ML20126F589
| ML20126F589 | |
| Person / Time | |
|---|---|
| Issue date: | 12/17/1992 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| References | |
| REF-10CFR9.7 NUDOCS 9212310006 | |
| Download: ML20126F589 (92) | |
Text
--
.M%%%%%dd%f,%%%'n%%%%%%%%%%%1VgV6%%4%%ffgggggigg.
6
- At! SMIT'At. TO: /\/ Occument Control Dest, 016 Phillips g h -
2 The Public Occument Roem &
3 ADVANCED COPY TO:
- ATE
- /J// 8/'/ A
- SECY Correspondence & Records Branen g FROM: -
h 5
E Attacned are ccoies of a Comission meeting transcript and related meeting 5 document (s), They are being forvaroed for entry on the Daily Accession List and %
2, placement in the Public Document Row. No other distribution is reouested or i 5-5 recuireo. $
Yeeting
Title:
/$trY{, en fA . W 8s~o]) w ($;< n </h A
rf /C GLati _, & i>-t s i E
- j Meeting Cate:
f
/> // 7 /G A Open X' Closec f
l _
E
$ g
= '
!!em Oescriptien*: Copies -
"g '
C Advanced DCS 5
t_o POR C3 e
tg 1. TRANSCRIPT 1 1 "i, (L) [tuotweg 4_/Iv ) I3 lE d '
ii:
(W l
- 2. A zew L72 - 4'o 9 A l b
/ e E .- E Q %
3- .~
g
~
3 4 ;e 9 4 h
6 1 c
$ 5.
- r :
b h
w E
M, s. 23000.' - . _. _. 3 3,1 i
3 ^ 6 U
3.s
["
- DDR is acvanced one copy of each document, two of each SECY paper. l
@i 3 C&R Branch files the original transcript, with attachments, withcut SECY [
e gj ; apers, y; U ,f 3l 9212310006 921217 3[ PDR 10CFR
- gg
- 3 /RC '
PT9.7 PDR
%}
6
~ mmwwwwin *M
f' 4
.. g UNITED STATES OF AMERIC A NUCLEAR REGULATORY COMMIS SION m: . ' .
- 0. BRIEFING ON ROLE OF AEOD IN OVERSIGHT OF OPERATING REACTORS bCCiI' CI.* ROCKVILLE, MARYLAND b3I6l DECEMBER 17, 1992
=
Fig 55: 63 PAGES t
NEAL R. GROSS Al!D CO., IllC.
COLRi REPORTERS AND TRA%5CRIBER$
1323 Rhode Island Avenue, Northwest Washington, D.C. 20005 (202) 234-4433
+
DISCLAIMER This is an unofficial transcript of a meeting of i the United States Nuclear Regulatory Commission held on December 17, 1992 in the Commission's office at One l
White Flint North, Rockville, Maryland. The meeting was I
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 determination 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 herein, except as the Commission may authorize.
NEAL R. GROSS court RIPoRTIR$ AND TRANSCRIBERS 1323 RHoOf 15 LAND AVINUE, N.W.
WASHINGTON. D.C. 20005 (202) 232 4 600 (202) 234-4433
._ __ - _._._ - . . _ _ . . _ . _ _ . . ~ - . _ _ . _ . _ _ _ _ _ _ _ _ . . _
37 I
t-
^
1 UNITED STATES OF-AMERICA
.2 NUCLEAR REGULATORY COMMISSION 3 ***
4 5 BRIEFING ON ROLE OF AEOD 6 IN OVERSIGHT OF OPERATING REACTORS 7 ***
8 PUBLIC MEETING 9 ***
10 Nuclear Regulatory Commission 11 One White Flint North 12 Rockville, Maryland 13 14 Thursday 15 December 17, 1992 16 17 The Commission met in open session, . pursuant - to e 18 notice, at 10:00 a.m., the Honorable IVAN SELIN, Chairman 19 of the-Commission, presiding.
20 21 COMMISSIONERS PRESENT 22 IVAN~SELIN, Chairman of the-Commission 23 KENNETH C. ROGERS, Member of.the Commission-l
'24 FORREST J. REMICK, Member of the Commissio.
L 25 E. GAIL de PLANQUE, Member of the Commissio L NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 2344433 ' WASHINGTON. O C. 20005 _ (202) 2344433 _ -1
i l
2
- l 1 STAFF AND PRESENTERS. SEATED'AT THE C'OMMISSION TABLET-2 SAMUEL J. CHILK, Secretary- 1 1'
3 WILLIAM C. PARLER, General Counsel 4 JAMES TAYLOR, Executive Director for Operations 5 EDWARD JORDAN, Director, AEOD 6 R. LEE SPESSARD, -Director, Division of 7 Operational Assessment, AEOD 8 JACK ROSENTHAL, Chief, iteactor Operations 9 Analysis Branch, AEOD 10 PATRICK BARANOWSKY, Chief, Trends & Patterns 11 Analysis Branch, AEOD 12 13 14 15 16 17 18 19 20 21 22 23-24 l
! 25 l
! NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l- 1323 AsooE ISLAND AVENUE, N W.
(202) 2344433 WASHINGTON. D.C. 20005 (202) 2344423
. 3 1 PROCEEDINGS 2 (10:06 a.m.)
3 CHAIRMAN SELIN: Good morning, ladies and 4 gentlemen. This morning, the Commission will hear a S briefing by representatives of the Of fice for Analysis and 6 Evaluation of Operational Data concerning their role in 7 providing oversight of operating experience and 8 independent assessment of operational events and 9 incidents.
10 AEOD was established in 1979, to analyze 11 operational experience independently both for reactor and 12 for nonreactor facilities that the NRC licenses. It's 13 responsibilities were expanded in 1985, to include 14 incident investigations and, again, in 1987, to include 15 the independent diagnostic evaluations of operational ,
g 16 performance.
17 We're looking forward to- hearing about these 18 programs, their effectiveness in providing NRC with the 19 capability to-perform-independent reviews and assessments-20 of the safety performance of NRC license facilities, and
-21 there are likely to be some speculative. questions about 22 what other ways we might use AEOD and, Mr.-Jordan,-what 23 lessons you can draw from your experience as the,-if you 24 will, investigative or independent evaluation mission has 25 grown somewhat over time.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 234 4433 WASHINGTON, D C, 20005 (202) 2344433
4 .
1 Commissioners, do you have any comments?
2 COMMISSIONER ROGERS: No.
3 COMMISSIONER REMICK: No.
4 COMMISSIONER de PLANQUE: No.
5 CHAIRMAN SELIN: Mr. Taylor?
6 MR. TAYLOR: Good morning. With me ?t the table 7 from AEOD are Jack Rosenthal, Ed Jordan, Lee Spasard, and 8 Pat Baranowsky.
9 Today's briefing will concentrate on AEOD's ,
10 independent role and its assessment and investigation of 11 operational experience.
12 Mr. Chairman, you've outlined some key 13 historical points in the development of the of fice's role.
14 In my discussions with AEOD, I keep emphasizing their 15 specific role in reporting to me as providing a separate and independent safety conscience for overviewing what 16 17 happens in both the reactor and other technology fields 18 for which we have responsibility.
19 I will-now ask Ed Jordan to continue.
20 MR. JORDAN: Okay. Thank you, Mr. Taylor.
21 I've been fortunate to have participated in 22 essentially all phases of events assessment investigations 23 and feedback of operational experience from a regional 24 perspective, from the Office of Inspection Enforcement, 25 through.CRGR reviews of proposed generic communications, NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVCNUE N W.
(202) 2344433 WASHINGTON. D.C. 20005 (202) 2344433
,- 5 1 and to implementation of the current AEOD programs. ;
2 From that vantage point, I've seen the evolution 3 of both the nature of the problems and NRC's programs to-4 extract and feedback experience.
5 AEOD's role, as I see - it, I believe, as the-6 Chairman and Jim Taylor have identified, is tied to the-7 AEOD and the Commission and the public, added insurance 8 that important lessons are learned and retained.
9 The origin of the office recognized that this 10 necessitated AEOD provide through its reviews both 11 redundancy and diversity of reviews by the Program Office-12 and by the regions.
13 Our presentation this past September emphasized 14 products -rather than process. This presentation ,
15 necessarily will discuss process, and we will use current 16 examples of products in order to provide that perspective.-
17 AEOD's organizational feature is one- of' 18 reporting directly to the Executive Director for 19 Operations, 'of not being responsible for day-to-day-20' - inspection or regulation of licensees, yet maintaining a 1
21 close coordination with the Program Office activities in 22 order to utilize scarce resources as' effectively as 23 possible. In this way, AEOD provides a quality assurance 24 and - an oversighti function. I plan to illustrate that t 25 point during discussions of event assessment.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N.W. ,
(202) 2344433 WASHINGTON, D.C. 20005 (202) 2344433 l
. . - . - _ - . . ~ - , . - - - . . ~ - ,
q 6' .s i 1
a 1 In addition to describing past activities,:we 1 2 will also talk about changes ~ underway particularly in the 3 trends and patterns area, Pat Baranowsky's area. l
- l 4 Could I have the first slide, please? -(Slide)' -i 5 In my discussion, I plan to cover AEOD's 6 activities associated with _ operational experience 7 assessment, feedback of experience, -incident 8 investigation, diagnostic evaluation, and follow-up. of 9 recommendations, and. that will be the order of-10 presentation. I'll also have a summary at th~e end.
11 I don't plan to discuss incident response, 12 technical training, or operation center activities, since.
13 they do not directly contribute _to AEOD's role in
~
14 independent oversight of operating reactors.
15 Next slide, please. (Slide) 16 The ori.ginal focus of the of fice was operational 17 experience ' assessment. The current components of=that-18 assessment can be broken into six topics managed by two 19 -' branches. Jack Rosenthal and' his Reactor Operations 2 0._ _ Analysis Branch _ have oversight over events analysis, 21 accident- sequence; precursors, and . human performance i- 22 activities.- Pat Baranowsky's Trends and Patterns Branch 23 has oversight over.the activity.of trends and _. patterns
~24 which inclu' des performance. indicators, l abnormal' 25 occurrences. And:now I'd like to discuss each of those NEAL R. GROSS COURT REPORTERS AND TRANSCRl0ERS 1323 RHODE ISLAND AVENUE. N W.
(202) 2344433 WASHtNGTON, D.C, 20005 (202) 234-4433
.- __ _. . . - _ . . _ _ _ . _ . _ _ .~ _.
. 7 -_ .
= .-
,1 7 1 components.
2 could I have the next slide, please. (Slide) 3 I'd like to explain how Dr. Rosenthal and his 4 staff provide the redundancy and diversity in an 5 independent perspective yet complement activities of NRR 6 and the regions,-in evrnts analysis,- and to do-so-I need 7 to go through-the sequence of their activities.
8 Each day begins with a conference call between 9 the Operations Center, the Reactor Operatione Analysis 10 Branch, and NRR, in order to screen operational esvents 11 that have occurred since the previous-discuasion -- and 12 that is, from workday-to-workday, and then, of course, 13 some-accumulation over a weekend.
14 This is a value-added type discussion 15 interactive- between the . individuals, and it- quickly; 16 identifies short-term actions. AEOD often self-starts 17 based on a new event added to existing concerns or work in 18 progress.
=19 Where a real-time-human' performance assessment 20 is warranted, the action is proposed _to'AEOD management 21 based on that - morning- call, - and may be part of an 22 augmented inspection team as_an alternative.
23 The next major AEOD step is classification 1of 24 significance - from detailed review of -LERs, the License 25 - Event Report, af ter 30 days. This identifies the level of.
NEAL R. GROSS-COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 234 4433! WASH NOTON, D.C 20005 (202) 234 4433 4
-- -y y - .- , ,
1 1
l 8 ,
1 subsequent review a.nd may be the basis for further 2 engineering study, a case study technical review. And the 3 last sequential activity is a quantification of risk 4 through the accident sequence precursor methodology, for 5 the more significant events.
6 And, finally, although this sequence is 7 complete, each license event report, each entry, remains 8 in a database, and subsequent analysis may come back 9 through that particular issue, combine it with another 10 complementary report, and identify a new lesson to dig 11 into.
12 I'd like to use a current example to show how 13 this morning call causes the staff to look again at an 14 issue. Recently, we had an event at the Salem plant, this 15 was December 13th, regarding an annunciator failure.
16 We've had a number of annunciator failures in the past 17 year and, over the years, annunciator failures have been 18 of concern to us.
19 I'd like to have back-up slide number 1, please.
20 (Slide) 21 The frequency of annunciator failures is of the 22 order of ten a year, and this is either computer or the 23 actual visual display of windows. And, so, across our 24 plants, that would give the likelihood at a given plant of 25 the order of 10" per year. And if one looks then at the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE LSLAND AVENUE. N.W (202) 234-44J3 WASHINGTON D C. 20005 (202) 234-4433
. 1 loss of both the annunciator, the window, and the process-2 computer that provides a printout'of alarms -- and could 3 I have the next back-up slide, please'-- (Slide) -- those 4 are less than one per year across the set of plants, and 5 so this would be of the order of a 10-' frequency.
6 Neither the annunciators nor alarm computers are 7 safety-related. They are not required for continued 8 operation, however, their unavailability does necessitate 9 compensatory action by plant operators. It's-a challenge 10 to them. And a simultaneous transient, while the total 11 annunciator system, the visual and the printout, were-12 unavailable, would make the operators' work much more 13 difficult.
-14 So, based on these recurring events, the staff 15 is-looking more deeply into does something more need to be 16- done with regard to annunciators. And, so, that's an 17 _ example of how we continue to accumulate information'and 18 that one is work in-progress. So, I can't givy you..the -
19 outcome at-this time, j 20 Could I have the next slide,-please?' (Slide) 21 COMMISSIONER REMICK: .And I assume - when ' they -
22 look at it, they'11 look at the probability'of having'the-23 process-computer out, losing the annunciator panels,^and 24 the transient. That probability is probably pretty l small.
25 MR. JORDAN: Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W-(202) 234-4A33 WASHINGTON. O C. 20005 (202) 234-4433
_ - _ _ _ - _ _ . ._ _. . .. _ _ _ _ - _. ~_ _ . . _ . _ . _ _ . _ _ .
- 10 .
1 COMMISSIONER REMICK: _I would think so. '
2 -MR. JORDAN: Depending on.how long,- and whether 3 or not the operator is aware of it -- I guess _a key point.
4 I did not bring up with regards to the Salem event and the 5 Calloway event was, in those cases, the_ operators were not 6 immediately aware they didn't have the annunciators. . And, 7 so, they were even further handicapped.
8 COMMISSIONER REMICK: Yes.
9 MR. JORDAN: And in some cases, the annunciators 10 have been difficult-to get back on.- An event at Palo' 11 Verdi a year ago, it was several days in order to recover?
And,
~
12 the annunciator system. so, certainly, it's an 13 interesting area. It's one that is related to the human-14 performance, the additional challenge to the operator, and -
15 we're really prepared for it.
16 CHAIRMAN - SELIN: In the spirit of - trying to 17 concentrate more on procedural ~'and substantive questions 18 in an issue like this, specifically the annunciators,-do 19 you have access to, and do you go into PRAs, to see if=we:
20 can calculate a difference in operator performance with 21 and without-annunciators available, or is that beyond your -
-22 --
23 MR. JORDAN: That's an attempt we'll make,.:and
- - 24 I'll ask. Jack Rosenthal to respond.
25 MR. ROSENTHAL: That's it, we'11 make an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 234 4433 WA$HINGTON. O C 20005 (202) 2344433 i
~. . , , _
.~ _ _ . _ . _ . - _ . _ _ _ _ . _ . _ . . . . _ . . _ . _ _ _ . . _ - _ _ _ > .. _ _ . _ . _ _ .
.j -
11
~
1 attempt. I'll use the _ accident sequence precursor program 2 myself, and I've already spoken to RES, and they're going _
3 to try something by using NUREG 11-50 models, since 4 they're higher, so we'll work together.
-5 CHAIRMAN SELIN ' I would hope you would get 6 beyond these labels, which you know I'm not very _ fond _of -
7 - safeuy required, safety related, et cetera -- and try to 8 get directly to the probabilities, and see what we,know _
9 about people's performance .with and without annunciators,.
10 specifically when they don't have annunciators but they 11 don't know that they don't have- annunciators, as 12 Commissioner Remick said, there is a sequence of events, 13 not all of which are- moderately unlikely_ but, 14 nevertheless, the idea that AEOD won't be misled by some 15 label and take a look and _ say, you know, .we are not-16 comfortable that this is'well within the envelope: of?a 17 plant, is very attractive.
18 MR. TAYLOR: Yes, sir. I initiated a call when 19 I saw the last Salem event, keeping.in' mind the previous 20 occurrences, and that they are not_ all identical, but- that 21 it-was an appropriate situation to take a look-.at. I 22 don't know exactly what will come out: of . it, ,but it-23- appears to need some _ attention not' only by us', but perhaps
.24 . by industry, to look for what's causing the failures, any 25 common' features.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 2344433 WASHINGTON. D.C. 20005 (202) 234-4433 -
7 1 HR. JORDAN: And' my comments about not being- ;
2 safety-related and not being relied u'pon, is not an 3 indication that we don't examine - it. I mean, th'at's 4 simply the classification that has occurred traditionally.
5 And, so, we begin to say, now, based on this' experience 6 compiled, is there not a problem.
7 COMMISSIONER REMICK: The thing that surprised 8 me was an annunciator going off and not knowing it._ As 9 long as-the process computer is working, chances are they 30 are-going to hear the click, click, click of-the process 11 computer to know that they're --
-12 MR. JORDAN: Yes, right.
13 COMMISSIONER REMICK: -- and cause them to l'ook 14 up. But if you lose the annunciators, don't know you've _
15 lost it, and then lose the process computer, then-I think 3 16 you are kind of like in the position of not knowing things 17 are happening.
18 MR. JORDAN: That's a very awkward . position, 19 yes.
20 JiR. - ROSENTHAL: Again, almostl belaboring the:
21 point, technically,- I'm interested inlthose situations, 22- and we do have them,.in which there's an-electronic pinch' 23 point, so that- although the computer -is up and running and 24 executing the program, in fact, there's no information'to 25 the. computer at the very same time that the annunciator
. NEAL R. GROSS COURT REPVRit RS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W, (202) 234 4433 WASHINGTON, D C 200CJ - (202) 234 4433
- , , _ ,. , _ ,_,.-,..,, _ .,, - , . .....-._ , ._. _ ., __, - - . , . _ . _ . _ ~ . . _ _ - ~ . . . , . . _ . . . ,
. . - ~ - _ _ . . . _ , - _ . . .-
g .:
-- 13 :.
1 .has failed. From the, human' factor standpoint, wm.like to 2 use the term "alertment". Once -- I didn ' t make : it up ----
3 "alertment".
4 MR. JORDAN:- Is that a word? 4 5 MR. ROSENTHAL: Alertment.
6 CHAIRMAN SELIN: Whatever happened to English.
7 (Laughter.)
8 MR. JORDAN: Will my SpellCheck pull that up?
9 MR. ROSENTHAL: And that is, I think your 10 observation is very correct, and that is, once they_-are 11 aware that something's wrong, you can double crew, you're 12 at heightened level of awareness, et-cetera, .and you can 13 go. And, so, one of the concerns is, what' happens-when -
14 you think things are okay, and you're more relaxed an?, in 15 fact, you've lost one or more information - systems and.
16 don't know it.
17 CHAIRMAN-SELIN: Right. Commissioner Remick's 18 "observaments" are always right -- right on the money.
. 19 (Laughter.)-.
20 MR. JORDAN:- Could I have .the next slide,- -
21 plerse? .(Slide) 22 We discussed the accident sequence precursor-23 discussion at the September meeting. l And, previouslyl Dr.
l 24 Murley and myself had: briefed the Commission about;a-yearf 25 ago, on our uses of this - tool jointly between NRR and - -
i NEAL R. GROSS '
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W, l; (202) 234M33 WASHINGTON, O C. 2000$ (202) 234-4433 ,
]
i-
~ 14 - z.
1 AEOD. The tool remains a - very important element,=and 2 we're investing even further in it in order to-improve the 3 timeliness of information to reduce modeling weaknesses, 4 and to fill in the missing data for 1983 through 1984.
5 This past July, the accident sequence' precursor 6 output successfully identified the Shearon Harris safety 7 injection problem of 1991, as the largest contributor to 8 risk for the year. This was a value of the order-of 6 x 9 10 in terms of its core damage probability.
10 CHAIRMAN SELIN: Contingent core damage 11 probability.
12 MR. JORDAM: Yes, conditional core damage-13 probability.
14 CHAIRMAN SELIN: Are you involved, Mr. Jordan, 15 in an attempt to try to take this sequence of individual' 16 precursors and put them together and get some kind of 17 indicators about looking back over the year at each event-18 at its most risky point, if.you could somehow aggregate 19 over these events -- o 20 MR. JORDAN: Indeed.
21 CHAIRMAN SELIN: -- what the risk indicator risk -
22 was for the year,- whether - it's a sum of contingent 23 probabilities, or a post-year probability, or.just some-l 24 kind of replicable indicator? j
.1 25 'MR. ROSENTHAL: The staff is working on it.
NEAL R. GROSS -
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE N W ;
(202) 234 4 33 WASHINGTON. D.C. 20005 (202) 234-4433
.~I
y
. 15 1 -RES, Eric Becktro has the lead.- I attended a meeting,-
2 actually last week, with three statisticians and All--
3 Modaras f rom University of Maryland, and it was your idea, #
4 let's put this on a firm footing and to get something 5 published that the outside community woul.1 agree to, and 6 that is proceeding.
7 CHAIRMAN SELT.N: I think that -- I mean, I don't 8 want to mislead you. The most useful thing thet happens 9 out of these are the individual pieces of engineering.
10 This is, in effect, our practical, actual experience, and 11 models are not what we can learn from what really almost 12 happened or did happen is the most valuable thing,-but 13 some of these aggregated looks as well, to look at, saying-14 that is our empirical experience, what is the sort of sum 15 of how close we got to different accidents,-what's the 16 trend, and what are the effects would be useful, not as a 17 replacement for the detailed analysis that you do,'but.in L 18 addition to that analysis. ,
J 19 MR. JORDAN: I think it would be a beneficial-20 output-of some that would have more value. -The example ;
21 use of the Shearon Harris event being highlighted in its-22 significance by the aacident sequence precursor 181both 23 gratifying and frustrating. -And-Dr. Murley: and - I have 24 spent some time arguing and discussing the: issue.
25 In this case, the tool did what it was designed NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1,a3 RHODE ISLAND AVENUE, N W=
(202) 234 4433 WASHINGTON, D.C. 20005 (202) 2344433 j l
i _ __-_- - __--______ _- ______ _ _ _ . _
V 16 :.
1 to do, but our analysis was untimely, and we Lare - all 2 ' disappointed -- I think Jack Posenthal, that'his people-3 didn't find through the normal deterministic' reviews, the-4 true significance of the event. It was identified as 5 being "significant" by our classification, but not "really 6 significant" as the ASP did.
7 So, we are actively reducing this lag time, and-8 both AEOD and NRR are examining our own processes, to try 9 to make sure that conditions like existed at Shearon 10 Harris are treated in the same fashion as events in terms 11 of their risk significance and their interest to the 12 staff.
13 Could I have the next slide, please. -(Slide) 14 The Human Performance Case Study was published 15 earlier this month, and the important element ~here is-to 16 :be able to communicate back tc, industry as much as 17 practical of what we learned, in a timely fashion.- This 10- case study did not, in f act, como up with recommendations, 19 and so the discussions that we-have had internally and 20 with Tom Murley have identified:that the-best thingjwe :j.
21 can do is-to package the case study,- transmit ~it to 22 utilities w i t h - a p e r s o n a l- letter from . Tom and myself 23 indicating "it's available for your use", and not cause 24 licensees to respond to it directly, or to direct'them to 25 take actions.
NEAL R. GROSS COURT REPORTERS AND TRANSCHIBERS 1323 RHOOE ISt.AND AVENUE, N W.
9 02) 234-4433 WASHINGTON. D.C 20005 902) 2344433
- -.- =. - .- -- . . -. .. . - . - . -
.. 1.7 1 We recognize the complex natura of- the
-2 interaction between procedures,-training, organizational 3 f actors and, from the 16 events that we've examined, we're 4 most cautious about specific recommendations. -These +
5 combinations are different at different plants, and we 6 really.want to only convey our experience that utilities 7 may use as appropriate.
8 It does certainly identify the need for the 1 9 agency to be able to collect more information in order to 10 come to views about the shift technical advisors, about 11 the level of staffing, about the level of detail and 12 procedures. I think there are certainly some-principles 13 that can be extracted and fed back a little more strongly.
14 The work by the Office of Research led to'our 15 using this protocol, and there has been- a protocol
.16 provided to NRR that's in now the Human Performance 17 Procedure for inspectors to-apply in order.to attempt to- -
18- extract more lessons.
19 So, we're working with them in orderLto try to 20 collect information in a-little more efficient way, and 21 es.tablish a ' database that would be able to be ~ used' to 22 search and pull.up useful lessons.
23 Could I have the next --
24 COMMISSIONER REMICK:- Before leaving that, I was 25 a little surprised at your raising the issue of the dual.
. NEAl. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N.W.
(202) 234 4 33 WASHINGTON, D C, 20005 (202) 234 4 33
--.- -.- - .- ~ . . .- . - _ . . .. -.- - .~ . . - .
q, 11 8- . ..
l 1 - role again, although,you certain'ly should if there is a -
2_ question, but there - certainly . has been a .dif ference o f -- 1
-3 opinion on that forever.
4 Do you happen to know what the current 5 distribution might be amongst'the plants, of stand-alone 6 STA versus the dual role, how that stands?
7 MR. JORDAN: That information was just compiled.
8 Do we have an answer?
9 MR. ROSENTi:AL: We'll get that for you,-sir.-
10 MR. JORDAN- NRR did a survey of all plants to 11 get a clear understanding of how they wt.re vportioned.-
12 COMMISSIONER REMICK: I was c:4e who felt the 13 Commission did the right thing by going to dual rol'e,~so.
14 that in contrast to the cases - _I don't know if they 15 still exist -- where there was a young buck engineer
! 16 brought in, put in a trailer 24 -hours :a day,_ and if 17 something. happened,_ he was. called to the control room and,
- 18. if he ran, he could make it within the ten minutes, I 19 guess, that we required,_and so forth.- But I: alw'ays-20 Lquestioned how much help he_was really: going to be to-the 21 operators, and how much confidence --
22 MR. JORDAN -
I agree.
23 COMMISSIONER REMICK: -- and to that extreme.
24 And I always thought, well,-gee, having this person-on-25 . shif t where he's accepted, he knows what the plant l's, but NEAL R. GROSS .
COURT REPORTERS AND TRANSLRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 234 4433 WASHINGTON. D C. 20005 (202) 2344433 m--m.n -,
- c. -,.n.', g- <g .,,3.,ps,,
y .gyy-,,,.e99,,m.qy,.,"46'"H3 vg Wg- , v eT qW ys s'ipM+. y yyrg,y- - 7e w we -9 Mi-- vvg- ' - ' 7
-.: 19_-'
1 LI realize.that they f could have' dif ficulty in' standing 2 back. So, I don't know what the optimum is, but I' hope-3 when you're looking at it, you look at those two extremes, 4 if they exist.
5 MR. JORDAN: Exactly. And that's the problem we-6 have, is that the information is sparse in the way we are 7 collecting it now. 3o, we are unable to give any advice 8 as to'what generally works better and, you know, people 9 have opinions, but they're not really based on enough 10 experience to be able to convey.
11 COMMISSIONER REMICK: Yes.
12 MR. JORDAN: And we're very interested in being 13 able to collect more than sparse information. i 14 Could I have the next slide, please.
15 COMMISSIONER REMICK: Oh, I'm sorry. I blinked:
16 on your slido number 5. I had a-question there. You
, 17 point out that there are' failures, not always-in PRAs, and 18 I guess I'm not surprised at that'. What would.be more of 19 interest to me in the. failures that have been identified-j 20 -that were not considered in PRAs, what was-their,--using' ,
q 21 your words, "real significance" from a risk standpoint?-
22 MR. ROSENTHAL: :Let me give you an' example. One 23 of the top events was at Fitzpatrick, where we discovered 24 that the low pressure safety injection valves wouldn't 25 operate given a large break LOCA. Now, for a normal PRA l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W, I (202) 234 4433 WASHINGTON. D C. 20005 .(202) 234 4433
- . - - .- . .- ,--, .~ ..
- 20: .
1 modeler, you're not going to pay that much attention to! ,
2 large-break- LOCA, and you'll put more ' effort in the
~
3 detailed modeling of loss of off-site power, and certain
-4 transients, and odd small break LOCA. -Okay.
5 So, even in the Fitz PRA where they'had modeled 6 the low-pressure injection system and they-modeled the 7 pumps, and they_didn't even model the injection valves.
8 And that was the proper thing for the analyst to do,_as 9 long as he thought that the injection valve failure rate 10 was -3, that the pump failure rate was -2, and he's 11 talking about a -4 sequence in the first place.
12 The day you discover that you think that the 13 failure rate of-those valves given conditional on large-14 break LOCA is closer to 1, then that becomes the dominant 15 part of that train _ and, - in _f act, becomes an important 16 overall sequence because you perceive the system then 17 reliability low. So, that was son.ething where until the-18 operating experience revealed-it, it was just absolutaly.
19 neglected.
20 -MR. JORDAN: .That's a. good example.
21 COMMISSIONER REMICK:- From what you said, I 22 didn't get the impression-it was neglected, but-perhaps 23 the wrong conclusion had been drawn that it was the pumps 24 that were more important than the valves, and that's why 25 the valves were left out.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N.W.
(202) 234 4433 WASHINGTON D.C. 2000$ (202) 234-4433 i
vi.-= -w+ % -+r $ -g um m me
,- 21' 1 MR. ROSENTHAL: But if you use generic failure-2 data, then that is the right conclusion. Another one,.
3 again, discovered by operating experience, the electrical 4 cabinets for the diesel controls at a plant were not-5 ventilated. And through normal situations, you have room 6 cooling, and so that the system would run on test just.
7 fine. And-then it was recognized that if you, in fact, l
8 had a loss of off-site power, you would lose that -- the i i: -
1 cabinets.
9 normal room cooling, and then the electrical R 10 could overheat and, in turn, the very diesels that you're 11 relying on would fail.
12 And, so, you're picking up operating- experience l 13 insights that- way that just aren't in the PRA. That's one 14 of the terrific uses.
15 COMMISSIONER REMICK: I can understand that's 16 valuable. Two questions, follow-on questions, on that.
17 of those' two events,- what was the' ultimate risk l -
!- 18 significance? Was':it carried out, putting that in?
19 MR. ROSENTHAL: Y e s ,- these -were -4 type:
20 scenarios - -I mean, one of the' top ten' type events. . ,
21 COMMISSIONER REMICK: Okay. When you say_ -4, -4 ;
4 22 core damage frequency, or --
l 23 MR. ROSENTHAL: No, no. I'm sorry. These are 24 all -- I mean, all the numbers - that we're citing are g 25 conditional probabilities and, in fact, I can'see where L NEAL R. GROSS
. COURT REPORTERS AND TRANSCRIBERS 1323 RHOOE ISLAND AVENUE; N W.-
(202) 2344433 WASHINGTON D C 20005 (202) 234 4433
. _ ,_ . _. . ._ _ _ . _ . _ - _ . _ _ _ ._. _ _ _ _ _.~. ..
- 22. .,.
I this becomes tho' issue of-how dolyou relate thatEto the _
2 PRA.which is a frequency of core damage.
3 That was a --
the electrical cabinet problem 4 that I described andlthe. valve problem:1-described were 5 both conditions, and you multiply that by the likell' hood j 6 of the initiator event over the exposure period in order 1
7 to como up with'a measure of the' risk significance..
I 8 COMMISSIONER REMICK' Right, right. And I-9 didn't know if-that had been done or not.
10 MR. ROSENTHAL: Yes.
11 COMMISSIONER REMICK: .I'm not belittling the 12 fact that this information is extremely valuable. I'm 13 -just wondering, though,-from the standpoint.of either core 14 damage frequency or public risk, were these significant-15 findings or not? And I assume you don't have the answer.
16- T'.at 's okay.
17 The next question:is, suppose that.we-do-find 18 ;them to be significant, how do we thsn convey that-.out to 19- .the people who. will .' be doing PRAs next year -or.three 20 years down the line, that they think-about that failure?.
21 Do' we have a mechanism in those cases that; you . just 22 mentioned,'of getting that out?- Let's assume-that-they_
23 were ---- _ _
.24 MR. ROSENTHAL: You'll hear about the valves-in 25 a little while.
NEAL R. GROSS -
COURT REPORTERS AND TRANSCRIBERS 1323 RHOOC ISLAND AVENUE, N W.
(202) 234 4433 WASHINGTON, D C 20(X)$ (202) 234 4433 -
. . . . . . . . , . . , - . . . . . . , , , . , , , - .- - . . - - , , , . ~.
23 l 1 COMMISSIONER REMICK: Okay. .]
2 MR._ROSENTHAL: And generically, Ed?
3 COMMISSIONER REMICK: Same valves?
4 MR. ROSENTHAL: -It's through just -general 5 communication. I don't have a specific -targeted program.
i 6 MR. JORDAN: 'So, _if a particular --_let's say i 7 the cabinet was, in fact, a significant contributor to 8 risk, then we would recommend an action to communicate .it ~ )
I 9 through, at the very least, an information notice.
10 COMMISSIONER REMICK: Information notice, 11 something like that, yeah. Okay.
12 MR. JORDAN: And I think that is a key _we need 13 to keep in mind, that having information is useless. To 14 learn a lesson, you have to convey it to the industry or 15 to the regulator in the right. place.
16 COMMISSIONER REMICK: That's right.
17 MR. JORDAN: So, we're intent on that.
18 Could I have.the next slide, please ---Human 19 Performance. (Slide) Okay, we're done with that.-
20- Trends'and Patterns then. I'm-sorry. (Slide)
-21 The principal activities of trends and patterns-
-22 in the past, were aimed at identification of issues such 23 as to seek learning curve of-new plants in the early__
24 1980s. The development of performance indicators and 25- reporting of abnormal occurrences flowed from activities NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS .
1323 RHODE ISLAND AVENUE, N W.
(202) 2344 433 WASHINGTON, D.C. 20005 (202) 234 4 33
24 .
1 ' associated with developing, maintaining and using events 2 . databases.
3 Today, more analysis is being done with the data 4 in terms of statistical significance and risk perspective.
5 Emphasis includes conditions in addition to events. Dr.
6 Baranowsky's branch is actively developing tools to help 7 ' etermine d when a previously learned lesson was ineffective 8 or has worn off. Use of these tools include hardware 9 items, human performance, regulatory and safety issues, 10 and industry initiatives.
11 And, so, I think we're going to find that-the 12 events analysis side is more in the mode of finding, let's 13 say, new sequences or new problems, whereas the trends and 14 patterns is going to be refining or identifying cases 15 where previously learned, or previously: communicated, 16 issues have resurfaced. And certainly a clincher now in 17 order to have a response and a recognition of significance 18 is a meaningful assessment of-the risk associated with a
' 19.. particular issue.
20 And, Pat, since you've recently provided me'with -
21 .a program plan, would you like to elaborate on this?
22 MR. BARAN0WSKY: Okay. Basically, what we're-23 talking about doing is looking at, say,-an issue, whether 24 it's a hardware- or human performance issue,- and 25 identifying what level of risk was associated with it, or .
NEAL R. GROSS COUHT REPORTERS AND TRANSCHIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 234 4433 WASHINGTON. D C. 20005 (202) 234 4433
- - - - ---, -- . . . . - . . - _ - - - . . - -.. -~ ~ - _
s
$' ]
l
.- 12 5--
1- performance of a component, and then we initiate some sort _
'2 of regulatory activity, like a rule or a bulletin, or ]
3 maybe there's an industry initiative, and then tracking 4 performance subsequent to that, and identifying whether or.
5 not there actually has been an improvement in performance, 6 whether or not the types of events that occur subsequent 7 to the implementation at whatever activity occurred, are 8 consistent with the lessons that we had learned the first-9 time around when we came up with the fix, and then, given 10 whatever level of performance we identify, trying to be 11 able to put that into some kind of a risk context.
12 In this case, what we'd be doing. is using things 13 like the NUREG 11-50 PRA models and other PRA models, to 14 put things like equipment -failure rates, or- human-15 . performance estimates ia, that would be-based on the.most 16 current understanding ~that we have, based on-recent data 17 that we can get from LERs or other sources.- I t' d o e s n ' t .-
18 just have to-be LERs.
19 And-we're going to spend _a fair amoun't of time.
20 trying to develop a capability of doing ~ common cause 1
21 failure analysis in.a rigorous way, as opposed to saying L 22 in our judgment it has a certain'value,'which is one of 23 the weaknesses that we've seen-in some of the PRAs that l-24 were-done in the past.
25 We're trying to set up a database and' an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 2344 433 WASHINGTON. D.C. 20005 . (202) 234 4433
26 ,
1 analysis scheme that.makes use of the methods that were 2 developed by the NRC and EPRI. We've got the method. We 3 think we know how to do it. We just need to put them in 4 practice, and that requires setting up the data properly 5 and having the calculational capability in a software 6 package that we can use. And, so, we're working in AEOD 7 and with RES to set these types of things up.
8 HR. JORDAN: Okay. Can I have the next slide, 9 please? (Slide) 10 I'm not going to spend much time on performance 11 indicators. Because they do play an important role _and 12 are an objective measure of licensee safety performance, 13 I wanted to bring them up. They allow us to trend 14 individual plant performance,-and they also give us some 15 insights about industry perforrance. Since we've been 16 using the same indicators extracted in the same method for 17 some time, they are comparable.
18 CHAIRMAN SELIN: That's very good, but I wanted 19 to ask you if you've ever tried to systematically relate 20 thest- indicators to anything that's. more directly 21' concerned with risk._ I mean, can we correlate risks with 22 cmergency system activations? Is there any difference-23 between automatic scrams and manually-induced scrams, et
-24 cetera? In other words, have we carried the evaluation a 25 step further?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(20?) 2344 433 - WASHINGTQN. D C. 20005 (202) 234-4433
27 1 HR. JORDAtlt Yes. In fact, the origin of the 2 indicators was based on selecting Indicators that were 3 related to the frequency of arrival of transients and the 4 availability of the safety systems, and then trying to 5 look at organizational and human performance. The last 6 piece we still don't have a good handle on. It becomes an' 7 outcome. So, some of the indicators that we use do, in _
8 f act, relate to frequency of transients and safety system 9 reliability.
10 CllAIRMAll SELIN: That's all fine, but that has 11 more to do with you have plausible things to look at, but 12 my question is, have you gone the next step in evaluating 13 your own evaluators to go back over sometime and say, 14 we've been tracking these indicators, and we've been doing 15 nome -- whether it's your accident sequence or some other 16 calculations that are more empirical -- and it does, in 17 fact, look as if there is a pretty high correlation, or.
18 not much of a correlation, between risks or-performance 19 versus these performance indicators, or is that a 20 meaningless question, the answer is so tight you don't:
21 have to evaluate it.
22 MR. JORDAN:- Not a meaningless question.- Our 23 problem is, I think, _ insufficient data. The accident-24 sequence precursor is not dense enough in terms of data on 25 a plant basis and, by rolling it-up over time, the plant NEAL R. GROSS-COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W.
'(202) 234 4433 WASHINGTON. D C. 20005 (202) 2344433
r l
28 m !
1 performance changes --
2 CilAIRMAN SELIN: I wasn't thinking of plant, I 3 was thinking of the indus*try performance trend is where 4 you're basically averaging over the -- or aggrocating over 5 the plants to be looking for overall trends.
6 HR. -JORDAN: Oh. We have a pretty much 7 coincidence in terms of what we see of the aggregated a trends of performance indicators and what we see with the '
9 roll-up of the conditional core damage probabilities 10 across plants. So, where we were seeing steep improvement 11 in that conditional core damage probability, we saw steep i
12 improvement in the performance indicators as I've said.
13 CllAIRMAN SELIN: llave we published this _at some 14 point?
15 MR. JORDAN: Sir?
16 CllAIRMAN SELIN: llave we published this at some 17 point?
18 'MR. JORDAN: We convey it in a package fashion -
19 and, for - instance, - at- the last - briefing- we gave, we 20 displayed essentially without units, the- set of 21 performance indicators, showing generally the improvements 22 from the early '80s into 1990, 'and then a plateau. And we 23 can see f rom the accident sequence precursor data,'the I 24 same kind._of a slope during_the same time.
25 MR.-ROSENT!! alt. The same-correlation.
NEAL R. GROSS ,
COURT REPORTERS AND TRANSCRIBER $
13?3 RHODE ISLAND AVENUE, N W. -
.(?02) 234 4433 WASHINGTON, O C 20005 (202) 234 4433 V
. 29 1 MR. JORDA11: Yes.
2 MR. ROSEliTilAL: If you're looking for 3 quantitative statistical correlation, f or example, you can 4 take the ASP results, national average, and you know how 5 much of that is contributed by transients, you can 6 calculate trips as a function of time, and then you can 7 just do a correlation analysis and say, what's the 8 coefficient, and does it correlate and, in fact, it 9 correlated with a very high confidence limit.
10 C ll A I R M All S E L I ll: We've done this? You've done 11 that calculation?
12 MR. ROSEllTil AL: Yes.
13 MR. JORDAll: And similarly, in fact, in the 14 origin of the performance indicators, we were using SALP 15 as the benchmark, that if we could neo trends in plant 16 performance from year-to-year that were comparable with 17 trends in SALP, and show statistically a correlation, then 18 that was our validity check, and that was quite 19 successful.
20 coMMISSIor1ER ROGERS: Dut I wonder if that 21 really is valid. I mean, in the sense that when the 22 people are doing SALP evaluations, they have these 23 indicators as part of their information bane.
24 MR. JORDA!1: They didn't, at thc. time we were 25 creating the performance indicators.
NEAL R. GROSS 4
COURT AEPORTERS AND TRANLCAtBERS 1323 RHODE ISLAND AVENUE, N W (202) 234 4433 w ASHINGTON, D C 200M (202) 2344433
30 ,
1 COMMISSIONER ROGERS: Well, they do now.
2 MR.-JORDAN: They do now.
3 COMMISSIONER ROGERS: And, so --
4 MR. JORDAN: So, it could be a self-fulfilling 5 prophesy.
6 COMMISSIONER ROGERS Well, yeah. I mean, it's 7 hard to consider those as totally independent variables --
0 MR. JORDAN: We need another independent set of 9 something.
10 COMMISSIONER ROGERS: -- yeah, or assessments, 11 that if you're making a SALP assessment with this kind of 12 information in front of you, you know, you're going to 13 integrate those in your head, to some extent, and I don't 14 know that you can make that kind of a statement that one 15 validates the other, or vice-versa.
16 MR. JORDAN: Right, but the point I was making 17 was that at the origin of- the performance indicators, 18 -we'll use the existing SALP and backdate it in terms of 19 the events --
20 COMMISSIONER ROGERS . Yeah, from the very 21 beginning.
22 MR. JORDAN --
and that was - a very strong 23 correlation. ..
24 CHAIRMAN ' SELIN : - I'd like to make this quickly -
25 -
I'm not being critical of' what we're doing, and I'm NEAL'R. GROSS COURT REPORTERS AND TRANSCRIDERS 1323 RHOOL ISLAND AVENUE, N W, -
(202) 234 4433 WASHINGTON, D C. 20006_ (202) 2344433
_ =
. O l I 31 s,
S 1 certainly not being critical of these indicators as being 2 very plausible based on the information available, I'm 3 just saying in the spirit of evaluation, I'd like to see 4 some not evaluation, and Dr. Rosenthal has indicated more 5 of this has happened than I was aware of, where we look 6 not only at the plants, but we look at our methods and see 7 do these -- some of them are easily measurable, but 8 they're not directly related to end points. Others are 9 directly related to end points, but it's hard-to measure 10 them and get the quantity -- I mean, you can't look at the 11 core damage probability meter in a plant and say what is 12 it today.
13 HR. JORDA11 11 o .
14 CHAIRMA!1 SELIlit Dut it is important to continue 15 to do the evaluation or your evaluation tools, to see if 16 the surrogates are, in fact, as closely related to what 17 we're really interested in, as they appeared to.be when 18 they were first defined when you used the SALP scores, et 19 cetera. And this is just more of a -- as we go on, this-20 is something I'd be interested 'in as opposed to something, 21 gee, it's terrible that-you're not doing this, or that-22 there's reason to doubt that these performance indicators -
23 are pretty good, they probably are really quite good.
24 MR. JORDA!1 And-to be fair and direct, we have-25 reduced the developmental type ef fort in performance NEAL R. GROSS COURT REPORTERS AND TRANSCR;BERS 1323 RHODE ISLAND AVENUE, N W. -
(202) 2344433 . WASHINGTON, D C. 20005 (202) 2344433-
32 .
1 indicators. We say we have a set, they work reasonably 2 well. We have recently proposed peer groups and, in f act, 3 have now a view from the staff that a peer group breakout 4 in terms of three-loop Westinghouse plants, four-loop 5 Westinghouse plants, as peer groups, and CE plants and so 6 on, is a very beneficial cut, and gives us a little more l
7 insight about the set of plants within that group, and 8 also cut by fuel cycle, that if we're looking at 9 performance during operation --
10 CHAIRMAN SELIN: All I'm saying is that, on the.
11 one hand, the more detail you get, the more insight you 12 get into specific events, but the more you lose track of 13 whether overall safety is improving or not because -- for 14 the same reason, you're looking at smaller, smaller sets, 15 in more, more detail.
16 MR. JORDAN: Yes.
17 CHAIRMAN SELIN: When you look for these 18 aggregate indicators about, you know, how are we doing --
19 we're spending $500 million a year of eventually the 20 taxpayer and the ratepayer's money -- how are we doing?
21 Those get to be simple answers. And the more you try to 22 answer those, the more you lose the detail, and I just-23 want to make sure we're doing a little bit of-both, that 24 eventually we should be able -- have to come up with a -
25 relatively simple indicator of just overall performance, NEAL R. GROSS COURT REPORTERS AND TRANSCRIDERS {
1323 NHODE ISLAND AVENUE. N W.
(202) 2344433 WASHINGTON. D C 20005 (20 $ 2344433 a . _ _ . . . _ _.
33 :
1 oven though that obviously integrates over lots'and lots 2 of very interesting parameters.
3 MR. JORDAllt Yes. 1 i
4 Cl! AIRMAN SELIll And I'm somewhat reassutou that 5 we're doing more of that than I'm aware of, but --
6 MR. JORDA!1: We will convey more of it, yes, !
7 sir.
8 COMMISSIONER REMICK: One other question on >
I 9 performance indicators, and I can see how it can be 10 extremely valuable to AEOD, as I understand how you're 11 using it, but just as a matter of curiosity, what 12 indications do you have that other offices or regions 13 utilize the performance indicators? Do you have evidence P
14 of strong use? Do you get calls from regions that say "we 15 haven't gotten the most recent one, we need it"? ,
16 MR. JORDAN: I think the larger use is in 17 preparation for the semi-annual Reactor Safety meetings, 18 and so they are an input.- They help select which plants 19 get a full discussion, and help identify anomalies in 20 terms of here's a plant in performance indicator space
-21 that looks strange,.and we didn't have it on the-list to 22 talk about. -It's that kind of a check.
23 - We, of course, condition perf ormance indicators l -24 so that' the inspector -wouldn't go beat up on a particular 25 plant with respect to his outage-indicator declining. -So, L NEAL R. GROSS COUR1 HEPORTERS AND TRANSCRIBERS l7 1323 RHOOE ISLAND AVENUE, N W.
.(202) 234 4433 WASHINGTON, 0.0.20005 ' (?02) 234 4433
~
- +, ' - -, n -. L ~ . , .
i 1
34 ,
1 we purposefully direct the individual inspectors not to j 2 apply it directly. It's like the managers do, and wo do 3 get commenta from them in terms of its valuo.- :
4 COMMISSIONER REMICK: That reminds me, if I t
5 recall, back at the timo the pin were being developed, the 6 Commission directed that no other performance indicator 7 should be used by the staf f, they shouldn't have their own 8 set. Is there any indication that that's being fulfillod?
9 MR. JORDAN: Yes. We've had some occasions ,
10 along the way, and they have stopped thoso.
11 MR. TAYLOR: Yes, there were some homemado sets- .
i 12 that have disappeared by this time.
13 CilAIRMAN SELIN: I realize that I've asked most 14 of the questions, but'I'd like you to try to finish'by 1 15 about 20 after because I've - got _ about . ten minutes .of 16 related topic I'd like to ask you to speculate about when 17 we're done.
18 MR. TAYLOR: That's fine, Commissioner.
15 MR. JORDAN: Yes, sir. Okay. In abnormal 20 occurrences, the next slide -- (Slide) - -I- think the only :
21 thing I would say there is, this is a method for the 22 agency to communicate with Congress about the very worst 23 reactor events. And over the years, since 1987, those 24 events per year from reactors have declined to zeroes _and: +
'25 'ones, from fours-and fives.
NEAL R. GROSS
' COURT REPORTERS AND TRANSCRIBERS 1323 RHOOC iStAND AVENUE, N W.
(202) 234-4433 . WASHINGTON, D C 2000$ (202) 234 4433
- . - - , . _ , _ . . . _ , _ , _ . - - . . . , . _ _ _ _ _ . _ . . _ . _ _ - . . , _ , . . ~ . _ _ _ _ _ , . . _ , _ _ , , , .
4 35 1 Feedback of experience I'd like to spend just a 2 little bit of time on because that, I believe, is the 3 heart of our activity. That's slide number 10, please.
4 (Slide) 5 And this is a listing of what I think are very 6 important feedback that the AEOD office was instrumental 7 in through case studies aand having generated generic 8 letters, bulletins, and so on. And I would identify air 9 systems as being a nonsafety system that the staff 10 identified as being a significant contributor, and I think 11 that has been a great benefit.
12 I'd like to then jump all the way down to the 13 last one on the list, for which no action yet hac been 14 taken because we have only this month published our 15 report, and this is pressure locking of gate valves.
16 This is a recurring, continuing problem, "a 17 problem that hasn't been fixed" is the best way to 18 describe it, and so I think first I need to explain what 19 the safety issue is. These double-disk wedging gate 20 valves have the capability of a common-modo failure 21 because of a pressure locking of this wedging gate into 22 the valve body. The phenomenon is that the valve operator 23 cannot overcome these very high loads of the friction 24 forces and the actual direct force of this pressure.
2E Could I have the backup slide number 3, please.
NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
]
@ 2344M WisSHINGTON, O C 20005 (?O2)2344433
. - _ - . - - . . - ~ . . - - - - . - - - _ . - . - - _ . -
36 .
1 (Slido) . ;
2 A schematic of a gate valvo, and the issue is 3 that from the reactor side, which is not the flow side, 4 pressure over time and through a leaking check valvo, 5 - normally, causes a buildup in this valve bonnet, and it's 6 labolod excosa pressure. It would be building up to tho 7 same as reactor system pressure approaching it. Given a 8 depressurization incident, then - thic pressure that has 9 built up in the bonnet does not leak off in the time framo 10 that one needs to open this valve in order to inject low--
11 pressure injection RHR, or whatever the safety function 12 is, and this is a nonourvoillanco disclosed type event 13 because you don't do surveillance of that kind of valvo 14 while you're pressurized, nor do you do a surveillance in 15 the f ashion that would catch the kind of problem you have.
16 So, the instances that have been identified over 17 the years have been somewhat serendipity in terms of 18 identifying the problem. A simple solution exists, and 19 there is 'a drilled hole indicated on that particular 20 diagram that would vont then the flow side to the bonnet.
21- Another fix is a leak-of f line f rom the bonnot to tho flow-22 side of the valve, and there are other possibilities.- But 23 I think my main point here is that this is a relatively 24 low likollhood event, but one that is a common modo, if it 25 exists in one of the two trains, it likely exists in the NEAL R. GROSS COURT HEPORTERS AND TRANSCRIBEFIS 1373 RHODE ISLAND AVENUE, N W.
402) 2344433 WASHINGTON. O C. 20005 902) 2344433
, . . .. ,-- - .-- - - - . - _ , . 1.__..-.- -. , , . .~,-a . _ - _, - l- . . _ . , , - - - - - - - . - - - ._-u------.
e 37 j 1 second train,, and certainly the plant condition that ;
2 would get you there would be existent in the other system.
3 And, so, it's one that over years the NRC has communicated 4 to the industry. 3 5 The staff has done a survey -- and could I have 6 slide 12, please -- (Slide) -- at six plants and,-in this 7 case, the . pressure locking at those plants generally 8 remained a credible problem. And in talking to the 9 ongineering staff at the plant, they say, "Well, we've 10 never experienced a problem", and when knowledgeable NRC-11 staff explained in detail the sequence in which one could-12 get into the problem, then the utility people say, "Okay, 13 yeah, that could be an issue", and then they go look.
14 So, I think the real problem is that we haven't 15 communicated it adequately,-properly.- Tom Murley and I 16 have spent some time discussing this issue and how we 17 might properly communicate it in a way that will_get the 18 desired action.
19 We are proposing a bulletin or generic letter, ,
20 and we are planning a workshop with industry in one 21 regional area so that our-best engineers who are vitally-i 22 convinced of the significance of -the problem, can 23 communicate with plant engineers in that region, from the 24 utilities,- and exchange their understanding with the 25 experience of the utility people in order to come up with NEAL R. GROSS
- COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
..(202) 234 4 433 WASHINGTON, D C 2000$ QO2) 2344433 -
.___-_a.,
.._r__.,____,.. .,_,...# .___.u..._,.__ __ _. . . . _ - . .
38 .
I the proper fix. .
2 11ow , on slide 13, there is a list of 3 communications, starting from 1977, a circular, a 4 Westinghouse service letter, a previous AEOD study in 5 1984, an industry piece of feedback, and even a current 6 information notice.
7 CHAIRMAN SELIN: All of which suggests the 8 problem continues to exist.
9 MR. JORDAN: It continues to exist. The 10 frequency of occurrence of finding the problem, even 11 though it is, as I say, kind of a serendipity to find it, 12 has, in the last three years, been greater than previous 13 years. So, it still exists out there. And, so, I think 14 that identifies one of the roles of AEOD very clearly, to 15 continue to seek areas in which lessons simply haven't yet 16 been learned, or fully utilized.
17 COMMISSIONER ROGERS: Well, I'd certainly agree 18 with that but, you know,.it really strikes me as bizarre 19 that such a problem of this has existed for so long. I 20 mean, these valves have been used in industry for years-21 and years, and it's got to-be something that has been 22 experienced in many other contexts, and why on earth, you 23 know, it's still a problem, is beyond me.
24 MR. JORDAN: I wouldn't'want to spend much more 25 time on it, but there are some interesting --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N.W.
(202) 234 4433 WASHINGTON. O C. 2000$ (202) 234 4433 -
39 1 COMMISSIONER ROGERS: No, no, I think what 2 you're doing is absolutely the rignt thing.
3 MR. JORDAN No, I mean, in discussion. But 4 there are some interesting features. The upstream check 5 valve has to be assumed to be absolutely perfect, and the 6 gate valve has to be assumed to leak a little bit in order 7 not to have the problem. And the nature of things is that 8 the upstream check valve is going to leak, and we've 9 worked very hard on gate valves and leaks, and they seal 10 very well now.
11 CHAIRMAN SELIN: Congratulations.
12 (Laughter.)
13 MR. TAYLOR: So much for diligence.
14 COMMISSIONER REMICK: Again, a point of 15 clarification. You said something to the ef fect that more 16 recently you've found more of these.- Now, is this found 17 that valves did not open-when called upon, or found that 18 people were not aware of the problem to fix, or - -
19 MR. JORDAN: No,.these are actual cases where-20 the valve did not operate.. .
21 COMMISSIONER REMICK: Okay.
22 MR. JORDAN: . And then the study, the review at 23- six plants Lirientified that those six plants had done very :
, 24 -little, If.anything,-about this kind of an. issue. . 'And 25 this was a nonregulatory type review ---you-know, what is
.NEAL R. GROSS COURT REPORTERS AND TRANSCRIBER $ '
1323 RHODE ISLAND AVENVE, N W.-
(202) 234 4433 WASHINGTON. D C. 20005 (202) 234 4433
40 ,
1 your experience? What do you do about this?
2 COMMISSIONER REMICK: Okay.
3 MR. JORDAN: Okay. Next is slide 14 -- (Slide) 4 -- the incident investigation program. We have discussed 5 that program extensively with the Commission.
6 And I think I would make the point in terms of 7 sur role -- slide 15 -- (Slide) -- we are managing this -
8 program for the Executive Director for . Operations. We 9 maintain documentation. We provide training for a cadre 10 of people that are ready to do incident investigation, and 11 we provide administrative support, review the report for
, 12 completeness, and then -- I think the point I would want 13 to make then is, we continue to follow up on the 14 recommendations from the incident investigation program, 15 from the diagnostic evaluation program, and from review of 16 case studies, so that we are keeping track of those 17 recommendations issued under the EDO's signature, 18 verifying closure of those items, and looking at -- and 19 this is more recently -- looking at the adequacy of those +
20 reviews.
21 COMMISSIONER REMICK: How many IIT ao f ar,-
22 including the current one? Is it seven or eight'now?-
23 MR. TAYLOR: The current one will make it eight.
24 COMMISSIONER REMICK: Okay.
25 MR. JORDAN: Right.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 AHODE ISLAND AVENUE, N W.
(207) 2344433 WASHINGTON, D C 20005 (PO?) 7344433
. 41 i In the paper, you I 1 COMMISSIONER de PLANQUE:
2 talked about the option of using that in a manner where 3 you report directly to the Commission. Ilow do you 4 envision this? ,
5 MR. JORDAN: Okay. That's a new paper that's on 6 its way. i 7 MR. TAYLOR: It's signed.
8 MR. JORDAN: It's signed. Okay.
9 COMMISSIONER de PLANQUE: Okay. It's coming.
10 MR. TAYLOR: Right. It's in the mail. -
11 (Laughter.)
12 COMMISSIONER de PLANQUE: It's in the mail.
13 okay. ;
14 MR. SPESSARD: We had anticipated that one would 15 arrive before the other one.
16 COMMISSIONER de PLANQUE: Okay.
17 CIIAIRMAN SELIN: Another example of excessive s
18 diligence, obviously.
19 (Laughter.)
20 MR. JORDAN: I can talk ab'out it --
21 CIIAIRMAN SELIN: It's okay. No. We'd rather 22 have the paper in front of us - before we get into . a 23 discussion-on it.
-24 MR. JORDAN: .Okay. I'm going to' Lee, and I'm
- 25 going to hurt .your feelings . badly, skip over the l NEAL R. GROSS.
COURT REPORTERS AND TRANSCRIDERS
. 1323 RHODE ISLAND AVENUE, N W. -
(202) P344433 WASHEVOTON, D C. 20005 ' .(202) 234 4433 I
-- ,w - , - + , .-,-m,----,s--, ,, . ~ , . _ . - , . . -,,,-.,,.m.,-
, -,---m (
- - .- ~ - . . - - ..- - - - - ... - .- -- - - - -._- . - - .~ - -
i 42 ,
El 1 diagnontic evaluations.
2 COMMISSIOllER REMICK: I have a couple of ,
3 questions in that area, but --
.t I
4 HR. JORDAN All right, sir, why don't I respond i
5 to your questions then. ,
t 6 COMMISSIONER REMICK: Okay.
7 MR. JORDAlls Or maybe Lee should respond to your 8 questions.
9 COMMISSIONER REMICK: One -- and I'must edmit, 10 this goes back a few years ago -- I had some personal r
11 concerns about parts of the diagnostic evaluation, in the j r
12 human factors area, questionn that were suggested where 13 people quiz supervisors, what do they think of their ,
14 supervisor, and so forth. I had concerns about that, 15 those still continue. But some of the things that I have 16 heard about diagnostic evaluation --
and I'm not 17 criticizing the concept -- where there are indications.
18 that a plant might be heading for trouble, and the EDO.
19 approves a diagnostic evaluation to try to get at the root 20 cause and get a better understanding, that concept is1 21 good. i a
22 Some of the thingn I hear from people, though, .
23 that have .had them, are concerns 'about whether the 24 - findings and recommendations and the conclusions and recommendations 25 relate .to safety,. or. are they" just NEAL R. GROSS COVHT REPORTERS AND TRANSCRIBERS -
1323 RHODE ISLAND AVENUE N W, (202) 234-4433 WASHINGTON, D C. ?0005 (202)2344433 Y - vv'- 'r '" w yy ag--v % w, hapi,--
i c,'--yy ..-pp. -ywg .y
,g4.,y-i,-o,9,qpy-y-pp-ew----y-.yww-9 .ge_vg+,
-% p.n.--,--iey .swm gy +-,+-w--.. <9-'y yyWi i yvvqy, i-y
. r 43 1 Individual inspector's preferences.
2 Let me give you an example -- it's not an actual i 3 example, but it comes close to one that was related to me.
i 4 I'm on a diagnostic evaluation. The inspector is talking i 5 to somebody -- one of the trainers in the training group, 6 and he notes that they had intended to conduct some courses by the end of March, and they weren't conducted 7
8 until the first week of April. And the guy says, " Yeah,.
9 yeah, we had really hoped to do that, but we had a whole 10 group of new initial training candidates, and we have a I
11 sh'ortage of classrooms around here, so we had to delay it 12 about a week to get it done".
13 The inspector follows up with a training manager 14 or vice president and says, "I understand that you don't-P enough classrooms to conduct your training". And the guy 15 16 says, " Yeah, yeah, but we did get money for capital 17 improvements, it's now. out with-the architect, and we 18 expect in 18 months to two years, that. we'll have an 19 addition to our craining- center and have the classrooms".
20 The inspector goes_back and indicates that he-21 had found that there was -an inadequate number of 22 classrooms. .No mention that this had been-identified to- l 22 him. No mention of the f act that there was a program 24 under review. So, the finding comes out, -inadequate 25 classroom..
Conclusion:
~ Management . not paying - proper NEAL R. GROSS
- COURT REPORTER $ AND TRANScalBERS '
1323 RHODE ISLAND AVENVE.' N W. '
M P344433 WASHINGTON, D C 2000$ - (202) 2344433
__,._.._____.._.-_~.b. _ , - . . . . . . - . - - - - . , , , _ _ . , . . - - , ,. . - - - , - . - - - . . ,
44 ,
1 attention to the operational needs of the plant, something.
2 to that effect.
3 So, that goes out. The licensse then has to 4 spend some resources in responding to that, basically, 5 saying what was told to the inspector, and it has to go 6 through the review in their organization, through many 7 favored channels, making sure that the answer is correct.
8 Comes to the NRC, we go through the same process, and it 9 ends up the NRC says closed out, so no harm.
10 But when you take this, one, I think the fact 11 that it was not mentioned -- would have not been mentioned 12 that this was identified by the-licensee and there was.a 13 program under review or underway -- professionally and 14 ethically, I question. It might make the inspector look 15 like he did a good job, he uncovered something, and.so 16 forth.
17- Dut if you add all these up, and then the 18 question.of -- and I'm not questioning this'and whether 19 it's related to safcty -- but that there are a number of 20 things that people say "we can't find how it relates to.-
21 our regulations, and we can't-see how it really directly 22 affects public health and safety", you- add these up, 23 people responding, and then'the NRC saying, well, closed l- 24, out, they' are satisfied with the answer,. is just.
25 expenditure of effort.
NEAL Fi GROSS COURT REPORTERS AND TRANSCRIBERS j
1323 RHODE ISLAND AVENVE, N W.
I~
(202) 734 4433 WASHINGTON, D C. 20005 (202) 234-4433 L -
e-h r .
l ' 45
. , l 11 Hy question to you is, what do you do to make 2 sure that theco findings and recommendations do relate to-3J regulatory requirements, are professionally presented --
Il 4 'n other words, giving credit where they were told these 5 thingo, idencify r.rograms under review. and so forth, and .
6 that they do tave some bearing on public he:,lth and safety ,
7 rather than an observation. It might even be a _ good 8 recommendation. In fact, I think in every case that 9 per pie have made these comments, they say "we welcome good 10 recommendatior.a from anywhere", but when we have to 11 respond on things that we already told the people when 12 they were here, and we don't see the direct relationship 13 to health and safety, why do we have to spend all this 14 time answering these' things, then the NRC' reviewing them ,
15 and saying, well, closed out.
16 So, my question, what do we do in the diagnostic 17 evaluation, to make sure that we're on target with our 4
18 mission?
19 MR. JORDAN: Okay. The example you gave as an 20 illustrative example would bother me a great deal because 21 that's exactly wh a+- we don't want to do. -We don't 22 encumber the utillcy with nits and trivia. And, so, the 23 whole purpose of the diagnostic is to get an understanding 24 of what the actual- safety performance at a particular- >
25 plant is, and do it independently of.the normal program.
I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISLAND AVENUE, N W.
(202) 234 4433 WASHINGTON, D C. P000$ - (202) 2344433 r==--q'as srm wr-W e f am, e- t-wg re-t---mgeyy- i w v mews- wy:P re- 4=-w N 'V w a pP-W&W wsfaf r' Pgr*P WWe e-' "-W'**W8V-*"94=P** e' *15L TTEF 9 ' +PtM** *@N--*'*T P rvPD19 '*"'
N
- i 46 ,
1 We find out what the region han dono. We find 2 out what NRR has dono. And the team la composed of people 3 that have not previously reviewed that sito, boon directly 4 involved with the regulation of that site, not from tho' 5' region, in fact, and so are there to got an understanding 6 of that plant's status.- ,
7 And the first thing that the team is directed to 8 do is to understand what the utility's plann are, and most 9 of those plants, an you indicato, have a correctivo 10 action, a get-woll plan, already structured and perhaps l, 11 ongoing. And wo don't want to interrupt that, we want.to ,
12 take advantage of it. But wo do want to understand 13 whether or not it's the right plan, whether it doon have +
14 the right elements in it.
15 And, no, we're tryina to got a much bigger 16 picture than that. We want to know fundamentally why 17 things aren't going well or, if they are going.woll,-why 10 they are. And, no, the example would bo. foreign to what
- 19 I-expect. We do uso senior managers from the ataff, the 20 regional division director. level, as team leaders.- Leo 21 Spessard han. led some of those teams. And maybe he's the 22 one that had the detailed findings.
23 MR.-SPESSARD: I doubt.it. ,
24 COMMISSIONER REMICK: Are you saying that there 25 are not findings and recommendations that can relate back NEAL R. GROSS COOHT REPORTEnS AND TRANSCHIDERS
.1323 AHODE ISLAND AVENUE, N W.
-(?02) 2344433 WASHINGTON, D C 2000$ (?O2)2344433-
- 47 1 to our regulatory requirements?
2 MR. JORDAN: Let me get to what may be the 3 problem. Up until, I guess, about two or three years ago, 4 the diagnostics carried forward all of the findings. The 5 individuals wrote everything they found, and then in the 6 summary and in the issues, only those higher level issues 7 were then conveyed to the utility.
O What we didn't realize was that the utility was 9 then individually keeping book on those lower level 10 findings as well, and the region was in their follow-up.
11 And, so, we had burdened both the region and the utility 12 with nits that didn't really affect safety.
13 COMMISSIONER REMICK: That was another point-14 I've heard, it needs to track all these things which were 15 uncovered.
16 MR. SPESSARD: We stopped that.
17 MR. JORDAN: Right. We stopped that.
-18 COMMISSIONER REMICK: How recently?
19 MR. SPESSARD: Palo Verde was the last large 20 size. report that was issued, and from Zion forward we've 21 had a shorter version.
22 MR. TAYLOR . Rightfully so, I think the licensee 23 told us about it, and we tried to fix that. That was a 24 useless exercise'in some of the tr'ivia that came out in 25 the report.
NEAL R. GROSS COO 9T REPOATERS AND TRANSCR$ERS.
1323 RHODE ISLAND AVENVE, N.W,
)
I 48 ,
1 MR. SPESSARD: Can I-just add something?
2 MR. JORDAN: Yes. We thought we had burdened 3 the region with follow-up that we didn't intend even.
i 4 MR. SPESSARD I'd like to add that every team 5 that goes out is trained before they go, what the 6 expectations are of that team. They get a lecture from 7 me, and Ed Jordan as well. We emphasize strengths and 8 weaknesses when we see them. If the licensee has programs. -;
9 in place, the instructions given is, you'll assess that 10 program and give them credit where credit is due. It's-11 intended to be an objective evaluation -- in other words, 12 it's balanced. It's not skewed one way or the other. But 13' you report what you find and, if it's ' mostly negative, 14 then the report is going to be negative. So, we have mado ,
15 some adjustments. But I will be honest, our evaluation is 16 intended to be performance-based. It's intended-to look 17 at performance as it relates to safety, not.necessarily j 18 regulations, and that has some licensees upset, .but we're 19 strictly interested in performance as it relates to 20 safety.
21 COMMISSIONER REMICK - Is it casler to relate it.
22 to " safety" than it is regulatory requirements because, of 23 course, almost anything one does in a plant, one.could say 24 somehow affects safety, but how significant is it?
25 MR. SPESSARD: The difference --
' NEAL R. GROSS COURT REPORTERS AND TRANSCRIDERS 1323 F1HODE ISLAND AVENUE. N W.
(202) 234 4433 WASHINGTON. D.C. 2000$ (202) 2344433 rr- y y i g.wy -79+-+m 7 --'-'-'TT N""
49 1 HR. JORDAN . I can respond and say what we were 2 trying to do was to avoid making diagnostic a civil 3 penalty hunt --
4 MR. SPESSARD: Or to read like an inspection 5 report.
6 COMMISSIONER REMICK: Yes, I understand.
7 MR. JORDAN: So that it doesn't have the list of 8 " contrary to's" and that kind of a regulatory form.
9 COMMISSIONER REMICK: That makes sense. Yeah.
10 Well, I urge that you look very closely at those. It !
11 appears that you are reducing come of that trivia that-12 I've heard comments about, and the extensive effort it 13 takes then to close out things that should not have been 14 necessary to close out, that they were basically closed 15 out when the discussion --
I realize there could be 16 differences there, but I'd urge.you to continue to look 17 very closely at those.
18 MR. JORDAN Thank you.
19 MR. SPESSARD: If you used a real example, I'm 20 totally unfamiliar with it.
21 COMMISSIONER REMICK: No, it was not a real 22 example.
23 MR. TAYLOR: But there was some of them that we 24 had that problem.-
25 MR. JORDAN: We had that problem, I agree.
NEAL R. GROSS COURT RCPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W.
(202) 2344433 WASHINGTON. D.C. 20005 (202) 2344433
50 ,
1 I think the last thing I would like to talk ,
2 about is the follow-up of recommendations, and could I -i 3 have slide 22, please. (Slide) 4 And I'll use that as a typical of what we do for 5 both incident investigations diagnostics and case studies.
6 Where we compile and maintain a status of the 7 recommendations, and these are recommendations that were; 8 transmitted from the EDO's office to the Program Office 9 or, in the case of case studies, recommendations I made to 10 the Program Office, that we track and verify on an annual.
11 basis that we maintain the status.
12 Thus far, there have been something like 32 case-13 studies conducted, that had 120 recommendations, and there ,
14 are of the order of 15 remaining. Some of these have been 15 partially closed out since these numbers were generated.
16 And, so, once again, from my perspective, we 17 haven't accomplished anything until-we have conveyed ~ a 18 lesson and, in fact, the-lesson has taken. And I think -
19 one of the things we've learned over the last few years is 20 that it is becoming more important for us to look harder 21 at how well those lessons have taken because some of them 22 are a fraction, that some utilities did it,.some didn't, 23 some utilities aid'better than others and, of course, even 24 within the NRC, we ourselves learn lessons that don't-25 stick, and we-have to just. keep working at them.
NEAL R. GROSS-
~ COURT REPORTFR$ AND TRANSCRIBERS 1323 RHODE ISLAND AVENVE, N W,- -
~
- (202) 2344433 WASHINGTON. D C. 20005 . (202) 2344 433
, . ~ . , - , _ . . - , . _ . , , - . . _ __ __ _ - ._ _
4
,_- 51 1 Could I have the last slide, please. '(Slido) 2 In summary, our role is to provide the 3 independent review and assessment of licensee safoty 4 performance and, in doing so, we expect to reduce the 5 likelihood that a safety lesson will remain unlearned, or 6 that a safety lesson would become unraveled.
7 We provide some redundancy, diversity, and a 0-A 8 role -- and I left off of my summary slide -- we provide ,
9 appropriate feedback of experience both to the agency and 10 to the industry.
11 Our reviews complement the assessments done by 12 the Program Offices. We try not to duplicate -- I mean, 13 if the Program office is doing a particular activity, we 14 don't replicate that activity. We are responsible for 15 closing out our recommendations and, once again, .to hiake 16 the last point, we have to continually review the 17 corrective actions, to assure that the lessons have a 18 permanence or, if the lesson is changed slightly, that it 19 be re-established. That's all.I had prepared.
20 CHAIRMAN SELIN: Mr. Jordan, as you well know, ,
21 I'm very pleased in general with the operation of the 22 office, and even more-so with the concept that having a 23 group that's relatively independent, that can look at the-24 empirical results without being so wrapped up with the 25- arguments about why'we did this or that in the ; first NEAL R. GROSS COURT FIEPORTER$ AND TFANSCHSERS 1323 RHODE ISLAND AVENUE. N W.
(202 2344433 WASHINGTON, O C 20005 ~(202) 2344433 t-l
4 52 .
I place, is invaluable. And I'm struck, even though 2 occasionally wo do need the help of outuidor;; tc point mit 3 questions such as the thermal-add question, or the boiling ,
l 4 water reactor level indicator quoution that perhaps hasn't ,
l 5 gotton as much attention within the agency, nevertholoss, 6 it is impressive to me, given how complex reactors are and 7 just how much engineering there is, that so many of thoso 8 problems are found by our own efforts and, in large part, i
9 by your offico's efforts.
10 I'm equally impressed, but in a negativo nonno ,
11 rather than positivo senso, on the material sido, such as 12 the medical sido, how of ten ovidence shows up that wo just 13 don't know ourselvos, that comes -- and, you know, a 14 couple of reporters, in a few wooks, can find dozens of 15 canos of incidents that we really are not aware of, as an 16 institution, ourselves. And - bearing in mind that the 17 relativo effort in reactors versus, say, materials in 18 quito difforent, I think that's appropriato.
19 Do you see a potential role for--- I don't want 20 to call it the " incident investigation", or .the 21 " independent investigation", it's not that formal -- but 22 a rolo.that you think your organization could play in 23 providing - a certain amount of evaluation so _ that the .
24 Commission isn't so dependent upon the regulators, to tell 25 us how they are doing, - and would- have an indopondent NEAL R. GROSS COURT RLPORTERS AND 1RANSCFiiDinS 1323 RHODE ISLAND AVENUE, N W.
(202) 2344433 WACHIN010N. O C 20005 (20?) 234 4433
l 53 1 source of empirical information on some of the material 2 side? In this case, I'm obviously very much concerned 3 about the medical side, but I wouldn't limit it to that 4 question. This is the speculative question I've been 5 threatening you with for an hour --
6 MR. JORDAN Yes, sir, I was afraid --
7 CilAIRMAN SELIN: -- in case you hadn't noticed.
8 MR. JORDAN: You're correct, we do a-very small 9 offort in this area. We spend the order of two dedicated 10 individuals --
11 CllAIRMAN SELIN: Out of how many?
12 MR. JORDAN: Out of 116, sir. We, in f act, havo 13 just made a recent selection, so that those two people,-
14 one will look at the medical side and the other at the 15 nonmedical side of the entire materials program. And in 16 that way, they can only look at the tip of the iceberg.
17 They can only survey some of the statistical information, 18 and cannot provide true independent review.
19 We are managing for Jim Taylor, the incident 20 investigation into the oncology services incident, : and --
21 CflAIRMAN SELIN: Is this the first medical IIT 22 that we've done?
23 MR. JORDAN:- It's the first medical. We did a 24 materials, the Amersham lost source event'as-an incident 25 investigation, and each of those.has identified a large NEAL R. GROSS
= COURT REPORTfR3 AND TRANSCRIBERS '
1323 RHOCE ISLAND AVENUE, N W.
(202) 234 4433 WASHINGTON, D C. 20005 . (202) 2344433
. . . . . . . _ _ _ . . _ m .
54 ,
l' number of lessons. So, I am uncomfortable. I think I 2 would 'have to ha;re a discussion with Jim Taylor, and come--
L 3 back to you as to what we would want to propose.
4 CHAIRMAN SELIN: When you have the discussion, 5 one other thing I'd like you to think about is -- again, -
6 I realize that performance indicators are based on a huge 7 basis of corporate experience that we have. We have a lot 8 of data, we have a lot of people who know a lot of things.
9 You have SALP scores, you have all kinds of things you can 10 refer to and say it is plausible. But it seems to me 11 that, particularly in the medical- area, that we sorely 12 need some kind of overall performance indicators to really .
13 say, you know, how's the industry doing? H7w are.we 14 doing? Are we spending all this-time worrying about the 15 one. event in a million that is bound to- slip through where 16 humans are practicing medicine, or do we have a serious 17 problem that we've been sort of overlooking in'a lot of-
'18 quality assurance data?
19 So, it's not just. whether we can do the 20 specifics, which I think-is the more important of the two, 21- but whether we can have some way.of measuring relatively 22 objectively, - how ~ are we doing? How are our programs 23 doing? How are the people we regulate doing? Are:they 24 getting better? Are they - getting worse?. Is theorisk 25 involving nuclear medicine comparable with the risks -that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N.W.
~
(202) 234 4433 WASHINGTON. D.C. 20005 (202) 2344433
55 1 are involved in other. ways of practicing medicine at such 2 a high level? Is there -- particularly, I'm interested in 3 is there a big disparity with what happens in the devices 4 we don't license, like accelerators and x-rays, compared 5 to the ones that we do license? Are people avoiding using 6 medical byproducts just because . we cause them more 7 trouble, good or bad, than going to electrical sources of 8 radiation? There are all kinds of questions that are 9 really evaluation questions, and I know it's a lot to ask, 10 but I wonder if there's a role for AEOD and the kind of 11 evaluations you do, at least getting a little bit of a 12 start on these very troublesome questions.
13 MR. TAYLOR: Mr. Chairman, I'd like to come back 14 to you. There are a couple of sets of information, those 15 are in states in which we have jurisdiction and those in 16 the agreement states, where --
the division of the 29 -
17 agreement states where they oversee the-medical aspects.
18 So, there are several different sets of data and 19 information, but to your broader question, I think we need 20 to -- Jordan and I need a little time to work through that 21 and see how much effort we're able to do within the 22 current resources, but the absence of the information is 23 very evident.
24 CHAIRMAN SELIN: It may be quite premature, but i
25 this approach of looking at the data and not getting NEAL R. GROSS COURT REPORTERS AND TRANSO31BERS 1323 RHODE ISLAND AVENUE, N W (202) 234 4433 WASHINGTON. D.C. 20005 (2025 234 4433
. _ . - . -- -~ -. - _ . - . . .
56 .
1 involved in what was.It we were trying to do'in all the 2 defense of this stuff is so powerful, and-you've really 3 done it very well.
4 Commissioner de' Planque, you haven't- had . a 5 chance --
6 COMMISSIONER de PLANQUE: Just one more comment 7 along those lines, it always strikes me as interesting 8 that in the report we send to Congress, there's a: fair t
9 number of nonreactor incidents that-go in there. And if 10- you look at the balance in that report, reactor-versus 11 nonreactor, then you have to wonder about is the balance 12 here consistent with that?
13 MR. JORDAN: Yes.
14 MR. TAYLOR: That's right.
15 COMMISSIONER de PLANQUE: And in view of the 16 fact that it's not, what's wrong? Which one is-out-_of ,
17 balance?
18 MR. TAYLOR: Reasonable question, it's been 19 asked a number of times through the years. It's a 20 reasonable question.
21 COMMISSIONER de PLANQUE: Yes. -A couple of
~
22- questions. We haven't touched at'all on what goes on in-23 a similar way in the international arena. Are we getting 24 events reports from the international community, and how 25 helpful are they in your efforts?
NEAL R. GROSS COURT REPORTERS AND TRANSCHmERS -
1323 RHODE ISLANJ AVENUE, N W (202) 234 4433 WASHINGTON. D C 20005 (702) 2344433
. - _ ~ ,. ,_ ,,. _
,;- 57 1 - HR. JORDAN: Yes, and, Jack, why don't . you 2 answer that with the IRS. i 3 MR. ROSENTHAL: Well, we formally reviewed the 4 International Reporting System reports, and incorporate 5 them into our own products. We have a lot of data 6 ourselves, so you have to question how useful they are. .
7 But we are also involved in rare event analysis. An 8 example, on the slide there was power oscillation at a 9 boiling water reactor. And at the time, we knew of only 10 one in the United States, but there were two other events 11 that had been reported internationally, and that was a 12 deciding factor in whether to go forward or not, and wo 13 had written it up in the report.
14 So, although it's f ar less data, it's usually at 15 a higher threshold than the 2,000 LERs a year that we get.
16 So, we do find i_ a valuable source.
17 COMMISSIONER de PLANQUE: Okay. -Also, in the 18- international arena, do regulatory bodies in other ,
19 countries have a group similar -to yours and, if so, 20- anything to learn from that?
21 MR. ROSENTHAL: Yes, but- different. My 22 equivalent in Germany is GRS, who really.are not feds.
23 France has a very similar-group.
24 COMMISSIONER de PLANQUE: Is there much l 25 interaction?
! NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W, (202) 234M3 WASHINGTON, D C. 20005 (202) 2344433
58 .
1 MR. ROSENTEAL: Yes. Yes.
2 COMMISSIONER de PLANQUE: Okay. Based on'your 3 independent experience, would you like to offer some-4 comments on SALP, whether you think-there should be any 5 changes? '
6 MR. JORDAN: I think SALP is a very valuable, 7 beneficial tool, and the early correlations we made in 8 validating performance indicators -reinforced that view.-
9 So, I'm a supporter of SALP.
10 COMMISSIONER de PLANQUE: As it's curret ,1y 11 structured?
12 MR. JORDAN: Yes.
13 COMMISSIONER de PLANQUE: -Okay. In. the 14 diagnostic evaluation areas, I-had the impression that 15 this of ten confirms the notion of poor performance. There 16 are exceptions, of course. Is there any-way.to better use 17 this -- and you were touching upon this, .too -- in more of 18 a predictive mode?
19 MR. JORDAN: It's expensive. We spend two-to 20 two and a half full-term equivalents in -doing a 21 diagnostic. And we did one at a plant thatLwas not a 22 serious question, as a calibratior early on, and we have 23 gotten some good news from-diagnostics.
24 COMMISSIONER de PLANQUE: Yes, I know.
25 MR. TAYLOR: We've found it's better-than we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 234-4433 WASHINGTON, D C 20005 (202) 234-4433
m_. . - __ _ -
.- 5'9 1 really might have thought, when we've done them, ~ which is 2 also good.- That is, we came out with --
3 CHAIRMAN SELIN: If I might just break in on --
4 COMMISSIONER de PLANQUE: Go ahead.
5 CHAIRMAN SELIN: What I would hope you would do, 6 when you do this overall regulatory review that you've 7 proposed, that you take a look at how-the resources that 8 we apply to plants is correlated with -the perceived 9 performance of the plants -- obviously, the DETs as well 10 as the team inspections -- would be very high up on the 11 list of where do these go. I mean, the idea of doing a 12 two-stage process where you first sort of look for what a:
13 lawyer would call -" probable cause", and then put the 14 resources in where there is rather than just randomly, is '
15 clearly an interesting prospect to look at, and I would 16 ' hope you would include the DETs as well ' = as the team 17 inspections as being the very large discretionary 18 activities that we do.
19 COMMISSIONER de PLANQUE: That's all I-have.
20 Thank you.
21' CHAIRMAN SELIN: Commissioner Roberts?
22 COMMISSIONER ROGERS: Yeah.. Well,-just while 23 we're on the performance indicator- question, and somewhat 24 along the lines of Commissioner de Planque's question, a
'25 couple of years ago, you were attempting to develop a-NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 2344433 WASHINGTON. D C 2000$ (202) 234-4433
60 ,
1 leading indicator, and I know that that looked promising 2 -for a while, and there were 'some enthusiastic reports to
-3 u s ', and then it just sort of petered away and we haven't 4 heard anything more about it. And have you ' totally 5 abandoned that approach, or that attempt to find a leading-6 indicator, something that tells you in advance that the 7 plant is going to get into some trouble?
8 MR. JORDAN: Yes. I give up.
9 COMMISSIONER' ROGERS: Give up.
10 (Laughter.)
11 CHAIRMAN SELIN: As Yogi'Berra likes to say, 12 "It's hard to make predictions, especially of the future".
13 COMMISSIONER ROGERS: Right.
14 MR. JORDAN: No. Being maybe a little more 15 direct --
16 COMMISSIONER ROGERS: I think that was very-17 direct. ,
18 (Laughter.)
19 MR. JORDAN: Well, it's not --
20 COMMISSIONER ROGERS: It couldn't be much more.
21 _ MR. - JORDAN : We .certainly see the trends in
-22 various information, as being perhaps predictive that 23 things are - getting worse, but in finding some unique i
.24 feature of the plant that'we can measure that predicts 25 plants are going to get worse, I don't -- I give up.
NEAL R. GROSS l COURT REPORTER $ AND TRANSCRIBERS ~
1323 RHODE ISt.AND AVENUE, N W.
(202) 234 4433 WASHINGTON, O C. 20005 (202) 2344433
61-1 COMMISSIONER ROGERS: Well, it seems as if it's-2 a quest for something that's very difficult to find.
3 MR. JORDAN Yes.
4 COMMISSIONER ROGERS: Yes. I noticed in your-5 discussion you mentioned that you have a daily interaction 6 with NRR. No mention of any periodic interaction with 7 NMSS, certainly not daily, maybe not weekly, maybe not 8 monthly, but I suspect that seems to be-on an ad - hoc 9 basis, when you get to NMSS. And my observation relates 10 somewhat to this question of how we're following - 'the 11 incidents in the medical area -- we're hearing so.much 12 about that right now, just currently -- and whether we're 13 following up medical misadministrations in the same way we 14 would follow up some kind of a prob 1'em that's been .
15 discovered in a nuclear power plant.
16 MR. JORDAN: Right. Resource-wise, we clearly 17- are not. We do have a coordination meeting with NMSS on 18 occasion, not as often as a_ periodic one.with NRR. When-19 there is an event call related to materials,~we do share 20 and participate in the-call with NMSS. The frequency of
-21 their events that make a telephone notification-is-much, 22 much lower than reactors, so they are. not a'part of a 23 morning call. They do receive, of course, the _ daily
- 24 report information, but I think a good point.has been 25 raised. They-are treated differently. The nonreactor and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.
(202) 234 4433 ' WASHINGTON, D.C. 2tXe (202) 234-4433 1
62 ,
1 reactor areas are getting a different ratio of effort.
2 Now, I'd like to correct maybe a-number I gave 3 you of 2 to 116 as being more like 2 to the order of 50, 4 since another part of our office- is training and 5 operations center, and so that.wasn't -really a fair ratio, 6 but it's still obviously a large difference.
7 COMMISSIONER ROGERS: Well, it just'seems to me:
'l that maybe the general' question of how we're following 9 through longitudinally on- the areas of ' medical'and 10 materials problems maybe is something that-merits more 11 attention.
12 CHAIRMAN SELIN: Anything,-Commissioner?
13 COMMISSIONER de PLANQUE:- No.
14 CHAIRMAN SELIN: Commissioner Remick?
15 COMMISSIONER REMICK: No further questions, but 16 I would say that I think lt was a good decision some years 17 ago to set AEOD up in a' semi-independent role. I 18 personally take a lot of ccmfort from.the fact-_that'you 19 are out there.doing the job that you are, and it's.been a
- 20 pleasure for me to observe the-last-_ decade, a:little bit-21 more, what..I see not only as an improvement, but a-better 22 job all around, really knowing now what you are about and 23 so forth. So, I think the AEOD has come a long way from 24 those very early days when it was trying-to -find -itself in 25 the organization, and - I think you're making a real l NEAL R. GROSS i
COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVENUE, N W.
l G02) 2344433 WASHINGTON. O.C. 20005 602) 234 4433
. .. = - - .
-,_ 63:
-I contribution. .
.2 MR. JORDAN: Thank you.
3 CHAIRMAN- SELIN: I like your Dragnet phrase,-
4 "Just the facts, man, just the facts".- I mean, the fact 5 that you're so far from the policy and you're'really just-6 looking at what's really happening _is a great source of 7 comfort. Thank you very much. _
8 MR. JORDAN: We'll be pleased when we._are:next 9 door to you.
10 CHAIRMAN SELIN: Yes.
11 (Whereupon, at 11:30 a.m., the -meetina was -
12 adjourned.)
13 14 15
~
16 17-
"- 18 19 20 21 22-23 24 25
' NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHOOE ISLAND AVENUE, N.W.
(202) 2344433 WASHINGTON D.C. 20005 - (202) 2344433 l l
e CERTIFICATE OF TRANSCRIBER This is to certify that-the attached events of a meeting of the United States Nuclear Regulatory Commission entitled:
TITLE OF MEETING: BRIEFING ON ROLE OF AEOD IN OVERSIGHT OF OPERATING REACTORS PLACE OF MEETING: ROCKVILLE, MARYLAND DATE OF MEETING: DECEMBER 17, 1992 were transcribed by me. I further certify that said transcription is accurate and complete, to the best of my ability, and that the transcript is a true and accurate record of the foregoing events.
G Mata ~
v (f Reporter's name: PHYLLIS YOUNG NEAL R. GROSS COURT REPORTERS AND TRANSCRilIR$
1323 RHOOf 15LAMO AYtNUf, N.W.
WASHINGTON, D.C. 20005 (202) 232 4600 (202) 234 4433
y lNDEPENDENT ROLE OF AEOD COMMISSION PRESENTATION DECEMBER 17,.1992 Edward L. Jordan
- Contacts: . . . .
' Thomas' M. Novak, 492-4484
. R. Lee Spessard,:492-8577
+
~-
i-
,., - - y v- y, . . . .
, . - , . -. 4 - . , y ,..r..
L ;
OUTLINE Operational Experience Assessment Feedback of Experience I I
incident investigation Diagnostic Evaluation Followup of Recommendations Summary 2
6
OPERATIONAL EXPERIENCE ASSESSMENT
- Events Analysis
- Accident Sequence Precursors
- Human Performance
- Trends and Patterns
- Performance Indicators
- Abnormal Occurrences 3
EVENTS ANALYSIS
- Event Report Screening
- Daily Events Conference Call
- Classify LER Significance e initiate Engineering Study
- Initiate Human Performance Study
- Perform ASP Analysis 4
- ACCIDENT SEQUENCE PRECURSOR PROGRAM
- Systematic Risk-Based Evaluation
- . Estimates Conditional' Core Damage Probability.
- Discovers Plant Vulnerabilities Failures!Not Always in PRAs
~
e- -
-* Identifies important Events
- s.
-N-~_..
N.,_ .mau - - . a lY. a ..-n.---u - .- -_-.a
n
' HUMAN PERFORMANCE STUDIES
-i
- e- ' Case Study AEOD/C92-01, " Human Performance in Operating Events"
-e- Information Provided.- ,
e- . Control Room Staffing',fOrganization
- e. " Dual Role" Shift Technical Advisor Shift Resources, Crew Teamwork ;
e Task: Awareness, Use Of-Procedures .
.+- Human-Machine Interface i
6 k
9 i
$ .'4-3 m , ,- 446.4 - , .- # a , ,s.. -4 .,-,. . --..~..,- :,-<r.--. .
TRENDS AND PATTERNS
- Analysis of Operational Events & Conditions
- Development of Tools and Procedures
- Develop Trends for
- Hardware-Related items
- Human Performance
- Regulatory & Safety issues
- Industry initiatives
- - Assess Risk Significance of Trends 7
].
l:.
^
PERFORMANCE INDICATORS
- One of Several NRC. Management Tools
- Plant Performance ' Analysis e identify Performance Trends
- Identify Performance Strengths / Weaknesses t
e industry. P;erformance Trends j.
+
. "8 i
' b'
. u.
- ., -m.. . , , . ...s ..e c % v. , , , u
u 4-1.
I ABNORMAL OCCURRENCE REPORT l-e identification of Proposed ' Abnormal Occurrences
- - Coordinate with Program' Offices and ' Regions :
- e: Quarterly issuance r Develop Criteria'and Prepare Guidance b
i d '-
l
?
r7
)
i 9
s c ~. . s ' ., ;
,, ....u.-,:....
b.
y
.ii.
1 FEEDBACK OF EXPERIENCE
+
- Air Systems - Generic Letter 88-14'
- BWR Power Oscillation - Bulletin 88-07
- - ! Motor-Operated Valves - Generic Letter 89-10 -
- Solenoid Operated Valves - Generic Letter 91-15 o
- Pressure Locking 'of Gate Valves ,
'10
'.l',
- 6 'e .
'.. -- . - . . < e . i.'.-e ,_ , ,n ,x__._m.,____ _ ,_, , , ,__m_ _
PRESSURE LOCKING OF GATE VALVES
- Safety issue: Potential Common Mode Failure to Open important Valves
- Phenomena: Valve Motor Operator Cannot Overcome High Loads
- Cause: Bonnet Pressurization Locks Gate Valve Closed
\
l 11
3 PRESSURE LOCKING OF GATE VALVES
- AEOD May 1992-Site Survey Results
- Pressure Locking a Credible Problem
- - Plant-Specific Failure Experience Lacking
- Engineering Analyses of Phenomena Not
- Conservative or Complete
-*' Most Licensees Have Not implemented Recommended Valve Modifications 12'
_..i'r .m_[m' .,-m_._, _. ~ ..J..'_..' .u.ml. -- '
PRESSURE LOCKING OF GATE VALVES
- NRC Circular 77-05, 1977
- W Service Letter,1977 o AEOD Pressure Locking Study,1984
- Industry Feedback,1984
- Recent LERs Show Problem Continues to Exist
- NRC Information Notice 92-26,1992
- AEOD Special Study, December 1992 13
i INCIDENTINVESTIGATION PROGRAM l
PURPOSE OF llP j u
- Investigate.-Serious Operational Events
- - Full Understanding of the Issues
- Increase Effectiveness' of NRC Programs .
J i
i 7
-I I
'14. ;
-i
. .t
=w ~
,s., ~
.e g , -e- -w
INCIDENT INVESTIGATION PROGRAM AEOD'S ROLE
- Maintain Documentation and Provide Training
- Evaluate and Recommend Events for llTs
- Propose llT Charter and Members
- Provide llT Administrative Support
- Review Draft ilT Report for Completeness
- Propose NRC Actions and Monitor Status
- Document. Action Closecut in Annual Report 15
INCIDENT INVESTIGATION PROGRAM EXAMPLES OF TECHNICAL LESSONS
- Shutdown Risk Management issues
- AFW Turbine Overspeed Trip Deficiencies
- MOV Switch Setpoint Problems -
- Check Valve Reliability issues
- Fuel Facility Criticality Safety Control issues 16
.a
n :.. . ...
u l: q r- INCIDENTINVESTIGATION PROGRAM j EXAMPLES .0F REGULATORY LESSONS L -* : MOV Testing and Surveillance Requiremeots Event Classification and Reporting Requirements for Loss of:CriticalityLSafety Controls .
- -Shutdown Risk Management Requirements ,
- Emergency Plan. Initiating Event Guidance
- NRC IN/ Bulletin ContentLGuidance 17 e-
-,:-,.,,.....,,g
. , . - - -- ., . . . . ... . . . .i, .. ,- .., . .. . ..i..
. ..,...,i- _,....., ,,..- 5,,., ,.i i. . ...
_.,a.,1r
...c. .... . . . . .....r... .. .. ..g.s,_j, ,, -y
,-...,,,.y,- .. .
DIAGNOSTIC EVALUATION PROGRAM PURPOSE OF DEP
- Independent Assessment of Performance
- Augment SALP and Pi information
- Determine Root-Causes for Problems 18
DIAGNOSTIC EVALUATIONS AEOD'S ROLE
- Maintain Program Documents
- Evaluate Performance and Recommend des
- Provide DET Core Members
- Prepare DE Plan and Provide Admin Support
- Provide Oversight to DET Manager
- Propose NRC Actions and Monitor Status
- Generic Action Closecut in Annual Report 19
DIAGNOSTIC EVALUATION PROGRAM COMPLETED DET'S Dresden - Nov 1987 Arkansas - Dec 1989 McGuire - Mar 1988 Palo Verde - Mar 1990 Turkey Point - Jun 1988 Zion - Sep 1990 Fermi'- Nov 1988 Oyster Creek - Feb 1991 Perry - May 1989 FitzPatrick - Dec 1991 Brunswick - Jul 1989 20
, - , . , , , - .n._ , ~-.. . . -. u_ . _. __-, . , , ._m
DIAGNOSTIC EVALUATIONS EXAMPLES OF DET LESSONS
- Senior Manager Perspectives on Performance Changed
- Service Water and MOV Deficiencies identified
- Licensee improvement Plans Revised
- NRC Staff Reviews of IST Programs Accelerated 21
FOLLOWUP OF RECOMMENDATIONS i
AEOD CASE STUDIES STUDY YEAR . REMAINING
! 1980 Vital Instrument Bus Tie 1 l Breaker 1981 St. Lucie Natural Circulation 1 ,
Cooldown 1985 Pressurized Gas Storage 1 1985 ECCS Overpressurization 4 1987 Air Systems 1 1988 BWR Power Oscillation 1 1990 SOV Problems 6 i
4 22
. e FOLLOWUP OF RECOMMENDATIONS INCIDENT INVESTIGATIONS YEAR llT REMAINING 1985 Davis-Besse O San Onofre 1 Rancho Seco 0 1990 Amersham 2 Vogtle 6 1991 GE Fuels 15 Nine Mile Point 12 1992 Oncology Services -
36 23
FOLLOWUP OF RECOMMENDATIONS DIAGNOSTIC EVALUATIONS Year DET REMAINING 1990 Zion 1 1991 Oyster Creek 1 1991 FitzPatrick 2 24 O b
4
SUMMARY
OF AEOD ROLE
- Independently Review and Assess Licensee's Safety Performance
- Reduce the Likelihood that a Safety Lesson Will Remain Unlearned
- Reviews Complement Assessments Conducted by Other Program Offices
- Formally Closeout AEOD Recommendations
- Evaluate Adequacy of NRC/ industry Corrective Actions 25 t
i