare a legal analysis of the issues for the Commission. As part of this analysis, the staff should consider cituations where updates or changes to licensees' PRAs (such as the underlying assumptions) result in changes to PRA results, which would cause a previously approved action to become unacceptable. The analysis should also include a discussion of the type of regulatory decisions that might be subject to litigation, an identification of the problems that such litigation might pose for the staff, and an estimate of the level of staff resources and technical support that likely would be rcquired to address the issues in such litigation. (OGC) (SECY Suspense: 6/30/97) Risk Neutral vs. Increases in Risk The Commission has approved Alternative 1 which would allow for cmall increases in risk under certain conditions, for proposed ,_ changes to a plant's licensing basis. The legal analysis ( 'Jcquested above should address the legal ramifications and ' prospects for litigation in making this change. In addition, the

- - ~-_ =. _ _. _ _. _. _.. _ _. _._ _.. _ _ _ _. _ _. _ _.. _ _ _ _ _ _ 9 N tGrms 'Cmalla cnd~"undar cartcin conditiono* rcquiro moro prccico definition.; The staff should provide a sound rationale for judging small' increases.and provide.for explicit consideration of gneertainties.:. Criteria for judging.small increases in risk ' f t i J ? h -i 3 h 10 h t k a 'I t i d: i 3._. .+' d f +vqe g- ,r-,c ..v,+,, _, _ _ _,,.,..,. _,,

ch:uld be conoid: rcd in tho context of maintoining rocconable casurance that there is no undue risk to public health and cafety. The staff should establish procedures to monitor the ( ';umulative changes in risk for a given nuclear facility as the ' #rcsult of license amendments that are conducive to quantitative risk assessments.- The staff should. develop a methodology for casessing-changes in risk that uses statistical concepts and gives considerations to uncertainties. (OGC/EDO) (SECY Suspense 8/29/97) The staff should, in its development'of risk-informed guidance end review of applications regarding risk-informed initiatives cvaluate all safety impacts of proposed changes in an integrated manner including the use of risk insights to identify areas where rcquirements should be increased or improvements could/should be implemented. In this regard, the staff should encourage licensees to use risk assessments for purposes of improvement that may require additional activity or effort on their part, as wall as relaxation, in order to realize the full benefit of risk casessments. i The staff should also verify licensee activity in this regard, as cppropriate. (EDO) (SECY Suspense 8/29/97) Implementation of Changes to Risk-Informed IST and ISI Raquirements The Commission has approved Alternative 2 allowing the staff to uce the acceptable alternative provision of 10 CFR 50.55a (a) (3) (1) to approve the pilot plants' applications provided appropriate findings can be made. Where the findings n2cessary to approve the alternative cannot be made, then the use of exemptions should be considered. The staff should work closely with ASME and with the Code consensus process ex) as to expedite changes to the Code involving ISI and IST. cc i - Chairman Jackson Commissioner Rogers Commissioner Dicus Commissioner Diaz Commissioner McGaffigan OGC OCA 7,. OIG -Office Directors, Regions, ACRS, ACNW, ASLBp-(via E-Mail) ~

o o o i PERFORMANCE-BASED f REGULATIC'N 4 l l F l Joseph Murphy, Director Division of Regulatory Applications j i November 13,1997 j Presentation to Sub-Committee on Reliability and Probabilistic Risk Assessment i I i i

[ OBJECTIVE l CONSIDER PERFORMANCE-BASED APPROACHES THAT DO NOT EXPLICITLY REFERENCE CRITERIA l FROM.PRA l PLAN HOW THESE MAY BE PHASED INTO REGULATORY IMPROVEMENT AND OVERSIGHT PROGRAM l l SOLICIT INDUSTRY INPUT l 2 i i i

O O O APPROACH J [ SPECIFY SAFETY OBJECTIVE AND ACTIONS IF l OBJECTIVE IS NOT MET l l l l l LICENSEE DETERMINES HOW OBJECTIVE WILL BE MET i MARGIN REQUIRED REGULATORY GUIDES TO SUPPORT QUALITATIVE CRITERIA LICENSEE DETERMINATION INCORPORATED IN CONTROLLED DOCUMENT I i 3 i ,. ~. _ _ _ _ _ _ _ - _

O ATTRbTES O l MEASURABLE PARAMETERS i OBJECTIVE CRITERIA TO ASSESS PERFORMANCE 4 [ RISK INSIGHTS HAZARDS ANALYSIS j PERFORMANCE MONITORING I DETERMINISTIC ANALYSES i i LICENSEE FLEXIBILITY ~ CLEARLY DEFINED OBJECTIVES t i I INSPECTABLE AND ENFORCEABLE i i 4

O IMPLICklONS O .l i i. l SAFETY OBJECTIVE MAY BE QUALITATIVE [ PROVIDED OBJECTIVE CRITERIA (NOT j NECESSARILY PRA-BASED) CAN BE DEVELOPED i INSPECTION FOCUS ON OVERSIGHT OF l l PERFORMANCE MONITORING PROCESS AND. l EFFECTIVENESS OF CORRECTIVE ACTIONS l l l DEFENSE-IN-DEPTH CONSIDERATIONS MAY LEAD l l TO TRAIN-LEVEL PERFORMANCE MONITORING l l l 5 ~ d I

O IMPLEMiSTATION O i TIED TO IMPLEMENTATION OF DSI 13 1 SOLICIT INDUSTRY SUGGESTIONS FOR CANDIDATE REGULATIONS (OR REGULATORY ~ GUIDANCE) THAT MIGHT BE CONVERTED f ENCOURAGE PETITIONS [E.G. REGULATORY GUIDE 10.12 ON 10 CFR 2.802~ ~ EVALUATE NEED FOR PILOT STUDIES i REPORT TO COMMISSION BY END FY98 4 6 o .}}