ML22230A132

Tran-M791016: Public Meeting Briefing on TMI Lessons Learned Task Force Report

ML22230A132
Person / Time
Issue date:	10/16/1979
From:	NRC/OCM
To:
References
Tran-M791016
Download: ML22230A132 (93)

Text

RETURN TO SECRITARIAT RECORDS NUCLEAR REGULATORY COMMISSION IN THE MATTER 0~:

PUBLIC MEETlliG BRIEFING ON TMI LESSONS LEARNED TASK FORCE REPORT Place - Washington, D. C.

Date - Tuesday, 16 October 19 79 Pages 1 - 91 Telephone:

(202) 347-3700 ACE - FEDERAL REPORTERS, INC.

Official Reporten 444 North Capitol Street

• Washington, D.C. 20001 NATIONWIDE COVERAGE.- DAJLY

CR7703 1 DI SCLAI}IBR This is an unofficial transcript of a meeting of the United States Nuclear Regulatory Corrrrnission held on Tuesday, 16*0ctober 1979 in the Cornmissions's offices at 1717 H Street, N. W., Washington, D. C. The meeting was open to public attendance and observation. This transcript has not been reviewed, corrected, or edited, and it may contain

- inaccuracies.

The transcript is intended solely for general informational purposes. As provided by 10 CFR 9.103, it is not part of the formal or informal record of decision of the matters discussed. Expressions of opinion in this transcript do not necessarily reflect final determinations or beliefs. No pleading or other paper may be filed with the Commission in any proceeding as the result of or addressed to any statement or argument contained nerein, except as the Commission may authorize.

2 UNITED STATES OF AMERICA

-CP..7703 2 NUCLEAR REGULATORY COMMISSION 3

PUBLIC MEETING 4

BRIEFING ON TMI LESSONS LEARNED TASK FORCE REPORT 5

6 Room 1130 1717 H Street, N. W.

7 Washington, D. C.

8 Tuesday, 16 October 1979 9 The Commission met, pursuant to notice, at 9:35 a.m.

10 BE*FORE:

11 DR. JOSEPH A. HENDRIE, Chairman 12 VICTOR GILINSKY, Commissioner

- 13 14 15 RICHARDT. KENNEDY, Commissioner PETER A. BRADFORD, Commissioner JOHN F. AHEARNE, Commissioner 16 ALSO PRESENT:

17 Messrs. Mattson, Case, Gossick, Sege, and Bickwit.

18 19 20 21

- 22 23

- 24 Ace-Federal Reporters, Inc.

25

-103 O I O I 3

~DA 1/  ? ~, O C E E D I ;~ G S 2 CriAIRMAN HENDRIE: If we coulj come to order. I 3 assume our colleaJues will join us directly. The Commission

~ meets this mornin~ for another in a series of the briefings

.) on the Three Mile Isla11d lessons from our Lessons v~arned j fasK Force. I ta'.<e it this is an intn1juct::iry oriefin::;i indictetinq the direction that Ro12r anJ his forces are 3 going in the forthcoming lon~-term rscommenjations paper, J *,1hi-:h I 'Noulj remar:< is not yet v1itt). us, in case anybody~s 1-J loo'.cing =1ro Jnd to ses what they missed.

1 li >~e welcome you, Lse, Roger, Ed, please go ahead.

L \Fi. GOSS IC:<: Than1< you, :,h. Chaicnan. As you 13 inJi:ated the report is not out yet. Roger tells me that he 14 expects to ~ave it available by the end of this wee~.

1::i (At 9:36 Commissioner Ke~nedy entered the room.)

1.S \\:~. GOSSICK: This is indeed a preview of the 1, paper on final recommendations of the task force. Ed, did 13 you nave an;thing?

jj !ki. CASE: No.

2J C(),'vP.{ISSID:'JER GILINSKY: These are the lonJ-term as 21 opposed to the short-term items?

22 MR. MATfSON: I wi 11 describe the difference. Let 23 me do that. And I think that wi 11 answer your question.

COMMISSIONER GILL~SKY: But it*'s complementary to 2j what we hea:-d before, rather than :;;ioing over that :J.Cound

103 OJ J2 4

\(upJA */

)

_)

• +

\U. :i:\ATfSON:

ltH. MATTSON:

Iha t" s ri :::Jh t.

A supplement. ,fall, the purpose

_) today is to tell you the content of the fi"1al report by the j Lessons Learned Task r orce. /'Je hope to issue the re {Jort by I friday night of this wee~. ~e have a proble~ in 3 guaranteeing that, which is so~ewhat of a syste~s

';) interaction proolem. I felt that it"s really a quality 1.J assurance proble:11. But it seems L1at the dehumidifier i'.1

] I the reprodu.:tion room causes the r9;Jroductio11 -- seem to I .2 kind of stick pieces of paper together. And you can t 13 guarantee, when y~u want 100 copies of something, tnat they 1+ won't be missing several pages. I think that"s a systems

]j interaction problem. We"re trying to solve several in

]j getting this report out.

1 j I thought it would be good to review the history 1::3 of the task force ta give you an idea -

l:i CHAIRMAN HENDRIE: It-'s dryer downtown. If we 2J could consolidate down here we won" t have a problem.

21 (Laughter.)

22 C0\1~HSSIO:-IER AHEARNE: Joe, can you say that the 23 way the roof"s been leaking?

24 CHAIRMAN HENDRIE: A different problem.

- 2:5 ~R. MATTSON: The task force was an

')03 01 03 :J k2pJA' int3rdisciplinary tsam of aoout 20 ~9ople. It was formed by J

.:i 4

the Office -::>f >luclear F~8actor Regul:ition in late May so we've had roughly five months to work o~ the lessons learned from Thr-ee .Hle Island. Our purpose was to identify and

) evaluate safety concerns from Three ~ile Island that wer~

6 within NRR cognizance.

I I And as I told you in the last oriefing, there wer9

• 3 so~e things we specifically exclud9J, li~e emerJency

) preparedness, fro~ the stats's siting and tne overall NRC 1J rol~ in accidents, those topics bein~ addressed by other l i people within the staff.

1::: ~e iss ed a short-term report in July known as 13 .'~UREG-OS 18, '1fhich had a numoer of recommendations f:Jr J.:j. changes in operating plants and licensing requirements for

- 1:.)

15 near-term DL applications.

As a result of those recommendations, reviewed by the ACRS, Mr. Denton an~ you gentlemen, some 24 short-term 1'

ld licensing re~uirements are now being implemented on

]J operatinJ plants.

2) (At 9g40 Commissioner Bradford entered the room.)

21 MR. MATISON: All but a large number of those are 2.2 to be in place by January I, I 980. The re st, for -'3 ll 23 practical purposes, a year later. In the course of working 24 on those short-term recommendations and developing long-ter~

2j recommendations that I won~t be describing today, we've had

-,oJ 01 04 5 an ~pportunity fa~ ~3ny meetings with the ACRS 3nJ its f~I-2 suocommitte~. S011ething on the ord3r of 10 meetings 3 altogether, of half a day to day duration, Just to ~iV9 you

- 4

)

.)

a flavor of the size of the interaction we~ve had with the ACR3.

(At 9:41, Commissioner Bradford left the room.)

I In addition, there is one short-ter11 J recommendation coming to the Commission from the office of

-) St~ndards Development, that involves rulemaking. This was 1;) one .Jf our short-ter11 recommendations in the f'JUREG-0518. It 1i has to do with tne new limiting condition for operation for 1-~' hum-3 n errors. I think you/11 recall we discussed that 13 be foe e.

14 At this stage, all but five members of the Lessons Jj Learned Task Force have been reassigned to other Joos. We LS ceassigned them to the implementation of short-term actions I I on operating plants, where they are part of the four review Id teams meetinJ with the four owners groups and speeding the IJ l~plernentation*of th~ short-term lesssons.

20 ~e have assigned people to the review for the 21 restart of fMI-1 ahj we have assigned people to the revi9w 22 of the near-term operating license applications. Several o~

23 the task force members have returned to their parent 24 offices. You/11 recall we had four of the 22 that were from 2.::i other offices. Two have now returned. And we have set

,-103 01 JS 7

~a piJAV asiae on9 person from the L9ssons L9arn9d fcisk Force to oe I

3 the nucleus of the numan factor control roo~ efforts within the office of Nuclear Reactor Regulation ovsr the coming

- 4

.:)

6 man th s.

With publication of the final report later this week or the first of next week, the rest of the t3sk force I will return to its other duties.

3 CD,\iMISSIOtJER GILii'6K'f: Cs,uld I just interrupt Y*'.)U

-) for a moment? Mr. Chairman. did you intend to schedule a IJ meeting aft9r we've had a chance to look at that r9port?

1I ,\FL MATTSON: I t:1.ink I can anticipate your 12 conc9rn. What we have said to the people that are being l .:)., reassigned and to their managers is that th~y have to oe l '+ available to call together again over the course of the next

- Jj

]j 1/

few months, for t',10 basic reasons.

implementation of that recommendations.

One is to advise on the Sometimes itJs hard for ~eople to read your woras and simply walk away with 1a that.

lJ I~e oth~r i~ to participate in presentations of 20 thei;- recommendations to whomever, the ACRS, the Commission, 21 othar interested people.

2~ COMMISSIONER AHEARNE: ThatJs half the -:1nswer.

23 Your answer is that it'll bs availaole. The other part of 24 the question is, are we going to h8ve a meeting?

- 25 CHAIRMAN HENDRIE: Clearly, when we get the report

'703 01 06 g ka p0A and the comnissioners have had a chance to reflect oath ~n the material at this oriafi~g and the discussions h3re and 3 the written re,port. I thin'.< the CorTI'.nission will want to 1 meet again with you and have further discussion of those

~ items when we've had a chance to think 3bout them a littl2.

MK. :V\A ITSON: That was the next subject I was going to turn to, which is what do we envision being done 8 with the report, ~ow that it's practically written?

') C~AIRMAN HENDRIE: If the machinery worked right, lJ why, you'd oe able to print it*Frijay night.

1i \(;?. MAfTSON: Yes. 1~hat *,* e would .:io is formally li transmit it to Mr. Denton, the director of ~RR, with copies 13 to you, to the ACRS, to the public, to the ?resident's 14 commission, to your special inquiry, an~ others that we know 15 to be interested in this suoject matter.

16 I would anticipate that Mr. Denton, much like in I/ the :ase of the short-term report, woulj refer the final 18 report to the ACRS for its review and comment. We've 1~ alre~dy talkej to the ACRS about having some time in the 2'.J t'fov3:nber me~ting, both with the full committee and the 2l subcommittee, to talk about the final r3port.

22 In 9arall3l with that, we would anticipat3 that 23 there is a need for an action plan. As I"ll describe later, 24 ther'3 's a lot of work portended in our recommendations and 2J thera's a lot of work already going on with short-term

't03 01 07 ka.pJA'/ recommenjatians and ather changes that have oeen mad9 since Three ~ile Island. ~e know that th9re is a Lessons Learned 3 I as k For c e i n the Off i c a o f I n s p e c t i *::rn ,3.. En f o r c e men t a 1 so 4 aoout to issue a report and there is a raal need to sit

~ down, develoo an action plan, consider the relative priority of t~ese various racommendations ana assign resources to th8m and Jet on with the onas that are most important.

'3 1

\* / s t h i ri 'c th at th i s rev i 9 ',v -:J y the AC RS --

J C0 1,\i'll!SSIO:H:R AHEMWE: One minute. There is also 1~ the ~ossioility that some of them ~ay be mo~ified.

1i MR. MATI'S()N: Of :ourse, yes, an.:l that.,s c;ioin:] to li say so right in the report, in fact, as 1~11 ex~lain later.

13 fhe thrust of the report is more of q policy nature than of 14 a detailed nature. And our recommendations are more in the 15 for~ of, This is the way we would start in a given area in 16 order to achieve policy objectives that we have tried to 1/ arti=ulate. It's quite possible that others will come u~

13 with better ways or permutations and combinations of our 1J reCD'THner::idations that would o,:! more to the point.

2J I shoulj also mention that in the month or so that 21 it should reasonably take to develop an action plan we 22 should have the Kemeny Commission proposals on the table.

23 And I think that.,s ::;iood, that we can have all of the 24 recommendations that have been placed at one time and 2~ consider them together.

703 0 I G3 10 I'~ sur3 there are ideas that we ~ad that they

~iq~t not n3ve had and vice versa. rt s auite possibl9 3 there are ~are ways than on9 to skin some af these cats.

I/d like to turn to an overview of the final

) report. As I said, in contrast to the short-term r9port, it

~ deals with more fundamental and longer term issues. You'll recall when I was here to dascribe the short-term report we j *:,3lled them our "hot coals,il our "burning issues," the J spe:ific things and desires in operations that we thougnt IJ needed to be ~otten on with in a relatively short time 11 f ra:-;ie.

12 fh9 final report addresses three 3reas: nuclear IJ power plant operations, design and regulation. The 14 structure of the re~ort is to give prominence to the policy 15 issues in these areas that we judge to be important and in 15 need of change or improvement. Having reported our thoughts I, and conclusions in these policy issues, then we turn to IS specific recommendations to begin motion in the direction of IJ achievin;J ti10se oo.jectiv.es.

2J The costs of our recommendations are large, at 21 least in terms of the people required to execute them.

CCP1!MI SS I ONER KENNEDY: i~hat does that mean?

23 WL MATTSON: I made an estimate myself last 24 night. I must caution you that it's not rigorous, cut I 25 took 70 plants and the numb9r of people I thought it would

103 01 '.)9 I1

~< 2.p OA'

- ~

j

-,.*1 Jidn't go tnrough them all in excrutiating ~9tail.

say on the order of 2000 people and two years could do the recommendations that we hav3 in the final re~ort.

I waul~

_) COW,1ISSI01\JER KENNEDY: That is 20JO people in

') addition tG existin;i staff?

I ':l.i. MATTSONg No, I don-"'t think that"'s ne*:essarily

.:i

• _.) tru3. 200J people in the industry. some of whom will be

) peo~le already enJaJed in other activities in the industry IJ whi~h, if phasea properly, will be comolete before the II lonJer ter:n thinJs are initiated.

1 :::' COMMISSIONER AHEAR:\JE: Are you really s-:1ying that 13 ;ou estimate roughly 4000 man-years?

I+ MR. \1ATTSON: I roughly estimate 2000 man-years,

- ] .)

]j 1I 2000 year over a two-year period.

COMMISSIONER KENN!::OY:

W~. MATTSO!'J:

Okay.

Again, that" s a rou~h number in 13 terms of N~C resources I) CO/fJISSIONER GILINSKY: fo do what?

2'.) !M-?e MATTSON: Implement the recom:nendations 21 contained* in the report. ~3tter to let me describe them one 22 at a ti me.

23 COMMISSim.JER GILINSKY: Oh, you--'re going to? All 24 right.

- 2:S CHAIRMAN HENDRIE: And the NRC resource?

103 01 10 12 ke.p:JAv' le 3 CO *f'A ISSI ON ER :<ENNEJY: fh3t" s in addition to 4 exist in:) staff?

J ~,~R. MATfSON: Som3 of it could be accomplished by j existing st3ff, priority adjustments. Again, there's a need I for an action plan to consi~er what we"re putting on 3 unresolved safety issues. fhe hundred people that we were

) given by the ConJress to address licensing cases that are IJ pending, those thinJs need to be sorted out. Some of the 11 110 would be new. But not all of them. I*'.J say th3 I~ majority of them would be new.

13 Some of it involves expertise that~s not presently 14 on the staff. If I could have the first slide.

1.5 C3lide.)

15 We've oeen saying what this slide says, and others I/ have been saying it for some months now. The fundamental 13 lesson from Three ~ile Island for nuclear s3fety regulation

1) li*es i*n the area.of- ..op,:n:*ations safety. I want to d9fin9 20 *Hhat I mean by . 11 operations safety,.11 out first let me put a 21 qualifier on the statement on this slide. ~irst, we hav3 22 concluded in our short-term report the continued belief that 23 the accidents ste:n from many sources, human e_rror, design 24 error, equipment malfunction and regulatory error. By 20 giving paramount attention to o~erations safety and the

703 01 11 13 i1UITT3:'1 element that it c:rnt2ins, we are riot intending, by an1 means, ta pl1ca sole blame on the o~erator 3t Three 1/2ile 3 Isl.:md. It/s important to recognize that qualifier as we tal~ about operations safety for the future.

~ The essence of our conclusion is that thsre are J no such separate things as operations safety on the one hand and ~esign safety on the other hand. There has been a

,3 'nistake in the past in nuclear react()r regul3tion. It,.s J been the overwhelming emphasis placed on design safety and 1J t he l a c '.c of ~e mph :::i s i s o n ope r a t i o n s s a f et y

• A nd the n the 11 natural consequence of ignorance of the juxtaposition of the ld two, the Joining of operations safety and design safety in 13 assuring overall plant safety.

14 So when we say place paramount attention on 15 operations safety we mean for the coming months and few 15 yea.rs in orJer to bring it up in pace with design safety I/ e f f*'.J r ts that have been ongoing for 20 years. That-' s not to 19 say that there isn't a need for design safety improvements lJ also. Could I have the next slide, slide two?

2J CS l i de. )

21 COMMISSIONER GILINSKY: When you talk of 22 operations safety, you mean basically the running of the 23 plant, the safe running of the plant?

24 MR. MATTSON: The safe runnin9 of the plant, and 25 the consideration in design for the eventual safe running of

703 J 1 \ 2 14 kapJAI/

2. ~ayoe it hslps if we look at the Jefinition on 3 this seconj slide, of what we mean oy operations safety.

+ It"s a oro.gd .grea, including human f3ct".)rs, personnel

~ qualifications and training and the ".)ther things listed J there. And we mean operations safety to include both normal and 3mer9ency op8rations.

3 In the report we describe or depict a matrix of

-.) the elements in operations safety. These four bullets at

1) the top of t~e slide are th~ principal elements. Of course, I\ many suodivisions in these 3lements -- and they"re 12 di s::u ssed 13 COMMISSIONER GILL\J'SKY: Ooes the last one include 14 gathering and analyzing safety information?

lj *'AR. MATISON: Yes, it does. But of cours-e, ths 15 second one does also. Training must have the lessons from 1I day-to-day evaluation from operating experience factored 13 baCK into the training. And then we try to articulate two

1) goals of ops~ational safety.

2J I said at the outset the body of this report is 21 policy or goal-oriented, and so in each of the three areas 22 we try to say what ought we to be striving for. We don't 23 claim that these are the final articulations of these 24 goals. But we think goal orientation over the next few

- 25 months is very important for all of us.

i03 01 13 15 ka-;JOAV The two go3ls we would suggest at this stage are

- )

3 reducing the challenges to safety and mitigating the challenges to safety; that is, the goals for operations

- ..j.

'.)

j safety. And we mean predominantly those challenges caus9d by human e.rror. And in mitigation we mean maximizing the capaoility of the human being to intercede productively in 1 transient u~set or accident event,. ~epending upon --

'.3 COMMISSIONER KENNEDY: At the same time, lessening

} the likelihood of improper interve0tion?

lJ )-AR. MATTSON: Yes. W9 think that reducing 11 challenges 9ntails a number of things. First and probably 12 most important, recognition of personal responsibility and 13 accountability at all levels of an operations organization.

14 You'll see in our specific recommendations we feel there is

- ]j 15 1I a need, not to just concentrate on the operator and senior reactor operator in training and qualifications, but *to conc8ntrate lower in the operations staff, instrument i.

13 technicians, auxiliary operators, higher in the operations

1) .s t a ff , pl.a n.t s u p.e .r .i n t e nd e n ts , v i c e ;..> r es i de n ts for o iJ e rat i on

• 20 And as we quote Admiral Rickover in the report, 21 one of the keys to his successful naval nuclear proposal 22 program has been a continual reinforcement of the need to 23 recognize this responsibility at all levels. Of course, 24 part of recognizing the res~onsibility is also understanding

- 2:5 attitudes and mind-sets that may hve crept into the

t03 0 I 1 4 16 kapOA com~srci3l nuclear power_program ov3r 1...ne ye'3rs.

.1-*

a little bit in the report about that mind-set and the n9ed 3 for commitment thfoughout nuclear technology to see that 4 people oelieve accidents can happen, that p':lople u7::Jerstand

~ the level they need to understand, the fundamentals of 5 nuclear technoloJy and the hazards of nuclear technology.

COWifISSIO;\jER GILPISKY: How tac do you carry -

• 3 woulJ you elaborate ,"Jn these goals, because these are th~

-J sorts of things that one could have written out a week after lJ the accident. And I presume you~ve gone far beyond that and 11 indicated how a new approach would differ from the current I~ or previous approach.

13 Can you give me some sense for that?

14 MR. MATTS()N: We go beyorid it in two ways. On:3 is 1:S a way that I hope to describe by quickly going through the I .:j policy matters. I'm just g~tting started.

I/ S=:3cond, by our specific recommendations, to give 13 more concrete examples of what it is we~re talking about

1) that .shoul j oe changed. And we hope through those two weys 2;) and through some preliminary articulation of the goals of 2i these areas to move the agency into a position of better 22 articulating for itself and for the industry it regulates 23 and for the people who depend upon us, what it is weJre 24 trying to achieve.

- 2:S So part of the answer to your question isn't up

703 01 15 I 7 kapOAV ther~, Commissioner Gilinsk;, and part of t~~ re3son is that we'r9 looking for opportunities with you anj with Congress 3 and others to consider what these goAls ought to be.

4 I'~ going to come back to the question of an

~ over3ll risk ~oal for reactor safety near the end of this 6 presentation, and I think it will give more the flavor of what I'm tal 1<ing ao:)ut when I do that.

3

-J 1J 11

\ 13 y 14

- lj 15 I I 13 28 2i 2.2 23 24

- 2:5

703-.02. 1 18 \.;.,_i_.',:.:~:. . . .

DH gsh *1 Now, quickly~ reducing challenges entails 2 recognition of responsibility, better training of personnel, 3 better control of normal operations, better control of 4 information, better training about the significance of 5 precursors, and better evaluation of operating experience to 6 identify precursors; cognizance of industry-wide operating 7 experience, and better founding in the technology and 8 the hazards that we discussed.

9 These are all things we think belong not only in 10 Washington in the lines of activities in the Nuclear Regulatory 11 . Commission, but also in the individual plants being improved 12 by the individual licensees.

13 Mitigating 6hallenges requires better training and 14 qualification of operators, better operational aids in the 15 control room, better technical and management support for 16 upset or emergency conditions, better emergency procedures, 17 better preparation and communications on the part of NRC 18 for its role in accident situations and, as 1*'11 discuss 19 later, some additional engineered safety features.

20 One last slide, number 3 -

21 (Slide.)

22 -- on operations safety policy, th~- question of the 23 NRC versus the licensee-'s role:

24 We believe that operations safety is uniquely 25 nuclear utility-'s responsibility. *'fhis isn-'t to say that

703.02.2 19 pH gsh NRC shouldn't set minimum criteria and shouldn't see that 2 incentives are there for people to do the right thing.

3 But if th~ hands-on operators, and operations is 4 a hands-on concept, do not practice the spirit of the 5 regulations, then no matter how they meet them in detail, 6 the regulations cannot have anticipated all of the operations 7 as facts that could cause difficulty or safety problems.

8 COMMISSIONER .GILINSKY: But your last bullet on 9 operations safety is technical and management support of 10 operations. And under that you would include the gathering 1*1 and analyzing and dissemination of safety data~

12 That's an activity in which NRC has a pretty 13 prominent role.

14 DR. MAITSONi: *It-'s clearly an activity in which NRC 15 has a responsibility. And you have directed the formation of 16 a new office reporting to the executive director.and supporting 17 offices or supporting organizations within the program 18 offices.

19 That~s true.

20 But there are other people who can and should and 21 are beginning to evaluate operating experience in detail and 22 industry-wide~

23 Those people include the Institute for Nuclear 24 Power Operations and pursuant to one of our short-term 25 rec-0mmendations, each and every nuclear power plant licensee

1 r03.02. 3 DH gsh is required to have a dedicated, multi-disciplinary team 20

- 2 3

4 with a site presence evaluating operating experience, engineers evaluating operating experience from their plant and plants of like design, factoring that back into their 5 training program and their engineering program.

6 So as you'll see in one of our recommendations, 7 there's a need to couple all of these things. We recommend 8 some sort of nation-wide network of operating experience 9 evaluation with NRC in the lead role pulling together these 10 various elements, so that they're learning with and from 11 one another.

12 So we don't -- this slide is a dangerous slide~

13 I realize that. It triggers various responses. Let me say t4 again what we mean by this slide.

15 We mean that operations safety is a hands-on 16 concept and no matter who*'s standing over your shoulder and 17 what details have been reviewed before, if the hands-on 18 person at each piece of equipment, at each level of the

,,J9 t:.1ti 1 f.ty is --not* dedic.ated .to operations safety, then there 20 won't be good operations safety.

21 It 1 s a fundamental point in safe operations~ in 22 our Judgment.*

23 COMMISSIONER AHEARNE: I certainly dorr't disagree 24 with that, Roger. But I think the word .1*uniquely-11 is what 25 certainly is bothering me. As you pointed**6ut, the Admiral

~703.02.4 21 DH gsh has several times stressed that you must have this pervasive 2 attitude on safety.

3 But as I recall, part of that stress-goes throughout 4 the whole system. And I don-' t disagree at all with-your 5 focus that the operators, the operational staff,* the plant 6 management, must have that.

7 But I'm a little worried that if we stress that so 8 highly, as some places have stressed the great reliance upo-n 9 nuclear safety, that there may become a backing-off on the 10 part of our people and the gradual feeling that, well,*

11 that's the utility's re spans i bi 1 i ty or the industryJ s 12 responsibility.

13 It isn't. It's across the board. We have to* feel 14 that we have a fundamental responsibility there.

15 So the word "unique ly-11 16 COMMISS TONER GILINSKY l Ne would be auditing their*

17 performance.

18 DR. MATISON: Yes, and also setting criteria and 19 conducting r~views of programs to meet those criteria before 20 they go into effect

• 21 . Ne*'re not in disagreement. It is the word "urJique,*11 22 I guess, that upsets the situation here:

23 I think yott'll see in the specific recommendations 24 that 'we do not recommend standing back and letting INPO do

• 25 its thing and the NRC endorsing it;

r 7703. 02. 5 22 1

DH gsh That*' s not the tac that we have chosen. Furthermore, 2 we do not recommend standing pat with the LCO recommendations 3 in the short-term report.

4 As we said, we need to increase the incentive for 5 them to improve operations safety. We're also saying in the 6 final report we need to go further and into more of the 7 details of how operational safety is guaranteed and all of 8 the elements that fJve described. And weJve got some ways 9 that we think are good ways to start to do that.

10 COMMISSIONER GILINSKY: What-' s LCO?

11 DR. MATISON: Limiting Condition of Operation. I'm 12 sorry. When you have a condition that leads to a complete 13 loss of safety function.

14 Dwelling on that ~oint just a moment, there is a I5 need to decide how NRC and INPO, the Institution for Nuclear 16 Power Operation, should relate one to another in the 17 coming years.

18 I donJt think that I*'d wait too long to have those

-*19 discuss ions. The i.nsti tute* -has ,form.ad a board of directors 20 and is seeking to place the director of the institute and 21 hire staff with a goal toward starting activity January t 22 of the coming year.

23 I think early in the coming year would be the time 24 to sit down as a policy matter between the commission and 25 the board of directors, perhaps, of the institute and think

~703.02.6 23 DH gsh through some of these role questions~

2 \"Te 11, the fourth slide ---

3 (Slide.)

4 And if your slides are numbered,. .. you-'11 hav-e*to 5 skip around here a 1 i ttle bit. This is- the one that" s no. 4.

• 6 The slide that isn-"t numbered, it-"s-the fifth -one in the 7 package 8 (Slide.)

9 if I didn-"t -say so already.' It deals with 10 improvements in reactor regulation. We come down to this 11 question of the ne.ed for an improved safety policy basis, a 12 reactor safety goal.

13 What we-'ve done the last couple of months in *our 14 final effort is stood back and asked ourselves the question 15 where we-'ve been, how we*"re going, and how we-'re going to 16 get there.

17 We pretty much list the problems raised by Three 18 Mile Island.

T9 (At 10:05, Commission Bradford enters the room.)

20 Our list is not dissimilar to other people-'s 21 lists. But the difficulty we see if one of reaching 22 decisions in the short-term report~ We came to you -- not 23 Just we came to you before Three Mile Island with individual 24 piecemeal changes in the existing buyer regulation~

25 In the report, we likened it to .a clockwork where

f703.02.7 24 DH gsh we had a piece at a time.

2 What we see is that reactor safety policy. has 3 evolved over the 20 years of regulations, intense years of 4 regulation, evolved generally in the context of the no-undue 5 risk to public health and safety, the statutory language, 6 but without either a detailed subject.ive statement of what 7 our reactor safety goals are, and certainly without a 8 numerical statement, a risk number, in our safety goal; 9 The ACRS has writ.ten a letter, _or a short letter, 10 saying, in their judgment, it 1 s time now to write and 11 promulgate a numerical risk goal.

12 COMMISSIONER AHEARNE: Yes. As I-'m sure you know, 13 they said that it would take at least a year to get there.

14 DR. MATISON: At least a year~* We see some 15 shortcomings in articulating a numerical risk goal: There 16 are places where you can do it and there are places where 17 you can't.

18 We~ for example, don't know how to write a numerical 1.9 * .risk .goal .for .shift technical advisors; yet, we know 20 improving the technical .e:apabi 1 i ty for dynamic response 21 diagnosis in the control room is a very important thing.

22 So what we recommend is the need to articulate a 23 clear safety goal in subjective terms augmented by quantitative 24 criteriai where appropriate.

25 And there are some places where it is appropriate;

25 DH gsh especially in equipment and systems analysis.

- 2 3

4 COMMISSIONER AHEARNE1 Could you give me a feeling, an example, or .some examples of .what you mean by .t*.c !ear and subjective terms*u?

5 DR. MATISON: Accept the premise that what we have 6 today is reasonable* assurance of no one to risk the public 7 health and safety. We have more than that, depending upon 8 the individuals you talked to. But as a clearly and 9 comprehensively articulated commission pol icy, I do~ t believe 10 that we have much more than that.

11 Now you can go beyond that and I think the 12 commission has. And we say so in the report since Three 13 Mile Island.

14 As we read commissi-on statements to the Congress 15 and commission actlon on short-term recommendations; there 16 apparently is evolving a policy of reasonable assurance of 17 no more Three Mile Islands.

18 That would be another form of subjective criterion.

1'9 COMMISSIONER AHEARNE: It*' s not obvious that that 20 would necessarily be clear.

21 DR.- .MATTSON: But it would be, in our judgment, 22 better than n.o undue risk to public health and safety. It.,s 23 a concrete example of what -constitutes no undue risk, if 24 that.,s the _right level.

25 COMMISS I.ONER GILINSKY: Where do you get this undue

7703.02.9 26 DH gsh risk. The law says adequate protection.

2 DR. MATISON: Adequate protection.* I*'m. sorry. The 3 words in the staff have evolved to "no undue risk."

4 MR. BICKWIT: It~s in the regulations various places, 5 not in the statute.

6 DR. MATISONz Thank you.

7 COMMISSIONER KENNEDY: Is there any reason to 8 doubt that they've been equated?

9 DR. MATISON: No doubt.

10 MR. BICKWIT: No doubt~

11 CHAIRMAN HENDRIE: The statute talks about no 12 unreasonable risk.

13 DR. MATISON1 There are some suggestions in the 14 report about ways that you could better articulate it. You 15 could come to concepts like 11 safe and reasonably achievable. 11 16 You could come to a concept ot nas safe as best practical 17 technology , 11 or 11 best available technology , 11 the kind of 18 language, the concept chosen in the Clean *Air Act and the 19 Clean Water Act.

20 You could come to a numerical statement *-- -you*

21 could say less than some numerical quantity of risk per 22 reactor year, per core meltdown.

23 ~*le don't try to tell you the preferred way of 24 setting that goal.

25 COMMISSIONER AHEARNE: You're saying that we ought to.

I' 7

703.02. 10 27 DH gsh DR. MATTSON: We see too many places where our

- 2 3

4 ability to make decisions on.new safety issues has been eroded by interpretation and change in the safety goal through a patchwork of regulatory criteria evolved through 5 the years, and itJs difficult for us to make decislons and 6 conclude on solutions to safety problems in a timely way, 7 especially backfitting.

8 COMMISSIONER AHEARNE: Let me see if I understand 9 that.

10 Is what youJre saying that. because of the 1 ack of 11 clarity or lack of these kinds of goals that when you find 12 instances where you believe, the staff believes that it is I

13 advisable to make these improvements; you donJt have tha 14 regulatory framework to use as a justification.

15 DR~ MATISON: Yes, thatJs true~\ Ba.ckfit policy is 16 one of the areas in which we need this common guidance to 17 bind together the regulatory requirements.

18 There are other reasons. In evaluation of operating 19 0 exper-ienee *you're .:g0i-rrg -to* -f;-lnd *many ne.w safety concerns.

20 The degree to which you fix them and the timeliness of your 21 .abi 1 i ty to dee ide upon the degree to which to fix those 22 operating pro bl ems require a we 11 art ic.ulated goal.* Your 23 unresolved safety issues, there are 19 of them being worked 24 very hard by a dedt.cated group of people in the .Office of 25 Nuclear Reactor Regulation', again, the difficulty in deciding

~703.02. 11 28 DH gsh to what degree to resolve them using what kinds of decision

- 2 3

methods, ATWS being a good example.

We 1 ve tried numerically. We've tried engineering

- 4 5

6 judgment. We've tried combinations of the two.

ATWS decision comes slowly.

Yet, the COMMISSIONER GILINSKY: Extraordinarily slowly~

7 DR. MATISON: Yes. We point out the value impact 8 shows you how to weigh the gains and the losses and various 9 alternatives, but it still doesn't help you ~hoose from 10 among alternatives.

l1 The most cost effective solution is not necessarily 12 the one required to meet your basic safety goals.

l3 Now leaving that and turning to the third bullet 14 here, we .list a number of things in the f Ina! report as l5 considerations and what I1 ve called here the reform of the 16 licensing process to improve its quality.

l7

• We've looked at how we1 ve r~viewed Three Mile 18 Island-~ . We've looked at how we*'ve reviewed other cases. We've 1-9 -,seen -th-ings that -we *think need to be improved. We need a 20 . better integrated system reYiew, including human factors, in 21 that integrated system review.

22 And we need to stress the system .level of reYiew~

23 not the component level of review, as we-'ve sometimes gotten embroiled in the pas.t.

24 25 . COMMISSIONER AHEARNE: Is that a shift or an addition?

703.02.12 29 DH gsh DR. MATISON: I think it*'s a shift; I think we have

- 2 3

to come to an era where we audit, to some small extent, how t,he d~t.ails of the design have been executed. And i.f we

- 4 5

6 find that they haven't been done well, we have to come to the discipline to be able to turn a licensing process off.

We cannot continue to get into the verification and 7 validation of the details of the design ourselves. We 8 cannot act as the software or design verifier and validator 9 for a utility at the OL stage. And weJve seen that happen in 10 the past. I can give you examples of where itJs happened.

11 And in order to get back away from that nitty-gritty detail 12 that is someone else*'s responsibility and get back up to the 13 system level of review is going to take some discipline on the part of the staff and some intentional organizational 14 15 creations, in our Judgment.

16 We*'ve -

17 COMMISSIONER .GILINSKY.1 Could you give us some 18 ipecific example that would take this discretion beyond the

,1,9 ,J;eve *,of* ,gene:r-al i t i es.

20 After a 11, we*'ve al ways believed we have to look at 21 the system.

22 DR. MAITSONa I don't believe, .Commissioner Gilinsky, 23 that we-'ve always believed that in the correct way; 24 COMM ISSI.ONER -GILINSKYa We 11, then; what* I'm asking

- 25 is what is the difference between the correct way and the

/03.02.13 30 OH ,;ish incorrect ~~y? I'm trying to g9t s~~e fe9l for what you're propos in9.

3 DR. MATTSON: Okay. We have a standard review ;Jlan.

~ It's written so that each individuJl branch in the jivision

.J of systems safety, for exami:-,le, can assign various sections of 5 the standard review plan to individual reviewers and tell them to go review a narrow portion of the design.

3 ~e've got, in addition in the standard review plan, J the specification of secondary reviaw responsibilities.

lJ These are the things which are intended to see that what one 11 ~9rso n from one branch needs of another person in another 12 branch is provided.

13 Down through the years there~s been di~ficulty in 14 accomplishing these secondary review responsibilities; that lj is, the coordination among the branches in accomplishing 15 integrating of the review has not been good.

1/ The manifestations of that are overspecification of 13 margin in some areas, underspecification in others, I) -..combinations of .m.argin specifications that, 'Hhen the whole 20 system comes together, are either inordinately large or 21 unsutisfactory, in some cas,as.

22 A way to do that differently is to use 23 intgrdisciplinary project r~view teams with strong project 2~ leadership drawn from some central technical review

23. organization, for example, but with the capability in the

/03.02.14 31 DH gsh proj~ct team itself to synthesize and integr,3te the

- 2 3

contributions of the individual technical experts to the overall wor~ product.

- -r a

'_J 5

In the past, that's been ~one by a relatively few isolated individuals within the division of project man,3gement and always at the tail end of the licensing process when the pressure is on to get to the ACRS to get to 8 the hearing boards. The staff is taking too much time to do

,/

(' ..; its *technical review and the discipline and thoroughness r 10 with which this system integration occurs is not good enough.

11 CCJ,\PHSSIO:'~ER GILINSKY: Are you suggesting that we 12 move to a t9am approach the way Saa~ builds cars rather than 13 have an assembly line where each worker or reviewer does his bit -- assemble a team that will handle an application when 14 I :5 that was completed, would move on to another application.

15 COMMISSIONER AHEARNE: But it sounds like there's more 1j of an emphasis on looking at the pieces as to how they are 13 put together in the system as a whole as op~osed to each 1; individual part.

2.) Tne impression that I'm g~tting is that you might 21 be taking for granted that a number of parts. weren't going to 22 work as designed and that, instead, you're going to look at 23 how they worked together.

24 2j

703 03 01 32

ngcDAV DR. MATTSON: Yes, given that you continue to have 2 assurance tha the parts are done correctly because our 3 experience teaches us that it hasn't ~een done correctly.

4 So there need to be devices created, and they exist. The 5 Defense Department has used verification and validation 6 concepts with its contractors, third party checking of the 7 details according to established criteria and procedures, 8 and it evidently worked. And there's some history of their Y development. There" s potentially there an e.xpertise that is 10 tapable.

11 C0MhlISSI0NER GILIHSKY: l*1ell, are you saying that 12 we woula delve less deeply into the various portions of the 13 design that we look at now, but somehow emphasize the 14 fitting together of the pieces?

15 DR. MATTSON: Yes, that"s a fair statement of what 16 we're trying to say. And as long as we understand that 17 11 less deeply" doesn"t necessarily mean that we're 9oing to 18 do less work. It's probably more work to do the system l'J level integration review of saf-e.ty. It just means that 20 we"re not going to get down into the wire by wire, nut by 21 nut, component by component level of review.

22 COMMISSIONER GILINSKY: But you are saying less 23 detail work.

24 DR. MATISON: That's true.

25 C0MMISSIONE!-t AHEARNE: It.J's a different type of

703 03 02 33 mgcJAv detail.

- L 3

DR. MATTSON: Yes.

co:,tv!ISSIONER GILINSKY: You"re talking about less

- 4 5

6 wire by wire review.

DR. MATTSOr~*: Instead of looking just 1:vire by wire in safety systems, for example -- forget the wire by wire.

7 Tell somebody else how to do that so you have assurance that b it"s been done. Then you expand your scope from safety i.j systems to non-safety systems as we have traditionally 10 defined them and look at the interaction of the two, and you 11 look at the reach and scope of equipment important to 12 safety. And you look at the ways that you can get into 13 difficulty through combinations and permutations of failures 14 and operator errors to get into difficulty in safety.

15 And you come at a broader and we think better 16 overall picture of the safety of the plant and especially 17 its integration with the human being.

I c:3 COMMISSIONER GILINSKY: I"m mentioning wire by lY wire obviously. That"s what our electrical branches do.

20 They in fact do look at the actual wiring diagrams, as I 21 understand.

22 DR. MATTSON~ That"s true.

23 COMMISSIONER GILINSKY: Are you saying they"ve 24 been looking in too much detail at those diagrams? In fact,

- 25 they~re the only ones who look at the actual working

703 03 03 34 mgcuA-1 drawings, as far as I know.

- 2 3

4 DR. MATTSOi~:

enough in anocher res~ect.

Too much in two respects and not Too much in the respect of having to go through a regulatory failure modes and defects s analysis to show that the single failure criterion has been 6 met in the design. Now some of that is necessary to audit, 7 buc when you find one, instead of taking the design, turning b it back to the utility and saying, 11 1 found a single failure

':I in this design; the criteria say there aren t to be any. Go 10 review the whole design and tell me its right again. 11 11 Inscead, the staff has the tendency, having found 12 one, to go find the rest of them, do a single failure 13 analysis of che complete safety system itself, then hand the 14 whole package back to the utility and say, *11 Fix all of these 15 things, and I' 11 approve the de sign", at the expense of 16 looking at the safety systems where, if instead of doing the 17 detailed failure modes and defects analyses for single lb failures to the safety system alone, chey had integrated I ':.i mor-e of .the .plant, the non-safety systems, the systems that 20 lie between totally non-safety and totally safety, and 21 seeing the interaction,*seeing the overall contribution to 22 the safe operations of the plant of other systems in the 23 plant not normally looked at.

24 It's a much better expenditure of their time.

- 25 COMMISSIONER GILINSKY: Are you saying that the

03 03 04 35 mgcuAV prasent staff effort coula ba allocated in such a way as to

- 2 j

produce a greater benefit, in terms of safety?

UR. MATTSOU: '{es.

- 4 5

6 COM/,\ISSIONER GILINSKY:

that.,s a reasonable level of effort?

Are you also saying that I mean did you in your analysis feel bound by that level of effort and feel that

-i you were comr:1i tted to moving it around and shif tinq it from b detuils to systems?

y DR. 1ViAITSON: First of all, we don.,t recommend a 10 revolutionary overnight change in the staff organization and

.11 process to accomplish this. ',le say with t.he couple examples 12 in the last few months where approaches like this have been 13 taken, this task force being one, with benefits arising from 14 this type of approach, we suggest. some pilot projects with

- 15 16 17 perhaps operating plants in the SEP program or operating license reviews that are either new reviews or midstream reviews with the intentional formation of dedicat.ed review lo tear;is to several of those reviews, to try it and see how it 1)I 20 COMMISSIONER AHEARNE: That sounds more like your 21 review of it. is being a marginal improvement as opposed to a 22 necessary 23 COMlliISSIONER GILINSKY: To what?

24 C0MMISSI0NEfi AHEARNE: As oppos.ed to a 25 necessary

703 03

.--. r-U'.::J 36 rilgcJJAV CCUN\ISSIOr~ER GILH:Si<:Y: '/,/hat *aas the first one?

- 2 J

COMi:\I SSIO:'-iEH Ai-lEAR;-JE:

DR. MATI"So1,f:

,,[arginal.

I think its more than marginal. I

- 4 5

6 think it is a nacessary chan~e, and perhaps as people think about this specific recommendation with some time more than we have had, they will do it in addition to the trial in the 7 way that you SU;Jgested.

b There/s another recommendation, another

';I consideration in here. Another way to achieve this

• 10 integration is to have an accident analysis function within 11 the staff different from the one we have today. You start 12 with the initiating event, carry through the dynamic IJ response of the machine and the operators to the mitigation 14 of consequences and eventually to the emergency procedure 15 and what have you, trying to integrate margin and 16 contributions from the individual review disciplines that 17 have been made along the way.

lb In the developnent of the Safety. Evaluation 1 t; Report, for ~x2mple -- that would be a way to cut across a 20 number of cases rather than trying the interdisciplinary 21 system level review on individual specific cases. So you 22 might want to try both simultaneously.

23 Part of the difficulty here is understanding when 24 and how to restructure NRR to achieve some of these goals, 25 and we have not written a roadmap for Mr. Denton to follow

l,oJ 03 06 37 mgcuAV in how to restructure iRR. He has a number of other

- 2 3

4 interests, a number or other policy interests, and this one we've dwelt on here is one ot the things we think he ought to bear in mind as he goes about choosing that new

) structure.

6 COi,iMISSIONER AHEARNE~ But to get back to the 7 point that Commissioner Gilinsky had made there, one of the b questions certainly that I will end up being interested in y is whether you see this provision as being a better way to 10 use the resources that you think are adequate, or whether I1 you chink that given the limitation on resources, this is 12 the best way to use them, or whether you feel that this 13 ap~roach ought to be added on top of the current approach.

Dr?. iilAITSON: I haven"t a good answer to the 14 15 question. I wi 11 say that in looking at how the job ought 16 to be cone, we didn.,t say to ourselves 11 with the six or 17 seven hundred people that NRR has today"; however, we did lb recognize the ability to lend, train, and use resources.

lY . .Our la.r,;;-e-r numb,~r is limited or has been limited in the 20 past.

21 Wall, forging ahead with this slide, attention to 22 unresolved safety issues is something that we just have to 23 mention. It.,s not really a lesson from Three Mile Island, 24 but Mr. Denton and yo~ shared in that decision earlier this

- 25 summer and decided to dedicate the people to unresolved

,*03 03 07 38 rngcuAv safety issues in the way that they were before Three Mile

- 2 3

4 Island. That's an excellent decision. It has to continue.

Prograss is being made, as I understand it, on unresolved safety issues.

5 And then turning to the next bulletin, we think 6 you're going to have to come to some analogous situation 7 with the lessons yet to be learned from operating 0 experience. It isn't good enough to stop with a really good

':I system and people for reviewing operating experience.

10 You've also got to have criteria and objectives for deciding 11 what to fix and be sure that you put the resources on it to 12 get it fixed. And I think we'll probably come to the same 13 concept that we've used on unresolved safety issues.

14 Finally, in improvements in reactor regulation,

- 15 16 17 there are many things going on, whether the staff -- as you are aware on the response capability and role of NRC, the executive management team, a number of things going on.

16 We added a narrow thought here that needs to be 1'7 factorea into the planning. There are NRR specialists, 20 analysts, experts of one sort or another that were called 21 upon at Three Mile Island and would be called upon in any 22 future emergency that need to do some preparing, some 23 thinking about who they are relative to one another, what's 24 the list of experts they_ought to have in their desk and

- 25 their phone numbers. How do they get mobile in a hurry?

1703 03 Ob 39 rngcLA\/ Should they conJuct some crills so t.hat they know ho1/1 to

- 2 3

4 interface with -che incident res;:Dnse center people, and what.

kind of arrangements are there in the Philips Building or elsewhere to do their ~*1ork?

..,.::; We're recommending that Mr. Denton identify the b technical specialists' group within NRR and give a couple 7 members of that group a few weeks of relief from other b duties to sit down, think through the things they need to y have including things in coordination with the Office of 10 Inspection and Enforcement., so that in the relatively short II future, within the next ~onth or so, this group of technical 12 specialists is ready to go and be depended upon.

13 COMMISSION!::!~ AHEAP.NE: Could I ask one perhaps 14 peripherally related question in the way of technical

- 15 16 17 support for emergency response? Did you give any thought anywhere in your deliberations on calculational support that might be needed as far as the NRC is concerned or NRR, its 18 ability to have runs made or analyses done rapidly?

l 'J .DR. MATTSON: Yes. The research program as I 20 recall, the briefing of a few weeks ago is already examining 21 the ways to put computer decks on existing machines in the 22 national laboratories, dclcks describing generically various 23 classes of re~ctors -- for example, a B&W deck that would be 24 available for running on command by the staff in the event 25 of a future emergency.

I 703 03 Qy 40 mgcUAV There,..s another area here that some of us reel 2 fairiy s-r:rongly about t.ha-c the Lessons Learned Task Force 3 does not. speak to that needs attention. It"s the question 4 of simulacors. Should NRC have one?

I think we need a simulator. I think we could use o a generic -- ana I"m tola a hybrid analog digital simulator 7 to help us in several respects. One is the evaluation of b operating experience, the ability to ask 11 what if 11 Y qua s-cions. Pennu ta tions and combinations of even ts are much 10 enhanced by a hands-on raal life kind of simulation, but 1l wi-ch acvanced software, you need the ability to couple in 12 the fast running versions of TRAC or RELAP or the more 13 advanced sophisticated codes. So if you"re generating 14 phenomenological consequences by your 111,vhat ifil questions 15 that are ordinarily handled in the course of the licensing 16 process, you"ve got codes that have some capability to be 17 modified to handle those consequences.

lo Another way it could help -- and one of the things 1'1 w.e .suggest for .. i.m,p.roving the licenstng -staff -- is practical 20 experience, in-house training in the use of simulators.

21 Even as .part of the licensing process, the integration of 22 procedures, for example, we say we oughc to review emergency 23 procedures as part of OL licensing. ACRS has insisted upon 24 it. We have agreed tci that.

25 The use of those emergency procedures, the review

1/03 03 10 41 1,19cu;W ot cne emergency procedures in conjunction with che design

- 2 3

4 reviev, proba:Jly doesn-'t add rnuch in terms of resources to the staff. ~e have to do the two simultaneously.

to integrate how it all works would be to use those And a way 5 emergency procedures on an NRC simulator.

6 There's also a third area of the use of simulators 7 during an accidsnc to diagnose or analyze alternative ways (j of ~etting out of und~sirable situations -- the way that y NASA has used simulators to understand alternative courses 10 of reaction to failures for spacecraft in flight.

11 I note the ~emeny Commission has looked into this 12 area. io some degree, I believe the experts is the Office 13 of Research have looked. As I understand it, if that will 14 ever happen in the future, that is accelerated time analysis

- 15 16 17 of real events, say ten times normal speed, there have got to be some pretty fundamental changes in the way we do the comiJucations with -..the big enorrilous loss of coolant accident lo codes like RELAP and TRAC, and there's consioerable 1~; development .required before that kind of use could be 20 realized.

21 It's something we~re continuing to think about.

22 These other uses of the simulator for more normal times I 23 think would be gDod.

24 You asked about analy.sis --

• 25 COMl,\ISSIONER AHEARNE: Well, I gather what you"re

r 42 703 OJ I I r;-igciJAV saying is tha:c in the more normal operations of ~RR, you

- 2 3

4 think it might be very valuable, but as far as specifically in :che emergency response mode. it sounds like you haven't yet reached a conclusion as to whether or not that might be 5 some thing.

6 DR. MATTSON: I understand up to this point 7 there's much development that has to be done there.

0 :che iemeny people have looked at it. Perhaps they have y found experts tnat see ways co do it that we haven't seen 10 yet.

1I The next slide, number five --

12 CO!,\MISSIONER GILINSKY: Before you go on, since 13 you mencioned Admiral Rickover, you cited him in your 14 report -- what is your reaction to his v.iew that simulators 15 ought not to be used in the training operators?

lo DR. MATISON: Well, as I understand, his basic 17 reason for discouraging the use of simulators is because he lb encourages the use of prototypes. That is, he says you need 1s, people :chat believe a ,dri.11 or exercise they're being 20 exposed to, rather than to sit back and say, 111

"/ell, if I 21 make a mistake on the simulator, it's just a mistake on a 22 simulator."

23 COMMISSIONER GILINSKY: He's gone beyond that. He 24 said, 11 We won-"'t have them. 11

• 25 DR. MATISON: But he has another alternative. His

703 03 12 43 mgc0AV prototypes are small. They're easier to cite. fhey were

- 2 3

4 there at the beginning of the program, when they were very useful in design decisions and things like that.

still used today for training.

They're We speak in the final report 5 to the use of in-plant drills to supplement simulator 0 training.

7 ~e're a little timid compared to some schools of b thought. There are people who suggest that in-plant drills y with some kinds of upset conditions might be reasonable to 10 conduct. At this point, I think that we would just II encourage that that continue to be looked at. We're not 12 satisified that that's safe, in the time that we've spent 13 looking at the use of drills. We think walkd-through drills 14 are important because you can involve many levels of the

- 15 16 17 operations staff in interactions among people and communications and things that you really don't do on a simulator with detailed training of operators.

18 CHAIRMAN HENDRIE; You get to go out and look at I~ the geography and remind yourself whdre it all is and which 20 way the valve goes, one thing and another.

21 I think if naval reactors did not have prototypes, 22 then they would have used simulators. And also, without 23 going into any detail, I think it's fair to say that the 24 Navy reactors, in view of their design characteristics --

- 25 you're able to run on the prototypes a set of exercises

703 03 13 44 nc;ciJAo/ thac, as Roger says, you would wan~ to think about very 2 carefully before you deliberately made that a matter of J normal training prac~ice on large power reactors which are 4 not set up to carry out some of the same evolutions that 5 military machines are.

6 7

b

':I 10 II 12 13 14

~* 15

'f 16 17 lb

]Y 20 21 22 23 24 25

r703 04 01 45 mgcDAV DR. MATISON: There's another possiblity. If the 2 reason for not wanting to depend on simulators is because of 3 the fact that they're not real consequences that you're 4 risking when you're on a simulator, then you might choose to 5 supplement simulator trainin;,.

6 Simulator training is important becau~e it's the 7 mock-up of the exact control room. At least some of them 8 are. That aspect is very important.

Y CHA IR MAN HENDRIE: You could have the fire sprays 10 come on if they blow the plant.

11 <Laughter.)

12 DR. MA Tl'SON: What I was going to suggest is that 13 if reality of potential consequences is of concern, you 14 might consider supplementing simulator training and in-plant 15 drills with training on smaller reactors which are not 16 prototypical reactors.

17 COMMISSIONER GILINSKY: That-"s what I'm really 18 asking about.

19 COMM rss I ONER AHE-ARNE: You say SU pplement, but not 20 replace?

21 DR. MATISON: Yes.

22 COMMISSIONER GILINSKY: My question is, how do you 23 feel about operators taking over the controls of a complex 24 reactor without ever having operated any reactor before?

25 DR. MATTSON: Well, we speak to some of those

703 04 02 46 I

r.1gcDAV things in some recommendations for future reactor operation,

- 2 3

4 at least for the senior reactor operators.

question.

It's a good We havenJ't thought that deeply about the specific sugyestion, and we don't treat it in the report.

5 But clearly, there are going to be qualification 6 requirements and training requirements changed beyond what 7 weJ"ve suggested over the next year or so.

• ."\

0 COMMISSIONER AHEARNE: I hope so.

9 DR. MATISON: IJ'm led to understand that the FAA 10 is reaching a point in the near future where pilots will 11 step into a 747 with people in the passenger area, having 12 never flown a 747 anywhere except in a simulator prior to 13 that time, but having flown an airplane somewhere before.

14 COMMISSIONElt AHEARNE: I'm assuming that they will

- 15 16 17 allow a person to be captain without having a co-captain?

DR. MATISON: No, I didn't say that, and, in fact, we don't say that in our final report. One of the things we 18 say ought to be instituted for senior operators, the 19 *~ommander.or the captain if you will, is that there be 20 requirements for service as a re~ctor operator before being 21 a senior.

22 COMMISSIONER GILINSKYz But I think your 23 qualificaon is important. This man presumably has had many 24 hours in the air.

- 25 COMMISSIONER AHEARNE: Right.

703 04 03 47 mgclJAV DR. MATISON: Let me forge ahead through plant

- 2 3

4 design, and then I'll try to get you some specifics.

(Slide.)

We think thare are *three areas in plant design 5 that require change: first, better implementation of 6 present requirements in one particular area that has to do 7 with the classification and qualification of equipment.

6 Now, we're not suggesting IEEE 3/23/1973 versus 9 IEEE 1971. It's a different topic than that -- relatable 10 but different. \'/hat we're saying is that the distinction 11 between safety grade equipment and non-safety grade 12 equipment has been too clear a line. In reality, it's not 13 that clear a line. It's a fuzzier line. And in between is 14 a class of equipment important to safety in ways that you

- 15 16 17 1

might not even have thought about at this point -- that is, systems interaction ways.

Now, how non-safety grade equipment can help you Id in the event of an accident and ought to be improved to lY .saf.ety_u.rade qual.ific::ations .or .some b.etter qualifications 20 than none is the subject, at least in part, of Regulatory 21 Guide 1. 97. It's II Instrumentation to Follow the Course of 22 an Accident. 11 That's a guide that starts with 23 instrumentation in normal operations, carries it through 24 transients and precursors up through accidents and all the

- 25 way through core melt.

703 04 04 48

[l(JCUAV COMMISSIONER AHEARNE: A Regulatory Guide not

- 2

.;i 4

widely implemented.

DR. MATTSON: A guide, though, that we told you when we issued the short term report, we had some specific 5 things in there that were part of it. We need to get on and 6 do those. We're doing it, and we needed a very rapid, 7 thorou~h revision of I .97 and then backfit it for b implementation across the board of the necessary elements 9 o9f 1 .97 operating plants.

10 That revision is well along. It goes to the ACRS 11 the first of next month for its review. And soon after that 12 or in parall8l, I expect it to come to the Ratchet Committee 13 for consideration.

14 COMMISSIONER .A.HEARNE: Safety improvement.

- 15 16 17 DR. MATTSON: Okay. But it leaves a hole, this addressing of the interaction and contribution of non-safety grade equipment. The hole is the deleterious effects that lb non-safety grade equipment can have on safety. It's really 19 systems inte~aaction as defined in the unresolved safety 20 issues to the Congress. The staff has been working on this 21 for now for several years, and frankly I think in my view 22 and in the view of the ACRS and others, it~s slow coming to 23 grips with the question.

24 Steve Hanauer's here. He'il be probably shooting

- 25 arrows in my back, because he thinks that he's got it well

703 04 05 49 mgcUAV in hand. But the difficulty, in my judgment, is the

- 2 4

complexity of solving that problem, given its dependence upon the uni~ue characteristics of individual plant designs. For those reasons -- and I won't go into them much

(-

.) more at this point -- the task force is suggesting that we 6 require an industry wide, plant by plant systems interaction 7 review of all equipment.

d COMMISSIONEH AHEARNE: On what kind of time scale?

9 DR. MATISON: On a time scale of a year or more.

10 I don't think it can be done meaningfully in less than a 11 year.

12 COMMISSIONER AHEARNE: Would this review be that 13 you would lay down criteria for the review?

14 DR. MATTSON: Yes, and that's difficult. The ACRS 15 just wrote a letter two days ago saying, 11 We saw the one on 16 Zion. We want to know about the one on Indian Point. We're 17 interested in suggestions." We looked at that yesterday to 18 see, could we take that letter, send it to 70 licensees and 19 say, 11 00 this failure modes and defects analysis sort of 20 systems interaction problem," and expect to get back 21 reviewable *results that could be interpreted for all plants.

22 It's not good enough yet. And how we specify what 23 we want them to do and how we want them to report it and how 24 we expect it to be accomplished in that review nseds ~ome

- 25 work. But it's not bleak.

703 04 06 50 ITILJCDAV Westinghouse just reported through Salem that it

- 2 3

had reviewed non-safety ~rada equipment within its scope of supply for deleterious effects from accidents that could

- 4 s

6 lead to unreviewed safety quastions that is, design basis events -chat exceeded what they told us about before.

found three.

They I don't remember the specifics of it today.

7 I've talked briefly to the Wastinghouse people, and what 8 they did was set up a matrix, and they systematically went y through the plant. It's interesting that they did this by 10 virtue of a requirement that we placed on them in the DSS 1l some months ago to look at environmental qualifications for 12 this aspect.

13 So what we're really doing here in this 14 . recommendation is extending it beyond electrical to include 15 mechanical, and say it's all systems, not just the 16 Westinghouse or the NSSS scope of supply. And failure modes 17 and defects is part of it, and you've got to look not just 16 for -- assuming the comp6nent doesn't function in the event 1 S,' of an-upset commission, .and therefore you won't rely upon it 20 in the analysis.

21 Assume it does function and functions wrong, and 22 it takes an accident to change the course of the event in a 23 way you hadn't thought of. It could be that it changes the 24 way that other equipment performs, or it could be that it

- 25 gives the operator misinformation that he's not trained to

703 04 07 51 m9cUAV accomodate. If that's what it does, then the answer to that 2 particular problem may lie in training and man-machine 3 changes in the control room *.

4 There are a variety of ways to solve the problems,

~ once you identify them.

6 COMMISSIONER AHEARNE: That sounds like a fairly 7 complex set of reviews.

8 DR. MATTSON: It is.

COM1,1ISSIONER AHEARNE: Are the utilities really 10 capable of doing that?

II DR. MATTSON: Only in groups. Utilities in groups 12 and with technical support of their architect engineers. I 13 think some utilities are capable thoroughly of doing it 14 themselves or vendors. I had an opportunity sitting with 15 the AIF policy committee on Three Mile Island, roughly at 16 the vice-president level from utilities, AEs, and vendors a 17 few weeks ago to talk about some of the thoughts that were 18 on our minds.

IY Thf.s is what I laid on the table. I asked them 20 the same question that you just asked. They said yes. They 21 thought it was a good idea. They ought to get on with it 22 and do it. Westinghouse at that time sketched for me the 23 process they had used in the. work they had done. I think 24 it's useful. It's an enormous resource question. It's hard 25 to envision using less than half a dozen people per plant to

703 04 08 52 mgcDAV do this review. Even if you say you are going to do it over

- 2 3

4 a year or more -- and right there is 400 and some man-years of really gooci engineers. These aren't fresh college graduates that understand systems interactions and defects 5 of a variety of equipment. They're good people.

6 CO:,!MISSIONEH AHEARNE: I was wondering where they 7 were all coming from.

8 DR. MATTSON: The question that has to be answered y is *11hether there's a significant enough improvement in 10 safety to expect from this kind of thing to expend those 11 resources at this time when those resources are being spent 12 on something else.

13 COMMISSIONER GILINSKY: And you would perform the l4 equivalent review here for those plants that are not yet l5 licensed?

16 DR. MATISON: No. For those plants that aren*'t 17 licensed, if this became a requirement, I think you'd 18 require the license applrcanL to demonstrate that he had 1~ -performed. -this --k-i nd -of .reN-iew.

20 CHAIRMAN HENDRIE: Will a fast survey, fault tree 21 analysis be in advance of such things if they're done. That 22 is, would those results be available before people got 23 started on failure mode and defect analyses?

24 DR. MATISON: That's a good question. I'm going 25 to talk in this next bullet about failure modes and defects

703 04 09 53 mgcDAV' analysis. ~aybe I ought to say what I know, and then we can

- 2 3

4 bring the two points together.

Lessons Learned is saying that we~ve had good experience this last summer i'n using risk assessment, 5 probablistic assessment techniques as an overlay to the 6 deterministic criteria and deciding whether the auxiliary 7 feedwater systems in the Westinghouse and combustion 8 engineering designs were reliable systems. In that

'} exercise, we used competent people which helped to shorten JO the time - very highly skilled people. But the review was 11 accomplished very shortly, apparently with good uniformity, 12 and significant imptovements in reliability were 13 accomplished with relatively insignificant changes in the 14 design of aux feedwater systems.

- 15 16 17 Now we say that we ought to use some form of probablistic assessment of system reliability beginning now in the licensing process. And there are choices as to how 18 to proceed. The Office of Research is working on a thing 19 called the IntegratBd Reliability Evaluation Program, which 20 is a thing agreed to in principle with the Director of NRR 21 for simplifying from event trees and fault trees and 22 training cadres of people to apply these simplified 23 techniques and to look at operating plants or plants near 24 the end of the OL licensing process where design details are

- 25 available,. to extend the reach and application of the

703 04 10 54 r.igcUAV reactor safety study methodology. The Task Force completely

- 2 3

4 su p,;::orts that effort, and we encourage its use as an over lay to the exis-cin_g deterministic criteria.

The vibrations we received from the ACRS are 5 really more than vibrations. They are in their letter on 6 the response to the short term lessons, with strong support 7 for using these integrated reliability assessment techniques 8 from the reactor safety study.

9 (Commissioner Gilinsky left the room at 10:50 10 a.m.)

ll There are several ways to come at it. You can 12 take individual plants and do all systems, and you can do 13 particular systems and do them on all plants. The latter 14 approach was the one that was used last summer by the 15 Bulletins and. Orders Task Force, using people from the 16 research laboratories to look at aux feedwater. I think the 17 off i c e of Re sear c h pref er s to us e ind iv id u a 1 plan t s ,

18 probably plants different than the plants that were used in 19 t-he reactor -safety study. We don't have any special quarrel 20 with that, excep-c that to be alert, as you go through these 21 individual plants, as you identify systems that may have 22 generic problems that you can stop, look at that system 23 broadly in all plants like aux feedwater that was done last 24 summer and make decisions on those individual systems.

- 25 I'm sorry, Mr. Chairman. The two approaches are

03 04 l l 55 mgcDAV complementary to ona another on these two bullets. And the 2 way tha~ they are married and moved forward should be 3 carefully coordinated.

4 I think in the action plan that we recommend that 5 there's time in the next month or so for the Office of 6 Research and the Office of NRR to get their heads together 7 on what's necessary here and what the two things wi 11 8 accomplish, what approaches are going to be taken plant by 9 plant or system by system, and then make decisions on which 10 to do first, or whether to do it in parallel.

11 CHAIRMAN HENDRIE: Well, it seems to me that the 12 IREP results are likely to be incoming through the first, 13 probably by a fair margin in time. And I wouldn't like to 14 see that held up. I think it's an important effort because 15 the aim there is to see that if there are high probability 16 vulnerabilities out there that we haven't come to in the 17 classic review process 18 DR. MATTSON: I was not suggesting that !REP would 19 be the one to wait. I was suggesting that if there was 20 something useful that !REP would produce, the first 21 recommendation -- then perhaps we've got to consider that.

22 I suspect the answer is that you need to do them both in 23 parallel. The IREP is being done pretty much by the staff 24 and has a crosscut of system reliability and the failure 25 modes and defects analysis of non-safety grade equipment

703 04 12 56 mgcDAV being done as a sort of fundamental re-look at the

- 2 3

4 conformance with the regulations that speak to the goodness of equipment importance to safety.

COMMISSIONER AHEARNE: As far as the how-to-lay-on 5 to the utilities, do you expect though that the actual plan 6 will address how you would develop and flesh out the 7 instructions you are giving?

8 DR. MATTSON: Yes.

9 MR. LEVINE: I haven't seen the matrix that 10 Roger's talking about that Westinghouse used to do this 11 system interaction stuff. It sounds to me like there could 12 be considerable overlap between IREP and that program, and 13 both these programs are going to tax resources enormously 14 across the country. I would sugges~ that Research and IREP 15 get together and see if we can make an accomodation.

16 COMMISSIONER AHEARNE: It seems to be a reasonable 17 request.

lo DR. MATISON: Well, the last bullet on this page 19 .has to do with the ques*tion of whether design features need 20 to be added to existing plants and future plants for 21 mitigation of degraded core and core melt accidents.

22 The way we started with this problem was the 23 question of hydrogen, the amount of hydrogen generated at 24 Three Mile Island, and the fact that it exc.eeded the design

- 25 basis for hydrogen in all plants. It was a large dry PWR

703 04 13 57 mgcDAV containment at 'Three ;,Hle Island. The same amount of

(

2 hydrogen in other containment would have generated different 3 kinds of problems. Not all of them were difficult as the 4 ones at Three Mile Island, but some of them were difficult.

5 Smaller containments with s;naller design pressure 6 or with different kinds of equipment inside could have 7 experienced different consequences as a result of the 8 hydrogen burning inside of Three Mile Island, and what we 9 decided for reasons explained in some detail in the report 10 is that we should come ac this questiop of hydrogen 11 narrowly. So we backed off, and we started with the 12 question of defense in depth and its embodiment as a concept 13 in our regulations.

14 We say that we think there are three levels of 15 difference in depth, the first level being for normal 16 operations and expected events and the design objective 17 being as low as reasonably achisvable, off-site release; the 18 second level being the design bases for accidents, there the l'J objective-.bei.ng-otf~site cons_equenc,es less than the Part 100 20 guidelines. And the design basis, of course, sets the 21 design of certain engineered safety f~atures like emergency 22 core cooling and containment.

23 The third level of defense in depth is less 24 completely articulated, but it's there, in our Judgment, 25 never the less. That is consequence mitigation .or 1 imi ting,

703 04 14 58 mgcDAV including accidents beyond the design basis for enginsered 2 safety features. We would put in that level o~ defense 3 emergency preparedness, containment leakage, some of the 4 environmental qualifications that we require like radiation 5 for things inside of containment, and our siting policies 6 certainly give us consideration of this third .level of 7 defense in depth.

8 9

10

~ II 1

¥ 12 13 14

- 15 16 17 18 19 20 21 22 23 24

- 25

703.05.1 59 DH gsh *I lihat we see is a piecemeal mit-igation str::itegy. It 2 hasn't been thought through in terms of the. relation of 3 the design to the emergency preparedness and deciding in any 4 definitive way since the early 1960s.

~ And what we see in Three Mile Isl~nd that 6 significantly exceeded the design basis. And by looking at

, that* event, we see other ev9 nts in our opera ting expe ri en ce a that have involved sequences of failure beyond the design 9 basis in several respects.

10 One other example is sufficient to illustrate the 11 point - the Davis-Besse transient.

12 COMMISSIONER AHEARNE: You mean the earlier one.

13 DR. MATTSON: No, the Davis-Besse transient of 14 December, 1971. That one stands out more in my mind than l::5 others, perhaps.

16 We further, for r9asons explained in the report, 11 come to the judgment that TMI 2 is a significant precursor 18 of core melt and we come to the judgment for much the same 1~ reasons that we did in the context of Af~S that it is very 20 di.fficult to prove or to a.ssure to a sufficiently high

• 21 level that prevention measures satisfy the National Safety

• 22 Goal for nuc !ear power plants. Recognize that r-~ve a !ready 23 said this morning that that goals needs better articulation.

24* rt comes out different than what we're presuming it would 25 come out. And some of our recommendations might be

-/03.05.2 61 DH gsh unnecessary. But judging what we think that national safety 2 policy is, we retcimmend that mitigation of ~egraded core 3 events and core melt events should be required. And we 4 stopped short of recommending that they become design basis

'.) accidents.

6 lie, instead, come with* *a recommendation U1at the design basis accidents and the designs supplied in present 8 systems, as evidenced by Three *Mile Island, have some 9 considerable capability beyond the design. Except in the 10 case of hydrogen and perhaps in the ability of emergency l I core cooling system to handle debris, which is the smaller I2 point.

13 Now by mitigation, I don't mean limit the 14 consequences of degraded core cooling events or core melt 1:5 events to part 100. I don't think that that~s a reasonable 16 expectation.

17 By miti,;;,ation, we mean cuttin9 the consequences 18 once they have gotten beyond the design basis of the lY facility * .And we think we s_ee potential for feasible 20 engineering changes to accomplish a further capability to 21 mitigate catastrophic consequences.

22 And those two feasible considerations, two feasible 23 design features that we recommend for high and prompt 24 attention are hydrogen control inside of containment for 25 the degraded core event and controlled, filtered venting of

703.05.3 61 DH gsh containments for the core melt event.

2 Now the controlle~ filtered venting, in our minds, 3 is a way to achieve significant off-site consequence 4 mitigation by assuring that the pathway to man is through

~ failure of the containment by melc through the base rather 6 than by a prompt overpressure above-ground failure and prompt release of the gaseous fission products to man.

8 That does several things. It buys you time, which 9 gives you the decay capability. It gives you better off-site 10* people management capabilities. It also give you the 11 capability to interdict the source of release. If ;ou can 12 control it, you can filter it. If you can filter it, you 13 can writer criteria for how much you want to knock the 14 dose or concentration of radioactivity down before you want*

- 1:S 16 to release it.

Now I said that it looks feasible for us to move in this direction. I have to qualify that feasibility.

I' 18 We are unable, given the current state of technology, 19 to come to you and say, her~ -are specific recommendations for 20 specific design features that you ought to consider for 21 backfit to all plants. There's more information that needs 22 to be gathered, more people that need to be involved in the 23 discussion.

24 It needs to be coupled with this -- what is our 2:5 national safety goal question?

"/03.05.4 62 DH gsh S8 we recommend that you move soon, and I jonJt 2 mean tomorrow, but within the coming months, with a notice 3 of intent to conduct rule-making.

4 That would give the agency the opportunity to put in

~ one place what its policy on catastrophic consequence 6 mitigation is, how to do siting, how to do design features, I how to do emergency plans, emergency training.

8 They all fit together in a consequence mitigation

-; strategy.

10 In that notice of intent period, nine months or 11 a year, there would be an opportunity to accelerate and 12 collect information from the reactor safety research program 13 now ongoing for improved safety, which contains a large 14 element on controlled, filtered venting containments and an 15 opportunity to explore alternative ways of hydrogen control 16 in addition to inerting.

1, For example, burning of hydrogen upon its generation.

18 And provide the commission with a better body of information 19 to make a decision on a proposed .rule requiring such 20 mitigation features, say a year from now or a year and a half 21 from now.

22 _Well, I spent an hour and a half doing the policy 23 things. If I could switch quickly to Slide No. 6, 1--d like to 24 run down some of the details.

2:S CS 1 i de.)

703.05.:5 63 DH gsh If you look through 6 to 10 of the slide pages,

- 2 3

you will find that there are 14 areas of recommendations.

Some are heavier than others and some of them have many

- 4

)

6 sub~arts.

I'll just try to run down them quickly and I'll tell you that having thought broadly, the way we described, I we came to 3ach of these individual things and said we need 8 to change.

-; UnJer utility management involvement, it's more 10 policy recommendation of the need for utility management to l1 recognize the role it's playing in assuring that people at 12 all levels attend to safety.

13 One specific change there would be to require 14 csrtificatiori by the vice president of operations of a 15 utility of the competence and fitness of his operator 16 license applicants.

l' COMMISS !ONER AHEARNE: Wou1 d we also set up some kind 18 of criterion that could be against which he would be making l':-1 the c*ertification?

20 DR. MATISON: Well, we have qualification criteria 21 and we have recommendations to improve those qualification 22 c-ri teri a.

23 So one of the things that he would be doing was that I

24 the person was qualified according to those criteria.

2:S COMMISSIONER BRADFORD: ~~hat are you really saying

703.05.6 64 DH gsh there, Roger? Thgt he expects the person to pass the exam?

2 DR. MATISON: No. He certifies that this is the 3 person that he personally has met, talked to, reviewed his 4 credentials, reviewed his capabiliti~s and, in his judgment, J as the senior person responsible for safe operations, this is 6 the person to be at the controls in the control room.

I It's a little bit like Admiral Rickover saying that 3* he"s personally interviewed the en.Jineer-in-;1 officers of th3 J watch in his naval vessels.

10 COMMISSIONER AHEARNE: In a lot of cases in the JI Navy, they ,jo the interviewing even before they let them into 12 the program.*

13 DR. MAITS0N1 I would suspect that the vice president 14 for operations is not going to invest a whole bunch of resources and time on a person that heJs later going to find 16 he can't certify to the commission.

I7 COMMISSIONER KENNEDY: ,But he can-"t be interviewin9 18 if it precedes the manJs qualifications, can't be itself tha l~ ,basis for his tertifi~ation, which I think is the point.

20 DR. MAITSONz No. If I were the vice president, 21 I'd want to do somethin*g at the front end.

• I-'d probably 22 want to do some checking in the middle and then at the end, 23 all I'd have to do is make a certification.

24 COMMISSIONER AHEARN Et Speaking for myself, I-'m 25 hoping that NRC does something for that front end.

703.05.7 65 DH gsh OR. MATTSON: Okay, good point.

CHAIRMAN HENDRIE: ~hy don't we recognize that there 3 will be a paper to read and further reading, and th~ hand 4 tooling of each of these points at this time will r~sult in

~ our not getting through them because IJm going to break the 6 meeting at about 10 minutes of 12:00.

, OR. MATISON: Even worse, there are a number of task 8 force members in the audience. And when you gentlemen state 9 your personal preference, they'll want to change everything 10 that they've written and will never get the report out.

11 (Laughter.)

12 COMMISSIONER AHEARNE: You;re trying to 9et me to 13 believe that when we make a comment, the staff r-uns o.ff and 14 changes it?

b DR. MAITSONz I've seen it happen, sir.

16 <Laughter.)

I/ CHAIRMAN HENDRIE: I.,d just be pleased to know that la they had been listening.

I) (.Laughter.)

20 DR. MATTSON: In training programs, this particular 21 one is narrow to personnel other than operators and senior 22 reactor operators.

23 And what we recommend is that each licensee be 24 required to review over the course of, say, a year, using 25 position task analysis, every operations organization position

703.05.8 65 DH gsh and then thoroughly compare that position task 3nal1sis with the training received and make decisions on how the training 3 needs to be improved and report back on what has been done.

4 And that year is a good tLne, I think, because it

~ would be in parallel with the evolution of requirements and 6

• certifications and training programs from the Institute for Nuclear Power Operations. The two can build upon on~

8 another, the institute offecing advice and assistance to l individual utilities on how to go aoout meeting what the NRC 10 will expect in terms of criteria for these training programs 11 for other operators and senior reactor operators.

12 In plant drills - r~ve already mentioned that 13 briefly - each licensee establishes a program of in-plant 14 walk-through drills for the entire operational staff and the 15 technical management support staff.

16 A:;iain, we, I think, suggest the possibility of t, going further with jrills and prototypes along th~ lines that

(

18 Commissioner Gilinsky has offered.

IY In operator licensing, 8 specific recommendations --

20 I'll run through them quickly.

21 We think there's a need to identify by name the 22 sourc.es of operator errors in operating plants. ,Ve think we 23 have the information to do that. We think it could be 24 handled under the privacy act in an a.cceptaole manner for use by operator licensing reviewers and NRC.

703.05.9 67 DH gsh

• That is, when they have a person in front of them 2 who has exhlbi ted a tendency to make mistakes in th9 past, 3 they can try to discover the heart of the problem 2nd make 4 a decision as to whether he continues to be qualified to

'.) hold the li~ense.

6 In that regard, we think we would also improve

/ the interface or the collaooration between NRR and I&E in 8 making these license renewal decisions. There are ~xperiences

• ..; and knowledge held ny I&E inspectors by virtue of their 10 closer presence to the site about operating staff capabilities, 11 .;:>rooably ne-ad to be fed bact< into qualification examin-:1tions.

12 A unique recommendation is to require one week of 13 NRC staff training of all new reactor operators; 14 that is, establish a program where the rqgulatory 15 interest criteria, fundamentals, underlying policies, 16 expectations, what have you, are described and explained to I, all t' eactor operators.

18 It's a large resource problem. And we would 19 .antici,pat-e -that y:ou .don't as-si.gn so.m.eoody to that task forever, 20 that it" s a rotating assignment for people within review 2i organizations *

.22. It gives them som9 feedback and it makes sure that 23 you*have a continuing .. refreshing of that regulatory pers?ective 24 We would also require a multi-disciplinary NRC 2:5 staff oral examination of all new senior reactor operators and

703.05.10 68 DH gsh shift technical* advisors, and biennially thereafter. That is,

- 2 3

as part of the oral examination for senior reactor 8perators and shift tachnical advisors, we would subject them to Dore

- 4

..)

6 than an operator licensing examiner review.

ifa'd put them in front of a ooard, if you wi 11, of diverse skills and knowledge about the dynamic response of I nuclear power plants.

8 COM}liISSimH:R AHEARNE: And that would oe conducted

') oy the NRC?

10 DR. MATTSON: Yes. The passing of that examination JI would be a prerequisite to renewal of the license.

12 ~ow I should pause. We haven't got a license for 13 shift technical advisors yet. Down on I .8 at the bottom of 14 the page, we say that some people_ought to sit down and 15 thihk through a disciplined way of what other people ought to 16 be lfcensed.

I' ~e see a couple of prime candidates. A shift 18 technical advisor might be one. I tried not to prejudge what 1-J others might need to be. There's some expertise that ne~ ds 20 to be taken irito account.

21 Okay. We also waht to try to build another element 22 of this bridge between the staff and the operations crew. 1\/e 23 recommend an annual workshop of reactor operations to give 24 reactor operators an opportunity to come* back to the fount

- 25 of regulatory requirements, tell us what's wrong with them,

703. 05.11 69 DH gsh help us solve problems that they know aoout and help us help

- 2 3

them to understand what we-'re all about.

COMMISSIONER AHEARNE: you don't mean all operators.

- 4

)

6 DR. MATTSON: No. 1~hat *,1e would think would be one reactor operator and perhaps one senior operator for, say, a three- or five-day period oncg a year -- maybe you do it regionally, but perhaps you do it nationally. Probably you 8 do it nationally.

That would mean youJd break it into workshops. You JO would take on specific problem areas and current operating

-H experience with current 1 i c3 ns ing re qui re ment deve lop11ent s 12 or what have you and try to nuild that bridge a little batter.

13 There's a tremendous gap there. WeJve seen it 14 before Three Mile Island. I think you'll recall th:3 wor'<

1:5 done- at the 'fankee organization on understanding the needs 16 and interests of reactor oparators.

17 We've seen the gap and we think that a combination 18 of these things might accomplish it.

IY Another recommendation is that we know that the 20 Institute for Nuclear Power Operations is going to go into 21 the business of certifying. instructors for operations 22 training.

23 We specifically culled that area out as one that 24 ought to be looked at soon because there are in the paper 25 before you now, on the operator licensing branch, there-'s a

7 "i'03.05. 12 10 OH gsh specific recommendation on how operator instructors ought to

- 3

~ be qualified.

That's 3 small point.

- 4 5

6

~e recommend placement of operator licensing examiners* at regional training centers. ~e'd like to see some of these examinations at the simulators rather than in the office in Bethesda. And our presence in the operating 8 license in this area, where there are a numoer of tr3ining 9 institutions is prooably a good idea.

to We also eliminate part-time operator licgnsing JI examinations. You see, we're moving in a direction of 12 building a stronger, broader operator licensing function that 13 has resource implications.

14 Under NRC staff coordination on qualifications and b training 15 COMMISSIONER AHEARNE: I Just have a comment on your I/ proposal. It's an insightful, intelligent -- it's very good.

18 DR. MATISON: NRC staff coordination thing here 19 means that -we s-ee a yariety of a.ctivities in industry and 20 professional societies with their standards-writing activities 21 in the staff on p~rsonnel qualifications and training.

22 We really need to have a better coordination and 23 a game plan. And writing this action plan is a chance to draw 24 all this together.

2'.) We'll also have a chance to see Mr. Kemeny's

-,o3. 05. I 3 71 r H gsh recommendation in this area. And I suspect that th9y will

- 2.

3 dwell to some extent on presonnel qualifications.

Licensed reactor operators is a more significant

- 4 3'

5 area. I'm going to make recourse to the specific words.

rec~mmend tnat the shift supervisor -- that is, the person 1n charge of operations on shift at the station - should We hav3 'at least a bachelor of science degree in engineering or 8 the physical sciences and that he should also hold a senior 9 reactor operator's license issued under the new requiremants 10 and have served as a senior reactor operator for one year 11 before he is in this command position.

I~ I should say at the outset that these specific 13 things cannot be accomplished tomo.rrow. ,fa recommend in the J4* current draft over the next period of *the next years to 1:5 work for this. And that four years might *even change 16 between now and Friday.

1, We're tussling with whether four years is 18 necessary and sufficient time.

IJ C0.111-MTSSI Oi'*IER AHEKW E: You're pro ba bl y not go in*;, to 20 be able to pin it down much harder than that.

21 DR. MATISON: ,,e're probably going to put in thare 22 bachelor of science degree or equivalent training, but that 23 will put the onus on us and others to develop in some 24 detail what we mean by 11 or equivalent."

2:S I think that we've had some success with doing that

703.05.14 72 DH gsh with certified health physi~ists in the radiation protection

.2 area. And I think that we can prooably have the s3me kind of 3 success ~ith shift supervisors.

4

.)

5 I

8 10 11 12 l3 14

- l:5 16 I I 18 20 21 22 23 24

- 2:5

CR 7703 73 HOFFMAN t-6 -me 1 COMMISSIONER AHEARNE: You might do something like 2 requiring a degree, but with then a comment that it might be 3 modified or waived on a showing of equivalence; a little greater 4 requirement on them.

5 DR. MATTSON: Yes. The senior reactor operator, in 6 this same time period we recommend that senior reactor 7 operators should have at least the same general technical 8 education and training in transient and accident responses as 9 now required for the shift technical advisor. What that means 10 is roughly 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> of college-level training for all senior 11 level reactor operators, plus special training to the unique 12 characteristics of the individual design as to its dynamic

- 13 14 response to upset conditions.

In addition, senior reactor operators should have 15 two years of additional nuclear power plant experience beyond 16 that required of a reactor operator, with a minimum of six 17 months of the two additional years being served as a licensed 18 reactor operator.

19 There is another recornmendation .. in this area having 20 to do with whether or not 12 months to accomplish the funda-21 mentals course in operator training I'm sorry, 12 weeks

- 22 23 is sufficient. We don't think it is. We have our idea of what we think ought to be accomplished in that 12 weeks and

- 24 Ace-Federal Reporters, Inc.

25 we say somebody ought to spend some time and really think through that basic fundamentals course. We haven't had time

mte 2 74 2

to think through it and we think it needs to be.

The licensee technical and management support

- 3 4

5 COMMISSIONER AHEARNE: Could I switch to the gentleman on your left and ask him a question?

Ed, how does NRR view the recommendations coming out 6 of Lessons Learned with respect to the recommendations proposed 7 in the operator training paper? Do you intend to do any 8 revisiting of the operator training proposal?

9 MR. CASE: I think we'd revisit that based on input 10 from the Commissioners we've received from you, but not from.

11 the o:theri;;, as far as I know, and the Lessons Learned input.

12 So we'd take both of those into account in coming up with 13 recommendations.

14 DR. MATTSON: Licensee technical and management 15 support is really just the task force recognizing and supporting 16 the work going on with the Quality Assurance Branch. You 17 recall. Mr. Denton sent out a letter asking all licensees to 18 describe what they had in these areas. They are retaining a 19 contractor to review what the licensees have described and to 20 extract from work like ours and from other sources what the 21 criteria ought to be and g~t back to each of those licensees 22 with suggested required changes in their programs.

23 It's a little bit of a backwards way of coming at the 24 problem, although I think it was useful to start early to Ace-Federal Reponers, Inc.

25 heighten the sensitivity of individual licensees, if for no

rote 3 75 other reason than to make them think through what kind of 2 expertise might they need and where would they get it if they 3 needed it tomorrow.

4 This is something that the AIF Ad Hoc Policy 5 Committee on Three Mile Island has worked on fairly hard, the 6 capability of the industry to respond to accidents and, by 7 implication, then, building upon what individual licensees 8 ought to have in the future. So there is, I think, progress 9 being made by the Atomic Industrial Forum in this area. It's 10 an important area.

11 Licensing of additional operating personnel we've 12 described already. Let's go to the seventh slide.

13 (Slide.)

14 The staffing of the control room. We recommend two 15 things be done.

16 First, we need to get the regulations consistent with 17 more stringent staff requirements. We've been exceeding the 18 regulations for some years,for what we require for presence 19 in the control room. We need to keep bookkeeping straight 20 there.

21 But in addition, we need to consider whether what 22 we may require is enough. The Task Force recommends we 23 consider a requirement for an SRO. and two reactor operators in AceI 24 eral Reporters, Inc.

25 the control room at all times .. The senior reactor operator seems to be helped by the shift technical advisor in at least

mte 4 76 some plants. A number of operating plants will use persons 2 holding SRO licenses to meet the shift technical advisor 3 requirement beginning in January 1980.

4 Under working hours--this is not a problem that 5 flows directly from Three Mile Island; it's a problem that 6 flows from heightened sensitivity to operations, wh~ch flows 7 from Three Mile Island. And a number of people have said to 8 us, we need to do something about working hours for operations 9 crews in nuclear plants. We want them fit, we want them 10 capable and able to do what might be expected of them on a ll moment's notice.

12 We see evidence of places that use operating crews 13 and other places that abuse operating crews. We think there 14 ought to be regulatory requirements setting minimum criteria.

15 We suggest no more than two shifts of 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> on consecutive 16 days. That is, in any 24-hour period, you work 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> and 17 in the next 24-hour period you can only be expected to work 18 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, and then you're expected to return to a normal 19 8-hour shift duty.

20 Whether that's the right one or not, we're not 21 married to these numbers. We think the Operator Licensing 22 Branch and the staff people can come up with the criteria.

23 The point is there ought to be one.

24 Emergency procedures . . I've alluded to that already.

Ace-Federal Reporters, Inc.

25 First the staff review of all emergency procedures is the

mte 5 77 basic recommendation. The Bulletins & Orders Task Force 2 already has a significant start on that by virtue of their 3 review of small break LOCA procedures and guidelines and 4 analyses over the course of the summer. It's still ongoing.

5 The short-term Lessons Learned spoke to three phases 6 of reanalysis of transients and accidents, and in each phase 7 of the three phases there was a new guideline to review the 8 new procedures. So we're well on our way for operating plants.

9 We also have promised the ACRS that we will review 10 the emergency procedures on all new OLs. We said we'll 11 probably do the ones next year better than the ones we do 12 this year because this is a learning experience.

13 There are many suggestions, including an articulate 14 one from the ACRS, that maybe we write emergency procedures 15 the wrong way, that there's ways to use the written word in a 16 crisis situation that are preferable to other ways, and some, 17 in their fresh, renewed interest in emergency procedures, 18 reviewed the ones that* are there today and say, we've got to 19 be able to improve these things, they don't have enough 20 symptom instruction, they don't do this kind of instruction, 21 they give too much do-this kind of instruction.

22 People in the industry have suggested that maybe, 23 in addition to the instinctive response, rapid utilization e 24 Ace-Federal Reporters, Inc.

event-oriented procedures, it might be useful to have event 25 mitigation procedures which are more oriented to, I don't

mte 6 78 care how you got to the situation; if you're in the following 2 situation, here are alternative ways for getting out of the 3 situation. All of that with some discipline, with some 4 additional technical disciplines to the staff, I would think, 5 in education theory and crisis management theory, and consultants 6 or staff members if it's necessary to have them.

7 We need to go with improving emergency procedures 8 plant by plant the way we're doing it. But we also need to 9 come generically at, are we doing it the right way, to optimize 10 the way in which emergency procedures are handled.

11 CHAI.RMAN HENDRIE: You' re going to have to find *some 12 way in the typical plant to get them catalogued and simplified 13 a little better.

14 DR. MATTSON: Access is another question.

15 CHAIRMAN HENDRIE: When the board goes bang and after 16 you turn the enunciator horn off and contemplate what's going 17 on, it doesn't help much to face 12 looseleaf notebooks and 18 scratch your head over where in there is the procedure that 19 tells you what to do now.

20 Now, ob~iously, operators who have gone through 21 training and worked on a plant are in enormously better 22 situations to deal with the 12 notebooks than some character 23 who is trying to upgrade his own feeling for the thing and

- 24 has just been around a week in a simulator course. But there Ace-Federal Reporters, Inc.

25 is a guideline sort of problem there: How do you get speedily

mte 7 79 into the ones that you need, and so on?

2 DR . .MATTSON: One thing that helps that is the 3 addition of the extra person in the control room to stand 4 back and integrate and advis~ the shift technical advisor, 5 especially in the area of symptoms.

6 CHAIRMAN *HENDRIE: But whoever is to delve into that 7 collection of material, now, and find the right place, and 8 then once you've got one that seems to fit, time will often 9 press a bit. You're then faced with the proposition, really, 10 have I got the right procedure? Shouldn't I get on and follow 11 it, or should I spend some more time making sure that there 12 isn't a better one, or that there aren't two that I ought to 13 have before me to keep in mind as we go down this heretofore 14 unencountered peculiarity in the system?

15 DR. MATTSON: That's part of what I said we needed 16 to look more broadly at, this review, to make sure the right 17 technical information is in there plant by plant. We also 18 need to come back to some-of the theories of why and how 19 they're useful.

20 Verification of correct performance of operating 21 activities. Mr. Kennedy, I know this is an area in which 22 23 you've expressed to staff recently about double-checking components to s_ee that they' re aligned properly or performing AceI 24 eral Reporters, Inc.

25 properly after maintenance.

We've looked at this independent verification of

mte 8 80 correct performance of operating activities and think that 2 there's a need for two levels of verification. We would have 3 not only independent human verification by the licensee, but 4 also automatic verification by machine. Therefore, we're 5 recommending a backfit of Regulatory Guide 1.47 on status 6 monitoring of safety systems in the control room; and we're 7 recommending that each licensee establish a program for main-8 tenance, surveillance and other normal ~perating activities 9 verification.

10 We've already heard that some licensees plan to use 11 their shift technical advisor cadre in this capacity, that is, 12 this dedication to safety of the shift technical advisor 13 cadre, the things they do in addition to their prompt response 14 capabilities in the control room have to do with safety. In 15 some plants they'll do it the way we initially recommended it, 16 evaluating operating experience with that same group of people.

17 In some plants they'll also perform third party verification 18 of operating experience. I've alluded to thi's. Two points:

19 We completely support the establishment of NRC's 20 lead role and responsibility in evaluating operating experience.

21 We've already required licensees to have similar engineering 22 groups in place by the first of the year. The Institute for 23 Nuclear Power Operations will be along after the first of the

- . 24 year. All* of these people need to be tied together somehow Ace-l"iideral Reporters, Inc.

25 so that their lessons are shared, so that their insights are

mte9 81 shared, and we think that we should take the lead or the 2 agency should take the lead in establishing some sort of 3 network.

4 We also should pay attention not only to how these 5 lessons get resolved in the licensing process, but also to 6 how the lessons get down to the operator. I keep meeting 7 operations crews and I ask about this or that regulatory 8 pronouncement; they aren't familiar with what I'm talking 9 about. There's the need to keep communicating up and down 10 these lines of command.

11 COMMISSIONER AHEARNE: While you're getting those 12 educational psychologists to address how we'd better approach 13 defining emergency procedures, you might also address that 14 last question.

15 CHAIRMAN HENDRIE: If it turns out to be necessary, 16 we can keep in mind that each of the people that you want to 17 get to in fact holds a license from the agency, and we have 18 a direct string between this place and where that guy domiciles 19 and, if necessary, direct communication is an option.

20 DR *. MATTSON: That flavor is clearly on there.

21 CHAIRMAN HENDRIE: In addition to the periodic 22 workshop sorts of things, which provide a more generalized 23 COMMISSIONER KENNEDY: If you're going to do that, 24 you've got to be sure that you simply don't overwhelm the guy Ace-Federal Reporters, Inc.

25 with paper, which finally goes straight from his mailbox to

rote 10 82 his out box.

2 DR. MATTSON: The thing we're trying to achieve, 3 you :se~, is some significant advance in professionalism and 4 excellence of reactor operations as a career, as a life ~ork.

5 COMMISSIONER AHEARNE: Have you anywhere carried 6 with it any of your recommendations to go along with that, 7 some sense that the utilities might pay them more?

8 DR. MATTSON: I've carried it everywhere I've gone, 9 and I've been so abused for carrying it. Yes, sir. I don't 10 see how people will meet these criteria without paying them 11 more. And I firmly believe they ought to be paid more.

12 (Slide.)

13 The next page has to do with the man-machine 14 i~terface. I think one of the most useful, potentially most 15 productive recommendations in the task force is the 7.1 control 16 room reviews. Basically what we're saying is each licensee 17 perform a one year long review of his control room. We have 18 a list of things that are to be studied and accomplished in 19 the course of that review.

20 We're working now with the resources we have, and 21 the resources are growing, to develop criteria to promulgate 22 this requirement within the next few months. We think we 23 can actually initiate the review without completely definitive e, 24 Ace-Federal Reporters, Inc.

criteria, instead providing opportunities for exchange through 25 the course of the year as we have further learning experiences

mte 11 83 on control room design and control room backfitting.

2 I want to make it absolutely clear that we do not 3 envision a control room review where the licensee studies it 4 for a year, comes back and rationalizes why he didn't ma~e 5 changes. There are specific changes of a generic sort that 6 we know we will want to see.

7 One is the second point on this sheet, plant safety 8 status display.

9 (At 11:32 a.m., Commissioner Bradford leaves the room.)

10 DR. MATTSON: In the parlance of control room safety 11 engineers, we want them to describe to us a process variable 12 state vector, that is, a vector whose components characterize 13 the current status of the plant. That state vector we want 14 centrally displayed. We want it to be paramount in the 15 attention of training and accident response, training and 16 what have you.

17 It's also that state vector that the task force 18 feels is probably the set of information that gets communicated 19 to NRC. First, it ought to go to a central point in the 20 control room, and there ought to be good displays and diagnostic 21 devices applied to that information for the operator's use.

22 It also ought to go to the on-site technical support center, 23 which these people are all establishing pursuant to the 24 short-term lessons learn~d.

Ace-Federal Reporters, Inc.

25 On the on-site technical support center, you do a

84 mte 12 little bit different things to it. You want to be able to 2 record it so that when the people assemble in the on-site 3 training and control center, they can do some training without 4 having to go to the control room and bother the operators who 5 are still in the control room situation. And it also ought 6 to be transmitted off-site. 'There are people who suggested it 7 ought to go to the vendor. The industry, for example, suggested 8 that it ought to go to the vendor and the AE. The states will 9 probably suggest it ought to go to them. Staff will suggest 10 and I'm sure that som~ of you will, that it ought to go to the ll NRC.

12 Well, we're working hard and Saul has a research 13 program with Sandia that's vDrking on how to put things in the 14 right format and send them. We're requiring the licensees to 15 go out and thin~ about this state vector, because we think 16 that's the thing that ought to be sent to us. We're going to 17 hear from other investigatory groups as to the relative roles 18 of the state and Federal Government and the operators and what 19 have you.

20 By the time some of this technical work has progresse, 21 we'll know better who to send it to. At this point, we think 22 we'd like that state vector.

23 Disturbance analysis systems, we do not recommend 24 anything specific in plants today, and we have recommended Ace-Federal Reporters, Inc.

25 that Saul do some research.on disturbance analysis, and I

rote 13 85 think that's planned. These are the more sophisticated noise 2 analyses and diagnostic systems developed in Europe and in 3 this country. They have potential for operational aid in the 4 future.

5 Manual versus automatic operations. We tried to 6 come at the fundamental question of, should there be no.

7 reliance on the operator for the first 15 minutes, that is, 8 should you forbid operator intervention, or should there be 9 what we have now, which allows productive intervention but 10 still automates equipment that we think is so important that 11 we can't rely upon the operator to initiate it. Or I guess 12 the third possibility would go to non-automatic, all operator

-- 13 14 command.

There's been a standard under development in the 15 ANSI program for several years, and I think if I could charac-16 terize my view of the difficulty with that standard, it's the 11 reluctance on the part of the licensing staff to accept it.

18 It's a bunch of engineers trying to judge what's really a 19 human factors question.

20 Now, there are engineering elements to it, clearly, 21 and there are basic safety elements to it. But we recommend 22 that we do some serious study of manual versus automatic and 23 try to bring some clarity into this area of confusion down

- 24 through the years. We are not recommending that we prohibit Ace-rederal Reporters, Inc.

25 .operator intervention in the United States nuclear power

mte 14 86 plants, not at this time. I would say it's not even a strong 2 interest of the task force.

3 We talked about it. We looked at what others have 4 said about it. And we want the operator there for productive 5 intervention. We need to get to the question of how much 6 credit to give to the operator, not how much debit to take 7 away from him, because we're not that confident of his training 8 any more. When we get over that, then we can start talking 9 again about how much credit we give some of these training e-6 10 things.

11 12 13 14 15 16 17 18 19 20 21 22 23

- 24 Ace-Federal Reporters; Inc.

25

87

,oJ 07 OJ kapDAV Standard control room design, our outlook on the

• 2 3

-+

task force has been a retrospective outlook.

at what to do to operating plants.

ile*"re lookinq

~e 1 re doing the same thinJ in plants already de~igned and under construction, to

.) improve their safety ih ways we think are necessary, 5 expecting that if we do that Job correctly, and we I articulate out goals and objectives correctly, building on J

._) that experi9nce, then criteria for future plants will be a

-) natural consequence of our work.

IJ Standard control room designs we know are Il ongoing. They see~ to be learning from the experience of 12 Threa Mile Island and there seem to be significant changes 13 made in peo~le"s approach to control rooms for the futur9.

Ther~fore wa noted in passing in the report, and encouraged 14 1:5 that work to go forward.

15 (Slide.)

17 The next to the last slide, I have already treated 13 reliability assessments of final designs, the so-called IREP 1-) pro;vam that Saul is wor1cing on in concurrence with NRR.

2,J I'm sure we'll h~ve lots of discussions about the bast way 21 to do it, and I am sure we will eventually be up here with 22 you to te 11 you what we considered as al tern a ti ves and what 23 ways we've chosen.

24 Suffice it to say a recommendation on reliability

- 2j ass9ssment techniques be overlaid on deterministic critera

103 O l 02 88 kapDAV to identify outlyers and to identify outlyers in the sense of specific designs that aran't reliable enough compared to J the best available technology and other designs, for 4 example, and to identify weaknesses in- deterministic j For example, the single failure criterion does not

/ consider passive failures of components. In aux -fe3dwater 3 reviews this summer, we found some single passive failures J which were the key weak points in aux feedwater system IJ reliability.

11 Does that mean we ought to change the 12 detarministic criteria for application to future plants?

13 Review of safety classifications and qualifications -- I've 14 talked about that. Design features for core melt and Jj degraded core accidents, I've talked about that.

LS <Slide.)

I, The safety goal for reactor regulation on the last 13 slide -- again, we recommend a lot of attentio~ at the

1) policy level to this. ~e n3ed better subjective criteria 2~ and we think you can set quantitative criteria in some 21 areas. And we think we ought to get on with doing that. We 22 caution that there are plac9s where prooabilistic assessment 23 wor~s, in the licensing process. There ara places where it 24 doesn't work. We need to be careful *to recognize that fact 2j and try to distinguish the two-places. And we need to be

,*,03 0 7 03 89 kapDA' careful that probabilistic assessment doesn"t lengthen the 2 decision process. It has the capacity to do that when you 3 are reviewing somebody else's probabilistic assessment with

• 4

~

o your own. fhere are differences in methods; there are differences in judgment that have to be exercised in turning out those kinds of analyses. The goal ought to oe to set more definitive policy for making better and more d timely decisions and not to set a policy th~t makes the

') decisions harder and longer.

lJ Staff review objectives -- we started out a month 1I ago trying to draw an organization chart for Harold and we 12 gave that u~, decided we couldn"t do it. He had other 13 things besiJes Three Mile Island to be concerned with~

14 ~e"vs met with him. We've described some of the JS considerations that ought to go into his organizational 16 thoughts as we come off of these task forces and look to the II future.

13 We've listed eight objectives to meet in Jy restructuring the Office of Nuclear Reactor Regulation as 2J things we've learned about the way we do business as a 21 result of looking at Three Mile Island.

22 Finally, the NRR emergency response team, I've 23 explained the need for that and the preparation of the 24 technical backup and the management folks in an accident

- 2j situation. And that's all.

703 07 04 90 kapDAV COMMISSIONER KENNEDY: Thank you very much. That

• 2 3

was splendid.

COMMISSIONER AHEARNE: I look forward with great

• 4 6

interest to seeing the report.

CHAIRMAN HENDRIE: I think in view of the time, I'd be glad to defer any substantive further discussion until the meeting which I expect John, we ought to look 8 forward to in three weeks.

COMMISSIONER AHEARNE: That's fine with me.

18 \\"R. MATTSON: If I could offer a suggestion. It's 11 the same one I offered to ACRS in my last meeting with 12 them. Policy-makers and policy advisors have been used to 13 dealing with the same set of people in relation to fhree 14 Mila Island matters. ACRS hasn't seen much of anybody 15 except Mattson and Denny Ross. On lessons learned, you can 1.5 talk to me. It's necessary at this point, in my judgment,.

17 that we begin to broadly disseminate tha capability to speak 18 to these lessons learned within elements of NRR and 1~ *elsawhBre in the -stat~.

20 So, I will be pushing from my end, when naxt you 21 hear about lessons learned, and the action plan progressing, 22 you~11 be hearing from some of the line officials 23 responsible for the eventual execution of these 24 recommendations, rather than only from the people making the 25 recommendations.

103 07 05 91 kapDAV CO MMI SS I ON ER KENNEDY: A9r eed.

• 2 C0~AMISS!Oi'JER AHEARNE: I °touldn"'t want to let you 3 go away, though, without at least thanking not only you but

• 4

~

5 all the people on your effort, because clearly you have ~ut in a lot of work and I thin~ it's been a very useful one.

And I'm sure it will end up being quite significant for us.

I W~. MATTSON: I.,d like to say that it"s been an 3 extraordinary group of people. As a manager, if you can get

~ 15 ~ercent self-starters that work all night and give you*

IJ ev8rything they've qot, in an ordinary organization, 'you"re l l doing well. This organizatic:ln it was well intd the 90s. It 12 was very committee, hard-working - excellg.nt group of 13 people, to sit back and take off the b1inders and look 14 broadly and, I think, deeply.

15 COMMISSIONER KENNEDY: The briefing gives every 16 evidence of that.

11 CHAIRMAN HENDRIE: We thank you for the briefing 13 today, and when you succeed in getting it through the print

1) ~ho~, we'll be even more delighted to have the report of the 20 task force. Thank you.

21 (Whereupon, at 11 :45, the meeting was adjourned.)

22 * *

• 23 24

- 2:S