ML20148G887
| ML20148G887 | |
| Person / Time | |
|---|---|
| Issue date: | 06/03/1997 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-3001, NUDOCS 9706060271 | |
| Download: ML20148G887 (139) | |
Text
. _ _ . . - . _ __ _- .
. Official Transcript cLf Proceedings. q 3
O NUCLEAR REGULATORY COMMISSION
/ CRS7~- 3co/
i l'
Title:
' Human Factors Subcommittee !
c i
TRO4 (Acg3) l RETURN ORIGIngt T
Docket Number: (not applicable) sawuIra i M/S T-2E26 i 415-7130 i TuANKS1 ~.
.l t
Location: Rockville, Maryland a D l f
Date: Tuesday, June 3,1997 - _
D l l
l .!
i h
9706060271 970603 )
T-3OO1 PDR j
, f y- e ;
Work Order No.: NRC-1130 Pages 1-125 i i
((ll{l!!!ilkk!!!!,IIIk!fk!! .
. i i
NEAL R. GROSS AND CO., INC. ;
Court Reporters and Transcribers ;.,'; _,4 , )
0G005,3 13B Rhode Island. Avenue p.W. , p 3 i
53 -
- 0Y%a m fg i O_it.le
- s u n30K41 a m ut..
1, ,
Official Transcript cf Proceedings I! O NUCLEAR REGULATORY COMMISSION
/ CRST~ 3o0/
Title:
Advisory Committee on Reactor Safeguards Human Factors Subcommittee -
TRo4 (ACRS)
TRETURN ORIGINAL Docket Number: (not applicable) 88HITE )
M/S T-2E26 415-7130 THANKSr j
{
1 Location: Rockville, Maryland i i
o !
Date: Tuesday, June 3,1997 I
,U l
I l
1 9706060271 97o603
$9o0i 3 PDR Work Order No.: NRC-1130 Pages 1-125
)
[lEllllIll,$l!Il!OIIE!!I
]
NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1
OG0055 on Rimac islandmenue p.g.! , K")'f i
h A, fJ. b% vi tfMiasiain[ogjhyf,30 icamuwu- a m. bg l ;f t (5g,'j.
g,
[0 $ 6 3 0 Of. th e 3 0 5..,,7
. . . . , . . . . . nt t!m@ l J
l 1
O !
F DISCLAIMER j PUBLIC NOTICE -l BY THE l UNITED STATES NUCLEAR REGULATORY COMMISSION'S l l ADVISORY COMMITTEE ON REACTOR SAFEGUARDS !
JUNE 3, 1997 ,
1 i
t l
i
- j. The contents of this transcript of the i t :
t j proceedings of the United States Nuclear Regulatory l r
i Commission's Advisory Committee on Reactor Safeguards on JUNE i 0
\
l 3, 1997, as reported herein, is a record of the discussions i recorded at the meeting held on the above date. ;
l This transcript has not been reviewed, corrected j i
and edited and it may contain inaccuracies. j i
l 1
4 i
O 1 1
NEAL R. GROSS COURT REPORTERS AND TRANSCR!llERS 1323 RilODE ISLAND AVENUE, NW (202)234-443 6 WASil!NOTON, D C. 20005 (202)2344433 I
1 1 UNITED STATES OF AMERICA f ~s 2 NUCLEAR REGULATORY COMMISSION 3 + ++ + +
4 MEETING 5 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS (ACRS) 6 HUMAN FACTORS SUBCOMMITTEE 7 + + ++ +
8 TUESDAY 9 JUNE 3, 1997 10 + ++ + +
11 ROCKVILLE, MARYLAND 12 13 The Subcommittee met at the Nuclear Regulatory
,ew i I V' 14 Commission, Two White Flint North, Room T2B3, 11545 1
15 Rockville Pike, at 12:00 p.m., George E. Apostolakis, l 16 Chairman, presiding.
17 l
l 18 COMMITTEE MENBERS: l 19 GEORGE E. APOSTOLAKIS CHAIRMAN 20 JOHN J. BARTON MEMBER i 1
21 DANA A. PO'WERS MEMBER 22 ROBERT L. SEALE MEMBER 23 24
() 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344433
2 1 ACRS STAFF PRESENT:
I p 2 MICHELE KELTON Tech. Secretary
.] 3 SAM DURAISWAMY 4 CAROL A. HARRIS 5 RICHARD P. SAVIO i I
6 NOEL DUDLEY 7 MEDHAT M. EL-ZEFLAWY ;
I 8 MICHAEL MARKLEY !
l l
9 AMARJIT SINGH i
10 l l
11 ALSO PRESENT:
12 JAY PERSENSKY 1
13 MARK CUNNINGHAM Ch
14 ANNE RAMEY-SMITH 15 WAYNE HODGES 16 CECIL THOMAS 17 JACK ROSENTHAL 18 ERNIE ROSSI 19 DENNIS SURIG 20 21 22 l 23 24
/~'N N,) 25 I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
1 (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
3 1 A-G-E-N-D-A
,s- 2 Acenda Item Pace l
/ ; -
3 Opening Remarks, Chairman Apostolakis 4 4 Introductory Remarks and Status of 5 RES Human Performance and I
6 Reliability Implementation Plan I 1
7 Wayne Hodges 5 8 Jay Persensky 9 9 Discussion of Plan 11 !
l 10 Committee Discussion 115 11 12 13 g3
'v)
+
14 i
15 !
16 17 18 19 I 20 l
l 21 l
22 23 24
( 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
4 1 P-R-0-C-E-E-D-I-N-G-S
,r 3 2 (12:09 p.m.)
i 1
'w)' 3 CHAIRMAN APOSTOLAKIS: The meeting will now 4 come to order. This is a meeting of the ACRS Subcommittee 5 on Human Factors.
6 I am George Apostolakis, Chairman of the 7 Subcommittee. ACRS Members in attendance are: John 8 Barton, Robert Seale, and Dana Powers -- who will join us ;
1 9 in a minute. I l
l 10 The purpose of this meeting is to hold I 11 discussions with representatives of the NRC staff l l
12 regarding the status of the Human Performance and l l
13 Reliability Implementation Plan. The Subcommittee will
,3
\ !
'd 14 gather information, analyze relevant issues and facts, and 15 formulate proposed positions and actions as appropriate, 16 for deliberation by the full Committee.
17 Noel Dudley is the Cognizant ACRS Staff 18 Engineer for this meeting.
19 The rules for participation in today's meeting 20 have been announced as part of the notice of this meeting 21 previously published in the Federal Register on May 9, 22 1997.
l l
23 A transcript of the meeting is being kept and 24 will be made available as stated in the Federal Register
/3
() 25 Notice. It is requested that the speakers first identify NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
1323 RHODE ISLAND AVE., N W.
! (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
5 1 themselves and speak with sufficient clarity and volume so x 2 that they can be readily heard.
(
)
3 We have received no written comments or 4 requests for time to make oral statements from members of 5 the public.
6 In response to comments by the ACRS concerning l
7 the Human Performance Program Plan, the staff has been 8 developing a comprehensive program plan. The Office of 9 Nuclear Regulatory Research has been leading an agency- ;
10 wide effort to develop a human reliability assessment and 11 human performance evaluation program plan which will serve 12 the Agency's future needs fcr confirmatory research and 13 developmental projects.
- ) 14 We will proceed with the meeting and I call 15 upon Dr. Jay Persensky of the Office of Nuclear Regulatory 16 Research to begin, 17 DR. PERSENSKY: Actually, what I'll do is I'll 18 turn it right over to Wayne first; he's going to do the 19 introduction.
20 CHAIRMAN APOSTOLAKIS: Okay, Wayne.
21 MR. HODGES: This is Wayne Hodges with the 22 Office of Nuclear Regulatory Research. I'll make a few 23 opening comments and then we'll turn it over to Jay.
24 Both in response to the ACRS letter, some (s ) 25 comments we had had from NSRRC, and from our own perceived NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
6 )
1 1 need to develop a coherent plan, we are in the process of 7s 2 developing a plan that we hope to be an agency-wide plan
( ) :
3 to deal with the human factor, human liability issues. l 4 What we have at this point is a work-in- j 5 progress. If you characterize the old human performance 6 plan that you have seen before as a catalogue of tasks, j l
l 7 probably what we are at now is at a catalogue with a 8 preface. So we're making some progress; we've got a ways 9 to go yet; and we want to get input from you on the type 10 of things you think need to be here.
l 1
21 What you'll from Jay is kind of an overview, a 12 very high level overview of what we're working on. We've 13 got members of the other offices here to take part in the J
Ns' 14 discussions because this is supposed to be an agency 15 effort, not just a Research effort.
16 Some parts of what the agency is doing are 17 relatively straightforward in that you may go out on X-18 many inspections a month and you can expect to do that 19 from now to you keel over. And so if that doesn't take a 20 fancy plan to do, but you ought to at least allow time for 21 it.
22 Other parts of what we're doing are much more 23 complex and we need input from a lot of different sources.
24 So at this point I'll kind of shut up and let Jay start l /'N q ) 25 the discussion, but the main point I want to emphasize is, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
7 1 that we're really in the early stages of the development (3 2 of this.
\ )
v 3 We have maybe two rounds that we've been i
4 through with the other officers trying to get some 5 comments. We're not at a point yet where we're ready to 6 give it to you but we will, by the end of the month, have )
7 something that we will give to you -- we're committed to 8 do that. That's what our target is, and so we'll kind of i 9 keep it at a high-level discussion as far as today's 10 presentation, if we can.
1 11 CHAIRMAN APOSTOLAKIS: So when is this 12 supposed to be finished?
13 MR. HODGES: June 30th I believe, is the
\ '! 14 commitment date -- to have something --
15 CHAIRMAN APOSTOLAKIS: That's for a draft?
I 16 MR. HODGES: For a draft to you, yes. It will 17 never be finished. It's a plan. We'll have to keep 18 updating it so it will never be finished, but we hope to 19 have something that resembles a final product, by the end 20 of June.
21 DR. PERSENSKY: We expect to have something to 22 the EDO by -- our commitment was to the EDO to have a plan 23 for him, that he would then circulate to the various 24 review committees. So that ACRS would participate (D
( ,/ 25 further, as well as the NSRRC, and who else might get NEAL P, GROSS COURT REPOF % e 3ANSCRIBERS 13235- ":. . AVE., N W.
(202) 234-4433 WA& rih ~ 21 20005-3701 (202) 234-4433
8 l
1 involved we're not sure. But at least those two groups.
i f x, 2 So our current target is June 20th.
%Y 3 CHAIRMAN APOSTOLAKIS: Is the plan going to l
4 change anything? I mean, are you going to make any l 5 decisions after you have the plan?
6 MR. HODGES: Yes. Part of my expectation is 7 that once we have it in place and you can take a look at 8 the whole plan, some of the work we're doing now will 9 probably drop off the plate, and some things that we're 10 not doing will be added. In our budgeting process, in i 11 meeting with the BRG -- last week I believe it was --
12 something that goes by fairly quickly, you have such fun l
13 at it.
(' )
'- 14 In a process with the Budget Review Group last 15 week, we basically told them that some of the money we had 16 in there were placeholders for things we didn't know what 17 we were going to be doing as yet because we're still going 18 to be part of the plan, and they accepted that.
19 CHAIRMAN APOSTOLAKIS: Okay. So even though 20 the plan will never be finished, you expect that maybe by 21 the fall you will start using --
22 MR. HODGES: We expect to be using it, yes, by 23 that time.
24 DR. PERSENSKY: Yes, within the FY98 budget we
(,,/ 25 will have, I think, used it as the basis for a number of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASH!NGTON, D C. 20005-3701 (202) 234-4433
9 1 things, and then when we're starting on the FY99 next g3 2 winter, it will be a much -- take a much greater roll than s_-
3 it has in the past.
4 MR. HODGES: I suppose I should clarify --
5 when I said we went to the Budget Review Group, that was 6 Office of Research in discussing our budget, and I haven't 7 discussed with the other offices how it was handled, but 8 as far as some of our research, we have some placeholders 9 for things that are not on the plate at this point.
10 DR. PERSENSKY: First I'd like to apologize 11 for the time on this thing. This meeting seems to have 12 been a moving target -- but so is the plan to some extent 13 -- and to thank you for your considerations in moving it
'\
(M
\- 14 to this afternoon rather than this morning.
15 I was down in Orlando at the ANS Meeting for 16 my first Human Factors Division Executive Committee 17 Meeting. I felt it was important to get some other 18 stakeholders involved. I think that we've been -- the 19 plan is of course, focused internally, but then knowing 20 what's going on in the rest of the world out there is very 21 important to us.
22 I'd also like to introduce the people at the 1
23 table with me here. Starting on my right, Cecil Thomas l
24 from NRR, the Branch Chief of the Human Factors; then
/ \
(,) 25 Dennis Surig from NMSS; Jack Rosenthal from AEOD; and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE. N W.
(202) 2344433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
10 1 where ar you from today, Mark? Okay, Mark Cunningham is 73 2 from both the PRA Group in NRR and RES; since he's on
( )
'~'
3 detail to NRR we expect him to come back.
4 We also have a number of staff members in the 5 audience who participated in various parts of the 6 development of this plan. Anne Ramey-Smith was the lead 7 in trying to pull this all together, 8 In our discussions over what we might do, I 9 think you and I ran into each other, George, at the --
10 when ACRS was presenting to the commission. We talked 11 about an in-progress review, and that's essentially what 12 this is intended to be.
13 We do not have a final plan or anything ready g)
\_/ 14 to show you at this point. You'll have the slides. It's 15 at a high level to get in the questions of vision and 16 purpose in some of the major programs and give just some 17 examples of how we think we're going to be actually 18 presenting the information.
19 MEMBER POWERS: Are you following any of the 20 structured planning processes that exist in this world, 21 like IEEE, systems engineering, or Air Force system 22 engineering?
23 DR. PERSENSKY: We're primarily using as the l
24 basis for this, the work that was done on the PRA A
() 25 implementation plan.
l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l I
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
11 1 MEMBER POWERS: I'm speaking of planning as an
,S, 2 activity, independent of the specific thing. There are 3 structured planning processes that seem to have these 4 kinds of words in them: vision, mission, then functions, 5 and requirements for those functions, interface 6 requirements, things like that.
7 DR. PERSENSKY: I'll let Anne -- since Anne's 8 eager to answer that; she was taking the lead on that. I 9 know we follow various --
10 MEMBER POWERS: If it's going to be covered in 11 her presentation I don't need to interrupt. )
12 DR. PERSENSKY: Anne Ramey-Smith.
13 MS, RAMEY-SMITH: Anne Ramey-Smith, Office of
('s ;
! )
k/ 14 Research, and eager is probably an overstatement, 15 Recently, I was at the Federal Executive Institute and was i
16 able to take some training specifically in the area of i
)
17 strategic planning. And what I tried to do with the first 18 version particularly, of this plan, is to make use of some 19 of those strategies for planning and to bring those into 20 this process.
21 And you're right; that's why you see things 22 about vision and the mission, because we try to stand back 23 and to take the big picture and what is it that we're 24 trying to do.
(p) _
25 Since that time, I have to admit that we've NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
l 12 1 gotten a little involved with the details of things, and i
,e 3 2 we need to sit back again on an agency basis, to revisit b 3 and to take the next steps that need to be done, from a 4 more systematic process.
5 MEMBER POWERS: It seems to me that systematic 1
6 processes are valuable and that this is one area where 1 7 they could be particularly valuable -- I J
8 MS. RAMEY-SMITH: I concur.
l 9 MEMBER POWERS: -- and in particular I can 10 imagine you doing substantial numbers of trade studies to j 11 decide on your activities. j l
12 MS. RAMEY-SMITH: You're exactly right, and 13 particularly when we -- one of the processes that we have i r^N
'- 14 in front of us and Jay will talk about it a little bit --
15 is developing what we're calling the objectives within 16 each of these programs that we'll discuss.
17 And these objectives, you can look at them and 18 think of them as criteria, if you will, of what fits this 19 program, what doesn't fit this program, and we need a lot 20 of discussion among ourselves to make the tradeoffs and to 21 prioritize those objectives in terms of, you know, what do 22 you have -- if you have to cut budget, how does reactor 23 safety and material safety compare? How does operating i
24 reactors and advanced reactors, and so on and so forth.
p)
(_ 25 MEMBER POWERS: Do you treat budget as a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344433
l 13 1 requirement or a constraint?
rw 2 MS. RAMEY-SMITH: At this point in time --
i
~s 3 I'll tell you what I'm doing then you can decide -- at j 4 this point in time we're trying to put together a plan l 5 that we think makes technical sense, and then once we have 6 done that, we will look at the budget constraints and see 7 what we'll be able to do first and when.
8 It may be that we have to stretch out some 9 schedules and cut some things off for now, and that's the 10 way we're approaching it.
11 MEMBER POWERS: It sounds like you're treating 12 it more like a constraint than as a requirement. Which at 13 this stage seems to be fair. I mean, at this issue we r
i
' 14 usually treat it as a requirement, but here I think -- I'm 15 comfortable with the idea of treating it as a constraint 16 at this point.
17 DR. PERSENSKY: That's the general process 18 we've gone through in the past. It also is always treated 19 as a constraint because our rollover is very limited.
20 MEMBER POWERS: It strikes me as maybe the 21 Commission Chairman is sympathetic with that point of 22 view, because some of those budgetary ideas came up in 1
i 23 connection with the shutdown risk presentation that we had 24 with them, and during that presentation she seemed to rm (
(_,/ 25 espouse the view that you treat it as a constraint at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
r
( 14 l
1 first rather than as a requirement in your planning.
2 DR. PERSENSKY: As Wayne said, this is not a g~
i\ J
3 complete work at this point. In fact, the other part of l
4 it is, we don't have complete agreement at this point 5 amongst the various players in this plan, and we've been 6 trying to put out various drafts to get feedback and try 7 to respond to that feedback the best we can, and we're 8 doing it as an iterative process.
l 9 It started out really, primarily, with RES, !
10 NRR, and AEOD because we were the ones that were in the 1
11 process of responding to the questions from ACRS. NMSS 12 came in a little bit later, even though they had been part 13 of the HPPP, but they came in a little later on this !
\
,\
( i
's '
14 schedule for this particular activity.
15 For the future we also hope to include some of 16 the other offices or other groups, like OI -- Office of 17 Investigations -- Office of Enforcement; some of the other 18 activities that are going on that would have some sort of 19 human factors component, perhaps.
20 Maybe not have human factors people, they may 21 not call it human factors, but I think we need their input 22 as to how it might be done, because, especially given the j 73 restructuring of the agency -- OI and OE are under the 24 same chain of command with Mr. Jordan -- so we would
((~% ,) 25 probably bring them in later.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 HHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
15 l
1 We're going to have to interact with IRM for zx 2 certain issues such as the database issue. So we will be f i
^'
3 bringing other parts of the Agency into this.
1 4 A little bit of background to expand on what 5 Wayne had said, was that this all stemmed -- or, this 6 particularly activity now stemmed from your response to 7' the Human Performance Program Plan which had been I
8 developed as a device for coordinating the activities I 9 across the agency.
10 You had concerns with regard to the fact that 11 it may be nothing more than a list of activities -- and it 12 was. It was a needed exercise that we had gone through.
13 You also indicated that there was a need for a model in
,n \ .
(_/# 14 order to better structure the plan, and we agreed.
15 And at this point we're focusing on ATHEANA as 16 a model. It's a technique for human error analysis, which 17 is described in NUREG/CR-6350. ATHEANA has a couple of 18 different places in this presentation so I'll talk to that 19 further later on.
20 You also indicated a need for ownership; that 21 didn't seem that the HPPP had ownership because we had 22 been ao ad hoc committee that was coming apart. A policy 23 decision was made by our management to assign the 24 responsibility for this new plan to RES, and Wayne I
! rS I
(. ,) 25 guess, is the titular owner of this particular function, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
16 l
1 but if I do wrong, I'm the one that gets hit as well, so I 2 have some personal -- and it all rolls downhill.
73 I \ l
~'
3 MR. HODGES: And this is not a research plan.
4 DR. PERSENSKY: But it is meant to be --
5 MR. HODGES: That's why we've got all the 6 offices present.
7 MEMBER POWERS: Are you saying that the 8 ownership is now clear?
9 DR. PERSENSKY: The ownership is now clear 10 based on management direction from the EDO --
11 MEMBER POWERS: From Morrison to Hodges to you 12 to Anne?
13 DR. PERSENSKY: Yes. At least parts of it.
m/ 14 CHAIRMAN APOSTOLAKIS: I believe all this time 15 that it's Wayne. Now Wayne --
16 DR. PERSENSKY: Wayne is the one that's 17 actually --
18 CHAIRMAN APOSTOLAKIS: Wayne can ask his staff 19 to help him, but it's Wayne.
20 MR. HODGES: I'm the one who's on the hotseat 21 if it doesn't work.
22 CHAIRMAN APOSTOLAKIS: He's the one who will l 23 be paying taxes because he's the owner.
24 DR. PERSENSKY: Okay, that does cause some --
, ~3
(,) 25 I think that part of that has a negative impact though, on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433 1
17 1 availability of the staff from the other groups, but
,m 2 they're all working -- well, we're working together the I 1 O'
3 best we can, given all the various constraints we're
- 4. under, one of which is time -- because of the very short 5 timeframe we were assigned this.
6 MEMBER POWERS: Have we gotten a copy of this 7 ATHEANA Report?
8 CHAIRMAN APOSTOLAKIS: Yes, we have this copy.
9 DR. PERSENSKY: The other aspect that came up 10 -- to some extent from ACRS but also very clearly, from 11 NSRRC -- was that we didn't appear to have a close enough 12 linkage with the HP.A activities -- the PRA, especially the 13 HRA activi t-ies.
/ \
\ )
V 14 Part of a response to that is that Anne, 15 Catherine Thompson, who were doing the HRA work in Mark's i
16 branch, have now been transferred into my branch, so that 17 the human factors people and the HRA people are working I l
18 directly together. Anne is the team leader for the human 19 performance and reliability team -- we keep changing the 20 name.
21 CHAIRMAN APOSTOLAKIS: So ATHEANA is under 22 Anne, now?
23 DR. PERSENSKY: ATHEANA is under Anne. Roy 24 Woods also came over in that transfer as a systems
,e C) 25 engineer because of the need to have that kind of inter-i i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
18 1 disciplinary background, to be involved with the --
fo 2 CHAIRMAN APOSTOLAKIS: So Roy is now with x.j 3 Human Factors Group? ;
4 DR. PERSENSKY: Right. ]
5 CHAIRMAN APOSTOLAKIS: Okay. l l
6 DR. PERSENSKY: The other aspect, and just to l
7 sort of round out the branch discussion of the current 8 branches, I also have the responsibility for the digital i l
9 systems area which includes both hardware and software. I 10 That group is also working on a plan in response to the l 11 National Academy of Sciences, but that is a research plan.
12 There is going to be some interface between the two as we 13 move along, simply because interface falls into both j l',,'s ,
U 14 areas. So that we will be working closely. j 15 MR. CUNNINGHAM: Excuse me. With respect for 16 ATHEANA, do you have anybody from your branch, Mark, 17 participating in it? Because that's a human reliability 18 effort development effort, right?
19 MR. CUNNINGHAM: That's right -- well, just to 20 be clear. Anne and Catherine had been detailed to this 21 branch and how it's all going to work out in the end -- l 22 right now we're relying on Anne and Catherine as the 23 interfaces.
24 CHAIRMAN APOSTOLAKIS: Okay.
,9 MR. CUNNINGHAM: With ATHEANA, with PRA
() 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
19 1 people, basically.
,7- \
2 CHAIRMAN APOSTOLAKIS: Okay.
r :
'wJ 3 MR. CUNNINGHAM: They have one foot in each 4 camp, if you will.
5 MR. HODGES: The human reliability work that 6 was being done in Mark's branch is going to be done in 7 Human Factors, in Reliability -- whatever they're calling 8 it now -- branch. Anne has that responsibility brought 9 over with her. Also with that is a mandate to talk 10 frequently with Mark and the people of his branch to make 11 sure that everything is coordinated.
12 DR. PERSENSKY: As Wayne indicated, at this 13 point perhaps, if you look at it, we've got an ts \
/
(. l 14 introduction, then we still have a list of projects, and 15 part of what we're doing currently is trying to better 16 organize the activities so that they more closely relate 17 to some objective criteria that we might use for 18 prioritizing.
19 That again, was an issue with the HPPP; is how 20 do we go about this prioritization? Prioritization is 21 very important, especially when it comes to budget 22 activities. Because we can list everything and if we got 23 funding for everything it wouldn't matter, but since we 24 know we're going to have limited funding we do need to O So we have various programs and we
'(_,/ 25 have prioritization.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
1323 RHODE ISLAND AVE., N.W.
I (202) 2344 433 WASHINGTON, D C. 20005 3701 (202) 234-4433
20 1 have activities within the programs.
i
,- 2 Now to get into more formal aspects of the
~~'
3 plan. We started this one off with a vision statement.
4 This is still in flux; we'd appreciate your comments. In 5 fact, during this thing, since it is a work-in-progress, 6 we held that it's never stopped you before, but I don't 7 want to just sit up here and talk to you; I want you to 8 interact with us as much as possible. When we get into 9 some of the more specific programs, the reason these other 10 people are at the table is to talk about their various 11 program areas.
12 But this is the statement that is currently in 13 the draft and really gets down to the issue of the i
\~- 14 contribution of human error associated with the use of 15 nuclear materials. It's supposed to be broad so it's not 16 just nuclear power plants. It gets into the NMSS 17 activities as well, and what can we do to control that 18 contribution of human error.
19 CHAIRMAN APOSTOLAKIS: I really don't find 20 anything wrong with this statement, but something is 21 bothering me; I don't know what it is. Maybe it will come 22 later.
23 MEMBER POWERS: What bothers you is that you 24 see a statement that cries out for quantification --
,m CHAIRMAN APOSTOLAKIS: Yes.
() 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
21 1 MEMBER POWERS: -- which is probably not 2 appropriate in a vision statement. But that's what grates 7y
\~'l 3 on your nerves. I have to say that my initial reaction to 4 it is, it's one of the better vision statements that I've 5 seen and it really honestly tells me what you're trying to 6 do.
7 So I kind of like it, but I have to admit that 8 there's -- " unduly and disproportionately" are such crying 9 out for quantification terms that -- but I think you just 10 have to wait to see the rest of the plan to see them get 11 quantified.
12 CHAIRMAN APOSTOLAKIS: Well, there is 13 something implied there that, if it contributes to risk
,.~.
i \
'd 14 proportionately to something, it's okay. And I'm not sure 15 I like that. You see what I'm saying?
16 DR. PERSENSKY: Yes.
17 MEMBER POWERS: But I think they cover 18 themselves when they say -- they avoid the unduly as well.
19 CHAIRMAN APOSTOLAKIS: Isn't the idea here 20 really, trying to understand what human performance is all 21 about and then control it?
22 DR. PERSENSKY: That's an element of it.
23 CHAIRMAN APOSTOLAKIS: That's an element of 24 this?
25 DR. PERSENSKY: The understanding of it is an
()
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
22 1 element of it, but from the standpoint of performance, 2 it's really to get down to, what can we do to reduce the
" 7 w} ,
3 impact? We know we're never going to eliminate human 4 error altogether, but to reduce the impact of --
5 MEMBER. SEALE: There is an element --
6 MR. ROSENTRAL: Reduce the impact where 7 appropriate.
8 MEMBER SEALE: There is an element of job 9 security in that statement because you never know when 10 you're through --
11 DR. PERSENSKY: Thank you, Bob.
12 MEMBER POWERS: I honestly think that you have 13 to look at the totality of things --
/~N i kl 14 MEMBER SEALE: Sure.
15 MEMBER POWERS: -- before you try to take 16 about these vision statements. You know, as a starting 17 point it didn't look -- I mean, it's pretty good compared 18 to the vision statements you get from the DOE laboratories 19 which tell you absolutely nothing.
20 CHAIRMAN APOSTOLAKIS: Would it be better -- I 21 don't know how the rest of the subcommittee has -- if we 22 delet3d the word " disproportionately"? See, that bothers 23 me. I don't know. To say notion of something is 24 proportional to something else -- why?
f%
'(_,)
25 As Jack said earlier, I mean, you may tolerate I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON D C. 20005-3701 (202) 234-4433
23 1 some human errors that may be disproportionate to
,ex 2 something. But so what? Unduly I understand. I don't i I
%J 3 know how everybody else feels, but disproportionately 4 bothers me a little bit.
5 MR. CUNNINGHAM: I guess I think of this, when 6 we use disproportionately in the context of how we talk 7 about accident sequences in PRAs if you will, but we don't 8 want an accident sequence to be the dominant -- in a 9 Eense, a big peak over all the other accident sequences.
10 And I guess I would call that disproportionate 11 also; that you don't wan ATWS to be 95 percent of the 12 core damage frequency, or 90 percent, or something like 13 that. That's certainly what I had in mind when we came up t i G' 14 __
15 CHAIRMAN APOSTOLAKIS: But wouldn't " unduly" 1
l 16 cover that?
17 MEMBER POWERS: It seems to me that when you 18 take it to NMSS, that the 18 abnormal occurrence reports 19 sent to Congress the other day, 16 of them are NMSSs 20 bailiwick; 100 percent of them are human error. You can't 21 get --
22 MR. SURIG: So we clearly have our share of 23 the pie.
24 MEMBER POWERS: It is non-- I mean, there are
(~g
(,) 25 always going to be dis -- there's no way to avoid it.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
WASHINGTON. O C, 20005 3701 (202) 234-4433 (202) 234 4433
24 ,
1 1 There are always going to be disproportionately human !
I rx l 2 error related.
\
(
l
\_/
3 MS. RAMEY-SMITH: Anne Ramey-Smith. If I can i l
4 throw another issue in here. If you look at that, in some !
l 5 of the discussion that we had, there's really two pieces !
6 of it. One is the disproportionate which as Mark says, is I
7 looking at the risk contribution. Now, it gets into the l
8 whole issue of function allocation and where are you !
9 putting your emphasis, and contribution to risk and 10 reliability associated and so on. l l
11 Unduly has a little bit different kind of !
12 perspective, and you can almost think of that -- it's not l 13 entirely this way -- but you can almost think of it as l r, !
( ) It may be that the
'N ' 14 having the cost benefit element to it.
15 contribution to risk is acceptable from some perspective, 16 but if there's something there that can very easily be 17 fixed or very easily be avoided.
18 For instance, you might say, well that's an 19 undue contribution, even though it is, you know, 20 proportionately small or relatively small. So we're 21 trying to get to, you know, both sides of it -- not a 22 vision that's completely driven by, you.know,
! 23 quantitative, human reliability assessment issues, but 24 also some of the less quantitative human factor sorts of em.
(' ,,\ 25 issues.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
' 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
25 1 DR. PERSENSKY: Do you have anything further fg 2 George, or should we just go ahead? It is a term that's
( )
v 3 raised some discomfort amongst other people.
4 CHAIRMAN APOSTOLAKIS: You know, it's not a 5 matter of being right or wrong here; it's a matter of 6 finding the right words. Unduly I can live with; 7 disproportionately -- I guess I can live with it too --
8 but I'm a little bit uncomfortable with it.
9 MR. HODGES: We will take another look at 10 that, and if come up with a word that you think might be 11 better, please --
l 12 CHAIRMAN APOSTOLAKIS: Human error should not 13 -- can we use the word " unacceptably" someplace, perhaps?
k/ 14 No , that's not --
15 DR. PERSENSKY: No, that gets into the same 16 kind of problem.
17 MEMBER SEALE: The problem, I think, is as 38 Dana alluded to. Particularly in the NMSS area there are 19 so many cases where human error is almost certainly the t
l 20 precursor to whatever else may have ensued that gave you 21 the problem. I mean, you can design a contbiner for a 22 radioactive source that will handle a fire, and most other 23 external events.
24 Generally they're not massive enough to be (3
(_) 25 involved in any kind of structural deflection problem or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 123 RHODE ISMND AW., N W.
(202) 234-4433 WASHINGTON D C. 20005 3701 (202) 234-4433
26 I i
1 anything. A human performance error is almost certainly
,o 2 the precursor to whatever else was the problem. So you're
, i ,
/l 3 going to buy into those problems whether you want to or 4 not.
5 Disproportionately, it seems to me, brings 6 down a disproportionate share of the blame on the human 7 error part of the problem. I mean, it's almost obviously 8 the first step.
9 MR. HODGES: Maybe -- and I don't know whether 10 that would work or not -- if we tried it as a positive ,
l 11 statement rather than a negative statement we might avoid l
12 some of that. So we'll think about that some.
13 CHAIRMAN APOSTOLAKIS: Yes, I was thinking
,/
YM 14 about that. Maybe start out, instead of human error say 15 human performance should be -- and then I don't know what i
16 it should be, but -- that would be positive.
17 MS. RAMEY-SMITH: Yes, that's why we have to 18 do some thinking about it, because it's not intuitively 19 obvious what the poaitive version of this statement would 20 be.
21 CHAIRMAN APOSTOLAKIS: No, it's not, it's not.
22 By the way, the traditional phrase in the safety matters 23 is that, you know, we find this not pose undue risk to the 24 health and safety of the public. So the word " unduly"
(. I don't think that the
() 25 probably is proper here.
NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON. O C. 20005-3701 (202) 234-4433
27 1 Commissioner of the ACRS has ever said that we find this
,/" 2 not to be a disproportionate contribution to --
N-]j 3 MS. RAMEY-SMITH: Although, you could think of 4 it in terms of the issues of defense-in-depth and safety !
5 margin and so on, has an element of -- that concern with 6 disproportionate.
7 CHAIRMAN APOSTOLAKIS: Sure, sure. But I 8 think undue contribution is probably -- well, but I think 9 a positive statement would be better, if there is one.
10 DR. PERSENSKY: The next statement here is the 11 mission statement. This brings into focus the current 12 emphasis on risk-informed and performance-based regulation 13 and oversight. I think the main thing that we wanted to 14 point out with this is that we are -- may be moving into 15 this area full force, to make sure that everything that we 16 do falls into risk-informed and performance-based I think, 17 in the broadest sense that we can.
18 It also covers the entire agency's activities 19 I think, in terms of the statement with the power plant as 20 well as the transport, storage, and use of nuclear 21 materials. So it's intended to say, this covers 22 everybody.
23 Since it's supposed to be a strategic plan, we 24 thought we'd have a strategy, or a couple of strategies k ,) 25 mentioned here. Generally there are three steps in this.
NEAL R. GROSS COURT REPORTERS AND TRANSCnlBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
28 1 One is first being that, we're going to go out and look 2 for problems. We're not going to wait for them to come to
(~)1 L/
3 us; we're going to identify them through some various 4 means.
5 We have a lot of sources already. We've got 6 the LER system, we've got the inspections that are going 7 done by the regulatory offices. We need to put it in a ,
8 form that is going to be most useful as far as identifying 9 where the issues are, so that the plan will be driven by 10 the activities of our licensees, in a sense, or the 11 failures related.
12 "'he next thing is, once we've identified them, 13 is try to figure'out what it is that there -- why these f,\- )
What is it that backs up; where the 1
14 problems exist.
15 problems come from. More the root cause.
16 MEMBER POWERS: If I read this strategy title l
17 as mission need, rather than strategy, would I be wrong?
18 DR. PERSENSKY: I'm sorry?
19 MEMBER POWERS: If I read this, instead of 20 being strategy, as mission need, would I be wrong?
21 DR. PERSENSKY: I'm not sure I understand what 22 " mission need" is.
i l
23 MEMBER POWERS: What you need in order to 24 carry out the mission.
(~T
(_,) 25 DR. PERSENSKY: Okay. No , you would not be l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
f l
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
29 1 wrong. This is how we would meet the mission; by 7,3 2 performing these broad activities.
3 MEMBER POWERS: If you could do everything on 4 there to your satisfaction, you would say, my human error 5 problem is solved; I'm on to other things?
6 DR. PERSENSKY: Each one of these things -- I 7 mean, if we could, in fact, identify all the human 8 performance issues that are ever going to occur, yes, we'd 9 be done. But this is, again, is going to be an ongoing 10 process.
11 One thing I've learned from being a 12 psychologist and working in human factors for close to 30 13 years, is that humans have a real, strong, innate way of
,r\
l / They can always come N/ 14 figuring out how to do things wrong.
15 up with a new, creative way of screwing up, so I don't 16 know that, again, as Dr. Seale said before, this could be 17 a lifetime job security here, because that identification 18 of issues will be an ongoing effort and as well as the 19 determination of what the root causes of those issues are.
20 The second one I think, is an important 21 element when we get into the question of models; how are 22 we going to use a model? We mentioned ATHEANA before.
23 ATHEANA is sort of the framework, the structure that we're 24 going to be using, but it also is an HRA model.
7_
(,) 25 And we've also talked about other types of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. j (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 ;
1
30 l
1 behavior models. The work that we've done in the past has i
. 2 often been driven Ly questions of behavior models.
l l~m
'~'
3 ATHEANA in fact, is driven by an information processing 4 model. And gone back and looked at the work of Roth, and 5 Rasmussen, and Jim Reason. And each time we do a project 6 almost, we get involved with a model because this gets us 7 into this issue, and we need to perhaps settle more on a 8 single model. j 1
9 Then the final step is to take some action; to i 10 do something with the information. You know, whether it's !
4 11 to come up with a new regulation, come up with further 12 inspection enforcement activities, new tools for the use 13 of our regulators. But it will also identify areas where
(_./ 14 there are holes; that we can't just go out and build a 15 tool today but that we'd have to do further research 16 analysis before we could go forward with it.
17 MEMBER POWERS: One of the questions that the 18 ACRS had in the last plan was, do you have the perception 19 that human error is too big of a component? The risk 20 worth of tha human in the system is too high, and that you 21 ought to reduce that?
22 DR. PERSENSKY: I believe that is a general 23 feeling, particularly as you mentioned, with NMSS -- the l
24 report that just went to Congress. Every one of them had l
7-s,
( ,) 25 a human component to it. Not all problems at power plants NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
31 1 have a human component, but at least on initial -- if you 7s 2 go back far enough in any root cause, you're going to find
(' ' ')
3 a person in there that did something wrong.
4 I think the basis, you know, why we are going 5 forward with this and why the agency has a Human Factors 6 Group at all, is because there are concerns with human 7 behavior and human performance amongst the licensees.
8 MR. ROSENTHAL: Let me draw the distinction 9 between the value being too high and the uncertainty being 10 too large. And at least in my mind, I think that you need 11 to work the uncertainty side of the equation. And my PRA 12 -- my equipment reliability, I think we're getting a 13 handle on that.
/3 5 ;
(_/ 14 We've had a lot of work going on in common-15 cause database and we've got this human performance 16 aspect. And in my mind, I think we need to reduce the 17 uncertainty or increase our knowledge. And that doesn't 18 necessarily mean to reduce the risk.
19 MEMBER POWERS: What I'm looking for is of 20 course, some measure of this so that you have a metric, to 21 say when have I done enough? For instance, in NMSS, my 22 reaction by the way, to reading all these abnormal 23 occurrences, in reading them I said, well when you think 24 of all the times that there were applications of nuclear
/
( ,T) 25 materials and they only had 16 events, gee, that didn't NEAL R. GROSS COURT REPORTERS AND TPdNSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASH NGTON. D C. 20005-3701 (202) 2344433 l
l l
32 '
l l
1 sound too bad to me. That was my initial reaction to it.
1 7s 2 So maybe they don't have a crucial, human
> l
\ ,)s 3 performance problem. Maybe they have a crucial I
4 uncertainty in human performance, I will concede, but it 5 would be comforting, especially in a mission need context, 6 to have some metric that I was shooting at, either an 7 uncertainty range that I wanted to narrow or an equivalent i
8 of a risk worth that I wanted to reduce, and as a metric 9 on my programs.
10 DR. PERSENSKY: Anne.
11 MS. RAMEY-SMITH: A comment on that. I think 12 what you've brought up is part of the reasoning behind 13 some of the changes in the organization that we've seen,
,/<3
\ \
\/ 14 because it's important to be able to look at human 15 performance and human factors issues and so on, in the 16 context of safety.
17 And what we're trying to do here is to marry 18 the behavioral sciences with human reliability assessment 19 in our work in HRA, to then carry that into, what does 20 that mean in terms of safety and risk.
21 That in a sense you can begin to understand i 22 the consequences, not just the cause of human error but 23 what's the consequence, and begin to get a feel for that 24 metric.
f3
( ,) 25 MEMBER POWERS: So what you're saying is that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
33 1 part of the strategy is in fact, to develop the metric?
<x 2 MS. RAMEY-SMITH: Yes.
l \
LJ 3 MEMBER POWERS: If I may ask one other 4 question about your strategy. It is an assumption of this 5 plan that correction to the problems identified in the 6 first two bullets are achieved by changing of regulations?
7 DR. PERSENSKY: That's one way of doing it, 8 not necessarily the only way. I mean, identification of 9 the problems may also be something that can be done to 10 identify areas that the industry itself may deal with.
11 MEMBER POWERS: See, one of the problems with 12 the strategy right now is, you say we'll identify area 13 issues, we'll understand those issue, but it doesn't say in)
14 you'll correct anything -- unless I assume that changing 15 regulations makes the correction -- changes the situation.
16 DR. PERSENSKY: To some extent that's the tool 17 that we have. I think, Wayne did you have --
18 MR. HODGES: Well, no. I was just going to say 19 that probably regulation, although it's listed first 20 there, would be the last one we would actually go to, 21 because it's difficult to regulate in that sense, people's 22 behavior. You can regulate something on a piece of 23 hardware, but to regulate organizational performance or 24 individual errors, that's a different matter altogether.
O
(_) 25 MEMBER POWERS: I will accede that I read the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1r3 RHODE ISMND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
34 1 word change regulation more than the 10 CFRs. It might be
,x 2 changing the Reg. Guide, it might be changing the
( I w/
3 inspection plans, it might even be changing the training 4 programs. I read the word broadly.
5 DR. PERSENSKY: And I interpret it broadly as 6 well, that's why --
7 MEMBER POWERS: You might want to make that 8 clear.
9 DR. PERSENSKY: -- some of these other things 10 that are included below in terms of the inspection 11 activities. But I also think that probably what isn't in 12 here is, if we had lessons learned that can be implemented 13 outside of the regulatory framework, something that the f I
14 industry itself would take on, is another approach to it.
15 MEMBER POWERS: We have a relatively proactive 16 industry when it sees things in its own self-interest.
17 DR. PERSENSKY: Right.
18 MEMBER SEALE: Well, earlier you alluded to 19 the fact that you were continually reminded of the 20 innovative ability of people to figure out how to screw 21 up.
22 DR. PERSENSKY: Right.
23 MEMBER SEALE: And you've also indicated -- or 24 Dr. Powers indicated -- that he was rather intrigued to g
( ,, 25 see how unsuccessful we were in challenging people to find NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
35 1 ways to screw up in the handling of some of these things. j ys 2 There had only been 16 in thousands of sources out there I ) ,
3 and things like that.
4 MEMBER POWERS: Well, in fairness, one doctor 5 can rack up a score.
6 MEMBER SEALE: I will grant you that, but this 7 goes back then to the question of, what is it that 8 motivates people to do these things? And quite often -- l 9 I'd say probably more often than not; certainly in the 10 case of power plants and so on -- it's the climate that j 11 the people live in in the plant.
12 MEMBER BARTON: I think it was the l 13 organizational cycle safety culture --
(D
%I 14 MEMBER SEALE: Yes. .
15 MEMBER BARTON: -- in the organization -- ;
16 MEMBER SEALE: And just the way the company 17 goes about evaluating people for promotion or whatever 18 else it may be. And some of those things have very little l 19 to do with regulation. And so if you're only looking fer 20 those things that are amenable -- and I mean in the 21 broadest sense of the word regulation -- if you're only 22 looking for those things that are amenable to regulatory 23 remediation, then you're only looking under one side of 24 the rock.
/s
( ,) 25 DR. PERSENSKY: Well I think, for instance, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1r3 RHODE ISMND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
36 1 1 the issue you brought up, we are looking at those issues l c 2 as well. That's one of the programs -- we go through this I
( ) l 3 in somewhat higher -
4 MEMBER SEALE: Okay, fine.
5 DR. PERSENSKY: You'll see that those are some 6 of the issues we're going to be addressing.
7 MR. ROSENTHAL: Perhaps you can provide come 8 guidance to us. I'm very comfortable looking at what goes 9 on inside the control room, for example. On the other 10 hand, I mean, the Russians believe in examining the 11 psychological makeup of the applicant for a position in 12 the control room; which isn't an area that we've 13 traditionally gone into. And maybe you can share some of t,,
O) 14 your thoughts about where we should slice this issue.
15 CHAIRMAN APOSTOLAKIS: In fact, from the last 16 conversation, I got this idea that maybe your vision 17 statement should be stated in a positive way and should 10 refer to the safety culture; which is broader than human 19 error. That would be a good statement; that you want to 20 have a good safety culture, and then find the right word 21 to bring --
22 MEMBER BARTON: That somehow gets you to 23 reducing human error -- will minimize human error --
24 something like that.
! ) 25 CHAIRMAN APOSTOLAKIS: The safety culture v
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
37 I 1 should -- the establishment and maintenance of a good i
s 2 safety culture -- whatever -- process of handling nuclear
( )
'~
3 materials, or something like that. That would be a ;
4 positive statement. It is using words that are very much 5 in vogue these days, and -- what else -- it is positive.
6 MR. HODGES: Yes. We clearly want to have one i 7 that is positive and includes everything we need to do l 8 here -- and I think there may oe something missing from 9 that. We also want something that we can agree on what it 10 means, and that may be a problem with safety culture.
11 CHAIRMAN APOSTOLAKIS: Yes, but it's a vision.
12 DR. PERSENSKY: And it's not disproportionate.
13 MR. CUNNINGHAM: Something that might be y\
i !
k/ 14 helpful in that regard, as I recall, a few years ago --
15 was it NSAC or somebody -- had a report that dealt with 16 the issue of --
17 CHAIRMAN APOSTOLAKIS: Yes, NSAC-4 defined 18 safety culture.
19 MR. CUNNINGHAM: Okay. And that might be 20 something. And there may be an analog of a vision 21 statement there, I don't recall. But it might be a good 22 source.
23 CHAIRMAN APCSTOLAKIS: Yes. Safety culture 24 would include things like a questioning attitude from the
(~
()j 25 staff and so on. So you know, I think NSAC-4 will give NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
1 38 1 you some ideas. Because they also address some of the l 1
zm 8 2 other issues that you're touching upon here. l
. ,.,Y 3 They talk about tangible evidence of a good 4 safety culture, and so on and so on. They recognize it's 5 hard to do it. So NSAC-4 probably would be a good --
6 MEMBER BARTON: There's also an INPO document, 7 96006, which talks human performance and safety culture --
8 got some stuff on safety culture as well.
9 DR. PERSENSKY: It's 96006?
10 MEMBER BARTON: I believe that's it -- 12 11 performance objectives and criteria.
12 CHAIRMAN APOSTOLAKIS: I think safety culture 13 would -- some positive statement involving the safety
,3
'v' 14 culture probably would be a good idea. I can't think of 15 it right now, but either of us can take some time. And 16 then when you come back here we'11 tell you no, human 17 error should be there.
18 DR. PERSENSKY: Right. The plan right now is 19 built around seven programs.
20 MEMBER POWERS: Would these be what I might 21 call functions?
l l
l 22 DR ., PERSENSKY: You might call it functions.
23 In general, there are two kinds of things we do in the 24 agency. One is the regulatory -- the inspection,
() 25 enforcement -- the kind of thing that you're going to find NEAL R. GROSS COURT REPORTERS AND TRANSCRIBEPS 1323 RHODE ISLAND AVE.. N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
l 39 l 1 in regulatory offices in NRR and NMSS. The other is more r3 2 analysis and review and research, which is in the program
- ! i l
\ /
3 areas, under Mr. Jordan.
4 These aren't currently organized that way; 5 we're still looking at best organization for these. And 6 right now they're not -- though if you read it, the first 7 blush is going to say, well that's an NRR function, and 8 that's an AEOD function, and that's a Research function.
9 It's not really limited to those particular groups. The 10 intent at this point is not to limit them to those groups 11 but to have in a sense, a lead for each.
12 The first is to conduct an operating events 13 analysis and database maintenance to support -- this is
! s
\' l 14 the going out and finding out what's going on. What are 15 the problems; what are the issues? This has traditionally 16 been an AEOD function. NRR has developed their database, 17 the HFIS, which they have a use for.
18 CHAIRMAN APOSTOLAKIS: Now, these areas are 19 presented here just in some sort of order that doesn't 20 mean anything?
21 DR. PERSENSKY: Well, they are -- in one view 22 of the plan, it was sort of a progression as to how we 23 would go through. The first being, looking at the data 24 that sort of followed the strategies.
(~\
(_,) 25 CHAIRMAN APOSTOLAKIS: Shouldn't though -- I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 13?3 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
40 1 mean, if you want to put it that way then it seems to me
,s 2 that programs 4, 5, and 6 really should be listed as 1, 2,
! \
3 3 because even though it's an iterative process, you 4 really have to have some idea of what you're looking for 5 when you go and look at the data.
6 You never develop the model without the data, 7 of course, and you never --
8 DR. PERSENSKY: Right. It's chicken and egg 9 and it's never --
10 CHAIRMAN APOSTOLAKIS: Okay, as long as we 11 understand -- it's no problem.
12 DR. PERSENSKY: There was a logic to it. I'm 13 not sure that in fact --
p)
(- 14 CHAIRMAN APOSTOLAKIS: No, that's fine.
15 DR. PERSENSKY: -- we've been talking about 16 the need to perhaps reorganize to meet a different logic.
17 We've talked with NRR about that at least, but I'm not 18 sure that we talked about it with all the other offices.
19 But I don't think they're going to change as 20 far as, these are the programs, they're the functions that 21 the agency currently carries out. As I said, the program 22 has in the past, been traditionally an AEOD, and will 23 probably remain in that.
24 CHAIRMAN APOSTOLAKIS: Now, I realize that
()
irn 25 this is a Human Performance Program Plan, but maybe it's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 41 1 early enough in the process to expand it a little bit.
7- 2 We're already talking about safety culture. Is that l )
'~'
3' something we don't want to do and bring the word safety 4 culture here? Although a lot of it, you know, you're 5 coming --
6 DR. PERSENSKY: Safety culture is going to l
7 come under Program 5.
8 CHAIRMAN ADOSTOLAKIS: Program 5.
9 DR. PERSENSEY: That's the primary -- one of 10 the things -- the management and organizational influences 11 -- that's where we're addressing it. If you took the l
12 various activities that we've been assigned, it would l 13 probably fall most likely, in there.
[~'T
(_) 14 CHAIRMAN APOSTOLAKIS: So I guess what we can 15 do, or maybe in one of your documents, and maybe next time l l
l 16 we meet we should have some working definition of safety l l
l 17 culture, because I suspect we have different -- l l
l 18 DR. PERSENSKY: DSI-23. !
l 19 CHAIRMAN APOSTOLAKIS: -- definitions. I go l
\ \
20 with an NSAC definition which I think is broader than just l l
21 --
22 MR. CUNNINGHAM: Much broader.
23 CHAIRMAN APOSTOLAKIS: -- organizational and 24 managerial factors. So, you know, as long as we eay, ,
<~s 25 this is what we mean by it -- because I think this is a
(\~-)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
I 42 1
1 good opportunity for the agency really, to make these I
,s 2 words part of their vocabulary, so people will at least i
\' ')
3 ask, what is it?
4 MEMBER POWERS: Let me ask you a question a 5 little bit out of context with the slide, but maybe you '
6 should go through the slide and point to it --
7 DR. PERSENSKY: If we're going to get to it 8 later I'll tell you.
9 MEMBER POWERS: The Chairman keeps asking us 10 about prediction, and she's interested in, for instance, 11 human performance. Can we identify characteristics of i
12 degrading human performance such that we catch it before 13 it has what you called it -- consequences, I guess -- in
(_) 14 the handling of nuclear materials? Is that an element of i 15 this plan?
l 16 DR. PERSENSKY: Yes. l 17 MEMBER POWERS: And where would I find that?
18 DR. PERSENSKY: Well, it's in a couple of 19 different places, in a sense. I mean, it's still in a 20 developmental phase so you're going to find it in areas l l
21 like number 4 and number 5.
22 MEMBER POWERS: Should it not appear in what I 23 call the mission need and you called strategy? I'm 24 continuously trying to r.ap this --
,m
( ,) 25 DR. PERSENSKY- You're going to make this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1r3 RHODE ISMND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
43 1 strategy bigger and bigger yet. No, no --
fs 2 MEMBER POWERS: I'm continuously trying to map I )
\m/
3 your program onto a template which -- if you want to know ;
4 what my template is, its the IEEE standard on systems 5 engineering. So I'm continually doing that mapping here 6 all the time -- and you need not pay any attentior. to my 7 mapping; it's just the way I think about things.
8 Every one of these structured planning 9 programs has its own set of words that it puts in the <
l 10 titles and they all -- what's in them is always the same; l l
1 11 it's the title at the top is different. But it seems to ;
12 me that predictive component is the mission need at least 13 the Chairman foresees, and maybe you want to make it i
\> 14 explicit. l 15 DR. PERSENSKY: In the strategy.
16 MEMBER POWERS: Strategy is your mission need 17 -- and it seems like that's what she sees as a need, and ;
18 you know, that predictive component shows up in the 19 discussion in the senior management meetings and things 20 like that. They're looking for ways of catching things l 21 before they get to the point they get her picture on the 1
l l 22 cover of Time magazine.
23 DR. PERSENSKY: Don't say that to her.
24 MEMBER POWERS: What you don't want is your (n. ,) 25 picture on the cover of Time magazine.
NEAL R. GROSS OOURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
44 .
1 DR. PERSENSKY: 1 think we are looking at -- I e~ 2 in fact, we're involved with the work that AEOD is doing G
3 in that area for the senior management meeting. In our i
4 elements here, if you look at a program, the work is ;
i 5 actually being performed -- if we were to put a FIN to it, 6 you know, how we do things in activity -- probably under 7 mostly program 5 -- but of course, the data that we're 8 going to be using is going to come out of program 1 -- out 9 of all of them --
10 MEMBER POWERS: Every one of the functions 11 ought to tie in to the --
12 DR. PERSENSKY: It all ties together.
13 MEMBER POWERS: -- mission needs someplace,
/ \
\
\s) 14 yes. I mean, there should be a line that draws right in.
15 DR. PERSENSKY: But from the standpoint of 16 this predictive aspect -- AEOD has a role; the data from 17 operating events has a role; the data that comes in from 18 the various inspection activities -- that has a role in 19 this whole question of, how do we build a template for 20 predicting downturn in performance of a licensee?
21 You know, this may not have anything to do 22 with power plants, but on the other hand if we try to 23 apply the same template, NMSS, the kinds of inspections 24 that they do. But somewhere it needs to be pulled
(,)
('~h 25 together, and it's pulled together primarily under program NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202; 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
45 I
1 1 in this activity. But the development activitier that
-s 2 go along with it -- you know, how do these things relate,
( \
/ l 3 how do they correlate -- are going to come out of the !
4 research activities as well. I 5 Mark, I'm sorry to keep --
6 MR. CUNNINGHAM: I was going to say, I was 7 just reading the -- going back to the strategy -- and I 8 guess the third word being proactive, I guess you could I l
9 take that to mean several things. And I guess one ;
10 interpretation of proactive is trying to be a little more 11 predictive in terms of human performance.
1 12 We could probably say, yes, we thought of that 13 and it's under proactive, but that may be a little I
~s !
/ 14 overstatement, or something.
l 15 MEMBER POWERS: I'd like to be able to sell it 16 to the Chairman. I'm reading between the lines on things j 17 that she's asked, but she's continuously asking us about ;
18 predictive capability.
l 19 DR. PERSENSKY: My shorthand for the first l l
l 20 bullet in strategy was, we're going to go out looking for 21 problems. That may be a way of getting into this issue of 22 prediction as well. Anne, you had --
23 MS. RAMEY-SMITH: Again, if, you know, one of j i
24 the recommendations that had come from this committee and
/~N
( ,) 25 others was to have a framework in which you look at all i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. ,
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
46 1 the work that's being done. And of course, you know, i
,s 2 we've chosen to apply the ATHEANA framework. .
t r i
'~' )
\
l 3 And what that is, and the purpose of looking i l
4 at human performance that way, is to do just that -- it's 5 to predict performance. So that if you understand what 6 goes into human performance and what is driving it, then l
7 theoretically what you can do is predict that and to put 8 in place regulatory programs and so on, that is in line 9 with that.
10 So it's definitely an important issue and I 11 agree with you; we probably need to highlight it a little i 12 bit. Whether it belongs in the mission or the strategies 13 or wherever, we need to take a look at it, f~~s, V 14 CHAIRMAN APOSTOLAKIS: Let me --
15 MEMBER POWERS: You could equally well put it 16 in the requirements for the function. Those are in the 17 tape. I don't think you're going to get to requirements 18 on these functions. Yes, it's interesting. You've done 19 the -- from the abstract planning basis, you're doing the 20 functional analysis before you do the requirements 21 analysis.
22 There is a ferocious debate in the systems 23 engineering community about which way to do it, and there 24 are exactly two schools of thought -- both of them n
() 25 disagree intensely. And you know what the two schools NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASH!NGTON, D C. 20005-3701 (202) 234-4433
47 1 are. i
,s 2 CHAIRMAN APOSTOLAKIS: I'm looking at the l' \
' '/
~
3 agenda here. Anne Ramey-Smith was supposed to talk about 4 vision, mission framework. Are you doing this, Jay? !
l 5 DR. PERSENSKY: Yes, I'm -- l 6 CHAIRMAN APOSTOLAKIS: So we're not behind 7 schedule now? j 1
I 8 DR. PERSENSKY: No, we're not behind schedule.
9 No, I'm taking the lead and she's going to be the good 10 question answerer in most cases.
11 CHAIRMAN APOSTOLAKIS: Oh, okay. Okay. Good.
12 So we're on time?
13 DR. PERSENSKY: Yes, we've got plenty of time.
- A x-) 14 If you want to get out by three we should --
15 CHAIRMAN APOSTOLAKIS: We will.
16 DR. PERSENSKY: You will get out by three.
17 Actually, with any luck you'll be out by two. I'm ready 18 to leave now if you want to quit.
19 MEMBER POWERS: Oh, you're not deriving a 20 stimulating inspiration from this interrogation you're 21 getting? I'm disappointed.
22 DR. PERSENSKY: Actually, I am enjoying it, 23 but I know you have to get back --
24 MEMBER SEALE: George insists that I ride in
~
(3) 25 an airplane for six hours today. And that's what's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
48 1 killing my back right now.
<3 2 DR. PERSENSKY: Well, I just got off of one.
I )
Q ,/
3 MEMBER SEALE: Is that a system requirement or 4 a system function?
5 DR. PERSENSKY: Back to the next program --
6 the next program is to provide the human performance ;
7 personnel with the concepts, the capabilities for the 1
8 inspection review activities for power plants. This is i
9 the, again, traditionally the NRR role.
I 10 Program 3 would be what we would consider the ;
1 1
11 NMSS role. Again, their inspection. But as part of these 12 things, RES, other groups, may help to develop the tools 13 that they need. They need to have tools to do their
,,m
! )
\/ 14 inspections and that's what we've done in the past.
15 Program 4 is the development of the technical 16 basis. This would be the traditional RES role. This is 17 where we go do the research to identify where the problems 18 are, to come up witn guidelines, come up with the tools 19 that are used. This is what we in RES, have'been doing 20 for many years now.
l 21 The next two -- well, the next program is sort
! 22 of this issue of safety culture; the issue of 23 organizational factors. It's changed terms over the 24 years, but it's organization management; it's the effect g
( ,) 25 of the people that own and operate the licensees on their NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 l
i 1 workers, and on the performance of their facilities.
I 2 MEMBER POWERS: If I look at the preceding l 73 N) 3 programs, I see the first one is collection of data, 4 historical function of the NRC mandated in the NRC mission 5 statement; I see programs on inspection. Again, 6 historical mission in the NRC mandated in its charter.
7 Human reliability analysis, part of o" t;chnology for 8 evaluating risk. It h s to be done.
9 Then I come down to this organizational 10 factors management. I don't see that flowing from statute 11 to mission, the need to do this -- this function.
12 DR. PERSENSKY: What I think drives this more 13 than anything else is the fact that in doing these other
/ 1 i /
14 activities -- inspection activities, the database 15 activities -- is that the issue that keeps coming up as 16 something that's been unresolved, something that we have 17 not dealt with -- in a consistent way at least -- has been 18 into an organization.
19 The whole question of the senior management 20 meeting, you know, they look at, what is the performance?
21 The Arthur Anderson report, a lot of what drives the 22 performance, especially the poor performers, seems to be 23 related in some way to factors that are more associated 24 with the management and organization than with the valves
,m
( ,) 25 and pipes and the wires and the control rooms.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
50 1 All of those things, if there's deficiencies
,_ 2 in that, seem to leak back to this management
\' ,)
3 organization. So that's safety culture. There are issues 4 that --
5 MEMBER POWERS: Is this -- I guess, is this a 6 more orthogonality performance-based --
7 DR. PERSENSKY: I can imagine a hypothetical 8 management organizational structure that produces perfect l
9 performance might be quite different. I mean, a variety 10 of management and organizational structures might yield 11 perfect performance. ;
i l
12 DR. PERSENSKY: And I think that's kind of ;
1 13 what we found in some of the research we've done, is that
,r q
\
ws l 14 there is no one perfect organization. There are elements 15 of a good organization, and we're not all satisfied on all 16 of those as yet. But you know, the current term of art or 17 the term that's in vogue right now is safety culture.
18 MR. ROSENTHAL: Cecil and I are waiting our 19 turns.
20 DR. PERSENSKY: Okay. These are all issues 21 that seem to keep coming up. Since my colleagues, they've 22 been quiet so far; I'll let them jump in.
23 MR. THOMAS: Maybe unlike some of the other 24 aspects that you mentioned that do have a regulatory
(~N \
( ,) 25 basis, management organization, management culture and l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
51 3 safety culture collectively, could be something that this 7
2 program drives.
)
3 There may not be today, a regulatory 4 underpinning for those areas like there is for inspection, 5 for example, but maybe this is some way that we can 6 influence -- if we can show a need -- we can influence the 7 regulatory framework and at least provide some kind of 8 guidance for evaluating, for regulating, or whatever, to 9 the extent that's appropriate.
10 So in this case, this is I think, part of the 11 proactive approach of the plan, as opposed to reacting to 12 what's on the books.
13 CHAIRMAN APOSTOLAKIS: Jack? No comment? I
,m i I N/ 14 thought you wanted to --
15 MR. ROSENTHAL: That's fine. I just wanted to 16 go on, and I think some of this is in the planning that 17 we're doing and we -- if I start out by saying, okay, I'm 18 interested in equipment reliability and availability, I'm 19 interested in common cause, I'm interested in human 20 performance. I l
l 21 And then I go another level down from there j l
22 and I say, okay, what management and organizational i
23 factors are driving my equipment reliability, let's say.
24 Then I think I'm on solid ground. It's outcome-oriented.
/~~'N
() 25 To take -- and that's very different from looking at a ;
N EAL R. GROSS '
COUP; HF?ORTERS AND TRANSCRIBERS 32? RHODE ISLAND AVE., N W.
(202) 234-4433 HINGTON, D C, 20005-3701 (202) 234 4433
52 1 management structure in isolation --
-s 2 MR. HODGES: I don't think --
( t
\/ ,
3 MR. ROSENTHAL: -- they're agreeing to take a l
4 look within the context --
5 MR. HODGES: This is not structure, in my 6 view. This is safety culture I think, as people have j 7 tried to use it, and it's not just the utility, it's the 8 NRC as well.
9 MEMBER BARTON: Sure. Let me --
10 MEMBER POWERS: I believe that I would be much 13 more explicit and careful in this area because, certainly 12 if I were a utility licensee, if I were an NEI, I would 13 say, these guys are getting in an area that's outside p
(_-) 14 their expertise. I know how to manage plants, I know how 15 to manage medical operations -- they don't; they never 16 have.
17 And they have no business and they have no 18 regulatory direction here. I would think of following a 19 template closer to what Jack has, which says that I would 20 take 5 and subsume it into 4 as an outcome, a finding, 21 rather than a target in itself 22 CHAIRMAN APOSTOLAKIS: And I would rename 4.
23 I think we should be using safety culture at some point, 24 because then it falls naturally there. I mean, if you
,-\
t
) 25 read the NSAC-4 report, they make a big deal out of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
I 53 l 1 organizational structure. ;
I gy 2 By the way, management and organization don't I'~') 3 go -- you usually see them together but they don't really 4 have to go together. You can look at organizational 5 structure without looking at management.
6 MEMBER SEALE: It's interesting. I'm tempted i
7 to say that what's happened in the past is that we haven't l 8 included the management and organizational structure and 9 that's been the wild card in almost any -- in many of the l l
10 assessments of what's gone on, and perhaps what you' re 11 trying to do here is to recognize that or identify that as I l
12 a full-time player in the problem. !
13 MEMBER POWERS: And I think --
/~'s
/ 14 MEMBER SEALE: There's a -- if I may complete.
15 There are other assessments, other evaluations of what has 16 gone on in some of the cases of problems that have 17 occurred in utilities, which strongly support the 18 identification of organizational factors, and management 19 is a part of the problem.
20 INPOs evaluations are loaded with examples 21 where management has had difficulties in perhaps conveying 22 the appropriate set of performance goals to the employees.
23 MEMBER BARTON: They call it -- expectations 24 and standards is the buzz words you hear from those folks.
(Q,/
25 MEMBER SEALE: That's right. And they've also NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
54 1 been quite outspoken in their protest, when they feel that g3 2 the NRC has overstepped its bonds in interacting. Not i j
'~#
3 that they necessarily, accept those bounds as being 4 theirs, but they feel that those bounds are not 5 appropriately the NRC's.
6 So we have a problem here, I think, in --
7 there's an organizational problem that runs down the 8 middle of this probably, if you will. And perhaps the 9 idea of merging these things together under safety culture 10 is a reasonable way to finesse the difficulty.
11 MS. RAMEY-SMITH: If I may make a comment 12 regarding this area. We've heard some discussion about 13 whether there's a regulatory basis. Earlier we talked r~w\
i 2 14 about uncertainties and the role of uncertainties in I 15 look.ing at human performance and human reliability. We've
]
16 talked about predictive capability, we've talked about I 17 models, and so on.
18 What I say regarding this area of management 19 and organizational influences is that I concur. We have 20 to tread carefully into this area from a regulatory 21 perspective, in terms of imposing requirements and this 22 sort of thing, onto the licensees.
23 On the other hand, however, part of our, I 24 believe cur mission and our responsibility, is to i
O
) 25 understand those influences on human performance, to
, NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE , N W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
55 1 understand those influences on facility risk. And what gs 2 we're trying to do here -- I think sometimes in the past -
! 1 3 - if I might -- we've been so concerned with the 4 regulatory vehicle that may happen, that we haven't 5 addressed the technical issue to really understanding what 6 these influences are. ;
7 This is what we're trying to do here, is to 8 understand those influences to help us reduce our 9 uncertainties in our decision-making, to be able to 10 improve our predictive capabilities, our trending, you 11 know, before something comes up and hits us in the face.
12 Now, whether or not we then take a step to 13 imposing some requirement, that's a whole other issue.
q t
k../ 14 But certainly we want to be able to provide to our 15 inspectors and reviewers, the information so that they can 16 recognize a contribution to human error if it's out there.
17 MEMBER POWERS: These programs are very broad.
18 This specifically talks about risk, and one of the things 19 we've gotten to the point with various research efforts 20 we've done in the management and organizational factors 21 area, was this issue of trying to accommodate 22 organizational factors in the PRA. So that's an element 23 of this.
24 MR. HODGES: And rightly --
()
25 MEMBER POWERS: It's also an element in terms NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 37.)1 (202) 234 4 433
1 56 ;
1 of one of the things that we've been asked to do, is to do l
1
,~
,. 2 some self-analysis. Within this program we're going to be !
\ )
3 taking a look at NRC activities, not only licensee .
I 4 activities.
5 MR. HODGES: And rightly or wrongly, the NRC 6 has been doing an assessment of management for years --
l 7 it's called the SALP process. 1 8 MEMBER SEALE: That's one of them.
9 MR. HODGES: That's one of them -- also the 10 senior management process. But I've been in the regional 11 SALP boards and I know what it's like to have to go
)
1 12 through that process and how ill-prepared I was to do some l
13 of that, at the time. I think if we can help even that 7-.
-) 14 process, we will have made progress.
15 MCABER BARTON: I agree with --
16 CHAIRMAN APOSTOLAKIS: Well, I --
17 MEMBER BARTON: I'm sorry, George.
18 CHAIRMAN APOSTOLAKIS: Go ahead.
19 MEMBER BARTON: I agree with you, Wayne, 20 because I look at this and I heard what Anne Ramey said 21 also. I think we need to understand the human performance 22 and the elements and where organizational factors fit in 23 and safety culture, etc. I think it would be a mistake to 24 regulate it.
,r-s
(. / 25 I think you need to look at the SALP process, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
57 1 senior managers process, and you need to adjust those 7s 2 processes to be more effective. But I think to go put s
3 regulations on them when you look at the performance of 1
4 the industry and you look at INPO indicators and NRC l
5 indicators, it says hey basically, industry is operating, l l
6 safety is improving, performance is improving. )
i 7 And there's some outliers out there and we're 8 cmbarrassed by some recent outliers. I think we need to 9 understand it and the NRC needs to get smarter also so the ;
10 outliers get -- you know, tweaked and pulled in before we !
l 11 have the issues like we do now. But I think, thinking 12 about regulations for this is probably a mistake.
13 MR. THOMAS: If I could put in my two-cent's !
77
% ) 14 worth on the subject. I have done a great deal of work 15 with IAEA and safety culture in developing guidelines to 16 assess safety culture based on NSAC-4. And one of the 17 fundamental things that comes out of that is, you cannot 18 regulate safety culture, and I think we need to admit that 19 up front.
20 But there's a difference -- there's regulation 21 in safety culture or management organization, and there's 22 regulation. And I'd like to just to share with you for 23 one minute, what we do in NRR.
24 It's important to know the causes of i,n) 25 performance that we see, and one of the things we do --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
58 1 and we have tools that we have developed -- is to, for
,w 2 example assess the effectiveness of licensee's root cause k)
3 determination programs, to give our own views for example, 4 on whether or not the licensee's corrective action 5 programs have indeed, determined that management factors, 6 organizational factors, safety culture, has contributed to 7 some performance problems that we're seeing.
8 And in that sense, we're not regulating, but 9 we're making sure that the licensee's own programs are 10 effective. And we use that a lot in NRR today. And I 11 think that's an important element of regulation that's far 12 short of having regulations that dictate culture, and so 13 on -- which are totally ineffective.
,e m s )
k/ 14 MEMBER BARTON: I agree.
15 DR. PERSENSKY: I think it gets back to Dana's 16 earlier point about the strategies. You know, how do we 17 define that third bullet? It's not just writing a new 10 18 CFR; there are other approaches that can be used to get to 19 the issue of improving, based on what we learn.
20 I guess one of the reasons we pull this out as 21 separate is because there has been so much focus on it and 22 it's such a broad issue -- it has risk components, it has 23 regulatory components, it has a number of different 24 components involved.
A kj 25 CHAIRMAN APOSTOLAKIS: Now, my understanding NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005 3701 (202) 234-4433
59 I i
1 is that the industry has been against this from day-1, and l
-~y 2 perhaps the reasons that they're afraid that there will be A_) l 3 some regulatory action at some point, you know, after this 4 goes past the research stage.
i 5 So perhaps communicating with the appropriate !
6 representatives of the industry that the idea here is not 7 to regulate would be a good thing. Because you cannot 8 have a major research program which the industry opposes 9 every step of the way.
10 At some point for example, you will have to go 11 to a reactor, the plant, right? And collect information 12 and so on. So if the industry official is against this 13 effort -- and I know firsthand, you know, even when we
! l x/ 14 were reviewing 1150 and we were writing the 15 recommendations for Federal research, there was a fierce 16 debate among the group, whether organizational factors 17 should be there.
18 And the industry representatives won the 19 argument, partly because some of us were still too young.
.0 But I think it's a miscommunication. They're afraid of 21 it; that we'll become a regulatory issue.
22 MEMBER BARTON: I think we'll find out, 23 George, because their comment period on this was up May 24 27th, on the safety conscious work environment. The y~
_) 25 subject's out for public comment on this right now. It NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
60 1 was --
,3 2 MS. RAMEY-SMITH: It's not us.
/ <
~~
3 MEMBER BARTON: -- OE, I think.
4 CHAIRMAN APOSTOLAKIS: OE?
5 MS. RAMEY-SMITH: Office of Enforcement.
6 CHAIRMAN APOSTOLAKIS: Oh, I was not aware of 7 it.
6- MEMBER BARTON: They're floating a document 9 out there right now for public comment.
10 MEMBER SEALE: Well, there may be a 11 considerable concern about regulation in this area, but I 12 think at least in organizations like INPO, there is a very 13 real recognition of the fact that management and
(~b e
(_/ 14 organizational influences probably have as much to do with 15 the problems in terms of plant performance, as anything.
16 And certainly the INPO people are very concerned about 17 these issues.
l 18 CHAIRMAN APOSTOLAKIS: I mean, every time 19 there is a public workshop where this issue comes up, the 20 industry immediately is against it, and I think you guys 21 ought to do a little bit of -- not quite lobbying -- but ;
I 22 talking to them and explain to them what the idea is here. i 23 Because the technical issues are never dis' cussed. 4 24 MEMBER SEALE: You need to reassure --
(_) 25 CHAIRMAN APOSTOLAKIS: It's one side that says l
NEAL R. GROSS l COURT REPORTERS AND TRANSCRISERS l 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASH lNGTON, D C. 20005-3701 (202) 234-4433
)
l 61 1 no, you shouldn't even touch it -- in fact, the last
!p- )
2 argument I had in Paris is, why do you worry about the 3 organizational practice here? When you build bridges, do 4 you worry about organizational practice? And that 5 derailed the argument, you know, for an hour. Now we're 6 talking about whether you worry about organizational 7 structures when you build bridges.
8 So I think it's a miscommunication and you 9 should get some of the key players like EPRI, NEI, and so j l
10 on, to buy into this and perhaps do their own research. I 11 DR. PERSENSKY: That's been suggested. This I 12 has been a -- several times we've gone through this kind I 13 of research in my history here, and it's been a cyclical
/~T.
\w/ 14 process. Right now we're at the peak, and it's been aimed I
15 at different aspects. It was aimed at regulatory tools at 16 one point; it was aimed at improving the SALP process.
l 17 The latest focus was more on the PRA aspects, i 18 but now because of the issues related to regulatory 19 effectiveness and regulatory excellence, it's become 20 again, an issue that's broadly discussed within the 21 agency, and there are a lot of people ir.volved with it.
22 AEOD is working on the senior management, 23 mentioned the work that Office of Enforcement is doing.
24 There is a new -- what's Martin's role?
/m (J) 25 MR. HODGES: He's the senior manager for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
62 l
1 regulatory effectiveness. ,
l 1
73 2 DR. PERSENSKY: Effectiveness. That's up in l
( )
'~',, 1 i
3 Jordan --
l 4 CHAIRMAN APOSTOLAKIS: Who's this? 1 1
5 MEMBER SEALE: He's talked to us -- Tom j l
6 Martin. I 1
7 DR. PERSENSKY: Tom Martin --
l 8 CHAIRMAN APOSTOLAKIS: Oh, yes, he talked to 9 us.
10 DR. PERSENSKY: -- for Ed Jordan. So he has ,
11 that function. We're trying to provide research, 12 technical basis for as much of that as we can, and 13 participate. So we're relying on things we've done in the
/ \
\d 14 past but also looking to do new things.
15 CHAIRMAN APOSTOLAKIS: Actually, yes. I think 16 what Jack said earlier, you know, looking at it as a 17 multi-level problem and how does organizational structure 18 affect human performance is the best way. Because if you 19 change the subject and you look for the best or a good 20 organization, then you never get anywhere.
21 DR. PERSENSKY: That has never been the goal, 22 I think, in all the time that I've been sort of I 23 tangentially involved with that -- it's never been the 24 goal.
y.m
) 25 CHAIRMAN APOSTOLAKIS: Have you read that NEAL R GROSS COURT REPORTER i AND TRANSCRIBERS 1323 RHODE I 3 LAND AVE., N.W.
(202) 234-4433 WASHINGTOfi, D.C. 20005-3701 (202) 234-4433 I
r
i 63 l l
l 1 paper by Todd LaPort or Barclay a few years ago, when he
.f3 2 found that -- he did some work with his colleagues on (s I
/
3 aircraft carriers which are a very high reliability i
4 organization. And he found that the way business is l
l 5 conducted there violated everything that organizational ;
6 theory said was the way you were supposed to run business.
7 Like, people would rotate every six months, 8 there seemed to be no clear assignment of responsibilities 9 and roles, and yet the aircraft were landing and taking 10 off reliably. So there isn't such a thing as a good 11 organization it seems to me. It depends on the particular 12 culture that prevails, and so on, so we shouldn't go --
13 MEMBER POWERS: There are some Navy studies on N s'\
14 the operation -- flight operations off aircraft carriers 15 and how to conduct them in a reliable fashion. I'm very ,
l 4
16 unfamiliar with them except they exist, and they're 1
1 17 portraits on how you set up a teamwork organization where I
18 there is not a hierarchy to ensure that the lowest radio !
19 operator is perfectly willing to tell the Captain of the 20 ship that he can't take the airplane off right now.
21 I mean, that's the way it's set up, is to 22 create a communicating structure that works well, by 23 eliminating hierarchy.
24 CHAIRMAN APOSTOLAKIS: But the reason why that O
( ,! 25 paper had such an impact was that it really found the case NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 2344433
64 1 where everything those guys were doing went against their
! .s 2 professional wisdom as to what a good and highly reliable
- s k
3 operation ought to be.
4 Maybe we should get going though, right?
5 Mark, do you want to say something?
6 MR. CUNNINGHAM: I was judt going to say in i
1 7 that respect, it probably relates too, to the clarity of j l
l 8 the mission on an aircraft carrier. There's a fairly i 9 clear mission that everybody knows, and as you get into l
10 organizations that aren't on aircraft carriers, the 11 missions become a little more diffuse and that changes how 12 organizations are --
,_ 13 CHAIRMAN APOSTOLAKIS: That must be part of
( ')
'v' 14 it.
15 MEMBER SEALE: There's one other factor and 16 that is, that if you have one of the airplanes turn into a 17 submarine, the Captain looks for a new job.
18 MR. CUNNINGHAM: Yes.
19 DR. PERSENSKY: The next program is actually 20 to develop -- continue the development to maintain and 21 integrate the amount of human performance and human 22 reliability which is the ATHEANA project that's been 23 pulled out again separately. It has two ATHEANA we're 24 using in two different ways here. One is a framework for
,es,
(_)' 25 this plan and the other is as a model for HRA work.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
65 1 The seventh program, some people have just g~s 2 said it's a ticket to travel, but it's really -- meant i \
\l 3 something that gets us more involved with the outside 4 world -- not only the nuclear world, not only in the U.S.
5 -- but there are a lot of international programs that l
l 6 you've mentioned here -- 1 7 MEMBER SEALE: This includes the Haddam 8 project and things like that?
9 DR. PERSENSKY: Yes. Things that we need to I
10 get involved with and know more about what's going on, 11 both in terms of identifying issues, but also identifying 12 fixes. I 13 MEMBER POWERS: Doesn't this follow directly l 14 from one of the direction setting issues? I believe 15 there's an --
16 DR. PERSENSKY: There is one --
17 MEMBER POWERS: -- international efforts, and 18 it seems like this explained --
19 DR. PERSENSKY: This goes beyond just the 20 international --
21 MEMBER POWERS: Then it looks like -- it has 22 language that just follows directly from it. I don't 23 think -- characterizing it as a ticket to travel -- I 24 mean, maybe that's true but the fact is, we have to now.
^
73
) 25 DR. PERSENSKY: Yes, we have to do it, wJ NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
66 1 especially based on the fact that we need leverage --
,- 3 2 financial leverage -- to do a lot of work. I think we've
('~' )
3 got a lot of it -- the Halden Project, for instance.
4 MEMBER POWERS: There's a lot more to mine out 5 of the Sizewell efforts.
6 DR. PERSENSKY: Yes, there are other things 7 that are going on as well in other applications. We're 8 trying to get involved with -- there's a technical 9 advisory group on human factors in the Department of 10 Defense. They get together every six months to talk about 11 all the various problems there.
12 I just had an opportunity to -- at Halden as a 13 matter of fact, invited the Director of the laboratory at
.rh i i
\~/ 14 Atlantic City for the FAA to present at Halden on their 1 l
15 development of a new laboratory. There's a lot of things l l
1 16 that he's doing there in the fat 1.ities he has that I 17 really wasn't aware of until recently.
18 CHAIRMAN APOSTOLAKIS: Who is this? I i
19 DR. PERSENSKY: It's the FAA Atlantic City l 20 Lab, Research and Development. But they have facilities 21 there and they're doing similar kinds of things. If you 22 go back down into specific activities like the question of 23 navigation between streams for --
24 MEMBER POWERS: A nice article.
[ h,
(_,) 25 DR. PERSENSKY: -- they're doing that, because NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
67 1 they're concerned with it for air traffic controllers and
,3- 2 for pilots.
! )
('~
3 MEMBER POWERS: A nice article in the 4 Scientific American for May, crew management on -- very 5 good.
6 DR. PERSENSKY: And that's human centered 7 automation is where that comes from. So there's a lot of 8 things that we need to be involved with. There's the CSNI 9 PWGs --
10 MEMBER POWERS: Let I ask you a question about 11 -- comes to mind from the-recent Academy study on digital 12 electronics. One of the implied criticisms in the written 13 part that was very explicit when they came in front of 3
kJ 14 this group to -- criticisms of the staff, is that they 15 didn't get out in the world enough.
16 You know, they didn't go to enough other 17 agencies, they didn't go to enough technical meetings, 18 they didn't present enough technical papers. Now, I 19 happen to think.they were wrong on that criticism, but you 20 have to take it to heart. I mean, this is a prestigious 21 panel of people that got together and looked at the 22 situation and they found it wanting.
23 Again, I think they were wrong. But it's
! 24 something you want to bear in mind when setting up your
( ) 25 program is that -- and certainly this task addresses half NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
68 1 the issue. The other half is participating in technical g 2 meetings and things like that. You need to bear that V
3 critique they have, 4
And the reason I give it any credence at all -
5 - because I do disagree with their finding, but I give it 6 credence, is that I've hear that criticism of NRC research 7 programs in the past in other contexts. I disagreed at 8 the time, but after you hear it five or six times from 9 different sets of people, you say, well maybe they're 10 hitting on something. And so I point it out just as 11 something that you may need to bear in mind in setting up 12 your program.
13 DR. PERSENSKY: To some extent that's why this
'w- 14 is in here as well. I disagree with that comment because )
l l
15 I think we're probably very well informed on what's going 16 on outside. I don't know that the people outside are i
1 17 necessarily as informed about what we're doing.
1 18 MEMBER SEALE: I think that's part of our --
I 19 DR. PERSENSKY: That's where we have perhaps a 20 problem, but as I said to somebody on that committee, I go 21 to ANS and IEEE meetings that are nuclear involved. That !
22 individual goes to IEEE meetings but they're on computer I i
i 23 software. It's a different one, but you usually focus -- !
1 l
24 when you give a paper you try to talk to the choir p
- V 25 unfortunately, because those are the people that want to i
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
69 1 hear what you're saying.
2 So the purpose here though, isn't just to
,f
(
3i x_/
3 travel, it's not just to make presentations, but to learn 4 from what's going on, as well as to perhaps, share that 5 information with others.
6 MEMBER POWERS: Recognize most people that i
7 criticize travel don't do it.
8 MS. RAMEY-SMITH: I want to, if I may, 9 reiterate one point that Jay made, speaking of the FAA. I 10 think that an awful lot of our travel that we need to do 11 is around the beltway. There's an awful lot of agencies !
12 out there that are very interested in the work that we're l
p 13 doing in human reliability assessment. 1 i \ l
\- 14 There's DOE, several different parts of DOE 15 that expressed a real keen interest in some of the work.
16 We've met in the past with folks from the National 17 Transportation Safety Administration, and the thing is is 18 that, you know, in days of unlimited budgets, you know, 19 maybe we could all do our own thing, but we don't have 20 that. None of the agencies in the government have that.
21 So maybe what we do is that we borrow from 22 some data that's available out there in the airline 23 industry to give us insights into the causes of human 24 error that helps in building our predictive models. Maybe m
i
(,)i 25 it doesn't give you a frequency count, per se, but it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
- 70 l
i 1 helps me, and in return we provide some insights from our !
l l
- ~3 2 research on the effects of culture, for instance.
'-] 3 So one of the things that we hope to put quite 4 a bit of effort into is to developing some informal 5 networks, if you will, of interest groups among different 6 agencies in the government, as well as talking with the l 7 industry groups.
8 CHAIRMAN APOSTOLAKIS: A good time to take a 9 break.
10 DR. PERSENSKY: Would you like to take a 11 break?
l l
12 CHAIRMAN APOSTOLAKIS: Let's say 1:50.
13 (Whereupon, the foregoing matter went off the
/ s
\- 14 record at 1:42 p.m. and went back on the j l
15 record at 1:57 p.m.)
16 CHAIRMAN APOSTOLAKIS: Okay.
17 MR. PERSENSKY: Okay.
18 As we had indicated, part of this effort that l
19 we're going through is to try to structure the program, l 20 both in terms of its organization, but also have a 21 structure or framework which we can use as a basis of sort 22 of making decisions as to what belongs here and what i
23 doesn't, what impact does a particular program or effort !
24 have.
,m
!, ) 25 (Slide) s.-
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
71 l
l 1 MR. PERSENSKY: ATHEANA has been selected as I
.,_s 2 the framework.
i \
~
3 MEMBER POWERS: When you say " selected," is -l 4 that a judgment that somebody made or did that go through 5 some formal trade study? 1 6 MR. PERSENSKY: I think it was more of a !
l 7 judgment that we made because it was an existing tool that 8 seemed to cover the elements that we had. It's also in a 9 development stage. So it's not something that's 10 necessarily frozen that we can take it and use it.
l l
11 Again, we're trying to respond to your l l
12 comments, the ACRS comments, of having some sort of I 1
13 structure. This seemed to have the elements that we i
(^' '
l
(_)s 14 needed, but we didn't go through a formal trade-off study.
l 15 MEMBER POWERS: Is that going to be a general 16 hallmark or to not do formal trade studies?
17 MR. PERSENSKY: No. I think that we will be 18 doing various -- depending, again, on what the need is, in ;
19 a way, it had been developed through some more formal 20 trade studies but not necessarily for this purpose. It 21 just fit this purpose.
l 1
22 CHAIRMAN APOSTOLAKIS: I mean, you can always 23 modify ATHEANA for the present purposes.
24 MR. PERSENSKY: Well, as I say, one of the
(~h 25 elements is, in fact
( )
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
l I 72 l
! 1 1 CHAIRMAN APOSTOLAKIS: Because ATHEANA has a
.,qa 2 specific goal.
N~'J l 3 MR. PERSENSKY: Its goal was to look at errors j 1
1 4 of commission --
5 CHAIRMAN APOSTOLAKIS: Yes. 1 1
6 MR. PERSENSKY: -- primarily so that they l 7 could be added or better structured into an HRA. This is l
1 8 not a briefing on ATHEANA. I just want to make clear I l 9 just wanted to give you an example of what it is. I think i
10 the elements that are most important in this model as it's l 11 laid out is that it talks about within the dotted lines l 12 here an error-forcing context. That was something that i
13 was kind of new I think to the concepts that have been i i(
(./
\i 14 used in HRA in the past.
i 15 These are the kinds of things that get into 16 the performance-shaping factors. How does training affect 17 this? How do procedures affect it? But also it puts some 18 of the limiting conditions of the facility, includes what 19 is going on in the facility. These things together are 20 what brings about some sort of human error.
21 If you have a degraded condition, you have 22 poor procedures, it leads to some problem. That problem 23 is going to be expressed through some cort of error 24 mechanism, which is generally a cognitive mechanism in n
( -)
m 25 this model. It talks about issues such as situation NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
T l 73 i 1 awareness, attention, the kinds of things that are going l
,y 2 on in the individual.
(
x_-)
3 An individual takes some action. An action in 4 the sense we're defining as an unsafe thing. It's 5 something that's had an impact or will have an impact back 6 on the facility.
7 CHAIRMAN APOSTOLAKIS: So the error mechanisms 8 are never observab3e?
9 MR. PERSENSKY: Right. It's their outputs.
10 CHAIRMAN APOSTOLAKIS: It's just the 11 hypothesis?
12 MR. PERSENSKY: Right.
13 CHAIRMAN APOSTOLAKIS: Yes.
,/ m k/ )
14 MR. PERSENSKY: But then it goes into the FRA 15 modeling, which is the more standard. One of the things 16 about this that allows for things that -- you know, this 17 is the focus of risk-informed, performance-based, but 18 there's also this error that comes along the top of the 19 dotted lines.
20 There are certain things that are going to 21 happen. We know they're going to happen. They are things I i
22 that are happening right now that we're dealing with in l l
23 terms of development of guidelines because there are tools 1 l
24 that are needed, tools that are needed yesterday. And l
/~N !
(_),
25 those may not go through this whole framework. So there NEAL R. GROSS COURT REPORTERS AND TRANSCRIDERS 1323 RHODE ISLAND AVE., N W.
j (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
74 1 is sort of an external process that goes along with it.
fS 2 As I said this is not intended to be a
, l L/
3 briefing on ATHEANA. If we want more information, we can 4 provide a more detailed briefing somewhere down the line, 5 but --
6 CHAIRMAN APOSTOLAKIS: Yes. We should do l
7 that. l I
8 MR. PERSENSKY: You all have copies of the !
1 9 documentation on it. It's in development. I '.1 fact, two 10 weeks, they're going to be testing it as an HRA model at a l 11 power plant. No. I think it's in July. l 12 MS. RAMEY-SMITH: Yes. It's in July in 13 Seabrook.
rh s
\ ~' 14 MEMBER BARTON: Seabrook?
15 CHAIRMAN APOSTOLAKIS: Would the fall be a 16 good time, Ann, to have a briefing on ATHEANA? Again, 17 that is work in progress, not a finished product.
18 MS. RAMEY-SMITH: Mid fall would probably be a 19 very good time. What you want to do is you want to wait 20 until the pilot application at Seabrook is over and the 21 lessons learned have been discussed and so on.
22 CHAIRMAN APOSTOLAKIS: Now, I have a couple of 23 comments here. First of all, nobody has objected to this 24 word " forcing," error " forcing" context? Shouldn't you be i (_
(")N 25 using something milder or --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
75 1
1 MR. PERSENSKY: This is something I inherited. I i
,x 2 So it's either Mark or --
( )
3 CHAIRMAN APOSTOLAKIS: -- error in using or 4 error-enhancing or something? " Forcing" is kind of 5 strong, isn't it?
6 MS. RAMEY-SMITH: Well, I guess you can look 7 at it that way. The intention, of course, with this is to 8 get the idea across that sometimes when you have a 9 combination of certain performance-shaping factors in i
10 plants, that the personnol or the operator in this case is 11 set up, has a very high likelihood of error in those 12 circumstances.
13 CHAIRMAN APOSTOLAKIS: No. I understand the
~- l 14 intent. It's just that the word " forcing" sounds a little 15 bit too strong because, after all, if you look at Browns 16 Ferry, for example, I mean, it goes way back, but the 17 operator has really acting in an admirable way, even 18 though the context was really pretty bad. They had the 19 fire. They didn't know what to do.
20 So the operators do not always do their own 21 thing because the context is bad.
22 MS. RAMEY-SMITH: Exactly. And that's the 23 reason why you don't -- error-forcing context, the 24 probability of error doesn't become one. It becomes a
) 25 high probability or a higher probability, as in that case.
HEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
76 1 They succeeded.
a 2 MR. PERSENSKY: And that's the balance where i
3 you include both the facility conditions and the 4 performance-shaping factors.
5 CHAIRMAN APOSTOLAKIS: No. I agree.
6 MR. PERSENSKY: That's how it comes together i 7 and balances that out.
s 8 MEMBER SEALE: The individual is almost the 1
9 victim in this case, rather than the perpetrator?
10 MR. PERSENSKY: Yes. ,
1 1
11 MR. CUNNINGHAM: Yes.
12 CHAIRMAN APOSTOLAKIS: Now, if you want to 13 expand this a little bit to cover for the purposes of the
,m i :
C) 14 HPPP, not for the purposes of the original in terms of 15 ATHEANA, I really like that figure that is in Jim Reason's 16 book, 7.9, which we also attached to our comments, the 17 comments of December 30th to Mr. Taylor, because it makes 18 it very clear. You know, you have the unsafe practice, 19 psychological precursors, the line management 20 deficiencies, and the decisions. And perhaps some of it 21 you can attach there to the performance-shaping factors.
22 For example, the line management deficiencies 23 may be fed into the performance-shaping factors. In that l
24 way, you will cover all seven -- well, not foreign travel
,a (L.J ) 25 -- all six programs.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
77 1 MR. PERSENSKY: Travel is for all of them.
g 2 CHAIRMAN APOSTOLAKIS: Yes. And adapt this to
(. i
'")
3 the present needs, although I'm sure the guys who were 4 de"aloping this had some of this in mind as part of the 5 formal --
6 MR. PERSENSKY: This package I believe has a l
7 lot to --
i 8 MS. RAMEY-SMITH: Right.
9 CHAIRMAN APOSTOLAKIS: Yes, yes. I know.
l 10 MS. RAMEY-SMITH: Actually, Dr. Reason was i 11 part of the --
12 CHAIRMAN APOSTOLAKIS: I know.
13 MS. RAMEY-SMITH: -- initial team to work on
(
\s) 14 this, as you know. Yes, the thing is that all this is, 15 it's a pictorial representation of the process. And there 16 is a great deal of depth here.
17 So when you look at performance-shaping 18 factors, most of the people in this room are familiar with 19 HPIP. And the discussion of the human performance 20 investigation process has a lot of information in there 21 about performance-shaping factors. Part of what we're 22 looking at in management organization in a sense is to 23 look at management and organization as a l
24 performance-shaping factor or doers it really belong n
t l 25 there?
LJ l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHlNGTON, D C. 20005-3701 (202) 234-4433
78 1 Is it an error mechanism that starts getting
-- 2 into motivation and so on? We don't know. We know that N] 3 it's somewhere in there.
4 CHAIRMAN APOSTOLAKIS: My point is this. I 5 .
mean, I like the figure. It depicts pictorially what we 6 really want to do. So if you go back to your six 1
7 programs, it would be nice to modify this a little bit and 8 say Program Number 4 comes to this bcx or to this box. It q l
9 may affect more than one box. I i
lo Because I'm sure a lot of the stuff is already ;
1 11 hidden in here. But since you have explicitly identified 12 six programs, that's why I'm saying that maybe by adapting 13 this to that, it will be easier to communicate what their
/-
\-) 14 high-level model was.
15 MR. PERSENSKY: That, in fact, is one of the 16 things that we're looking into.
17 MEMBER POWERS: I'm not sure I would do it 18 that way. I think I would come in and say, "I have 19 identified seven functions." Now, those functions have 20 requirements, and it's the requirements that map onto this 21 plot, not the function.
22 CHAIRMAN APOSTOLAKIS: Yes, the requirements.
23 But why do I have a Program Number 5, " Develop the 24 technical basis and guidance of management organization"?
,i-~
() 25 As I go to this figure and I see a box that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
79 1 feeds into performance-shaping factors and possibly error i
.pe~) 2 mechanisms, that says " management and organization." Then l N /l \
3 I have justified my Program Number 5. I 4 MEMBER POWERS: Now, I think they have -- l l
5 CHAIRMAN APOSTOLAKIS: And I know where it 6 feeds into the big model.
7 MEMBER POWERS: I think they have to justify l 8 the functions from the mission statement. .
l 9 CHAIRMAN APOSTOLAKIS: The mission. Let's go 10 back to that.
11 MEMBER POWERS: The mission and the mission 1 12 needs have to lead to the function. j l
13 CHAIRMAN APOSTOLAKIS: The strategy is what ,
s 14 you --
15 MEMBER POWERS: What they call --
16 MR. ROSENTHAL: Top down.
17 CHAIRMAN APOSTOLAKIS: Yes, but it doesn't 1
18 really go into details of this nature. I mean, yes, the l 19 mission -- well, the whole picture helps you do this, but 20 the mission --
21 MEMBER POWERS: Yes, but the mission 22 -identifies what the mission is. I mean, it does exactly ;
23 what it says it does. Their mission need, which is a 24 strategy, is what they need to carry that out. And the f
(3),
25 functions are the functions that have to be -- functions NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W, (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
80 1 are what they call programs.
m 2 CHAIRMAN APOSTOLAKIS: Yes.
()'_
3 MEMBER POWERS: -- are what they need to 4 achieve this mission need.
5 CHAIRMAN APOSTOLAKIS: Right.
6 MEMBER SEALE: And ATHEANA is the integrator.
7 CHAIRMAN APOSTOLAKIS: And what I am saying is 8 --
9 MEMBER POWERS: ATHEANA is a framework for the 10 requirements.
11 MEMBER SEALE: Yes. That's how you put it all 12 together.
13 CHAIRMAN APOSTOLAKIS: Yes. So why do I need 4
\- 14 to have a Program or Function Number 5 because that feeds 15 into the performance-shaping factors, the error-forcing 16 context? That's what I'm saying.
l 17 MEMBER POWERS: I think you're going at it the 18 wrong way or you've reorganized. I think they have a 19 Program 5 because they want to develop knowledge of the l 20 causes and consequences, in this case causes, of degraded 21 human performance and reliability.
22 That's the mission need. And the function to 23 meet that mission need is Program 5.
24 CHAIRMAN APOSTOLAKIS: Okay. But this
,\
' ( ,) 25 degraded performance occurs because of what? Because of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
81 1 the PSFs. That's where it comes into the picture. That's
,. 2 a high-level model that tells me why I need this activity
(
3 to help me understand human error.
4 MEMBER POWERS: Then this should intercede in 5 between strategy and function.
6 CHAIRMAN APOSTOLAKIS: Well, all I'm saying is 7 it would be nice to use a picture like that to justify the l
8 program.
l 9 MS. RAMEY-SMITH: Okay. If I can make -- ,
1 1
10 CHAIRMAN APOSTOLAKIS: So we want to rename 11 them is a different --
12 MS. RAMEY-SMITH: Again, we adopted this 13 framework because we did see it as a good integrator. And
,~
! )
k_/ 14 what's driving what? You know, is the model driving our 15 programs or our programs are consistent with our mission l 16 in this framework?
17 We have taken some approaches to looking at 18 how to organize our documents that we're working on. Do l 19 we organize them according to this HRA model or do we show 20 the relationship so that when you see the integration, you 21 see how things are integrated?
22 Now, we actually did have another version of 23 this that showed the programs and listed which program 24 seems to be primarily related to which area. We made a
) 25 decision, perhaps faulty, not to include that here today NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C 20005-3701 (202) 234-4433
82 1 because we knew that we were kind of going through a 73 2 process of reorganizing our programs and stuff. And we
()
3 were afraid that it would just cause confusion down the 4 road.
5 But it's a point well-taken. We need to show 6 the nexus between this integrator and to show how this 7 framework is an integrator between --
8 CHAIRMAN APOSTOLAKIS: Yes. For example, 9 let's take another program, Program Number 3, " Provide 10 human performar a evaluation and human reliability support 11 to inspection ar.d . eview activities for materials 12 licensees."
13 So now if this is my overall model, where does 5
/
\/ 14 Program Number 3 come into the picture? I mean, 15 inspection. I don't see any inspection there, really.
16 You want to have low probability of failure; right?
17 So you have to choose now the boxes where you 18 have something to inspect. And you may decide to inspect 19 something that aids the error-forcing context. You may 20 want to inspect something that's a precursor to an unsafe 21 action.
22 I don't know that you can inspect the l
23 mechanisms. Probably not. You can certainly inspect the 24 facility conditions, but that's not part of the human
/s is ,) 25 performance program; right, unless you can tell them NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
l I
83 1 something they haven't thought of?
f3 2 MR. PERSENSKY: As it relates to human t 1
(,/
3 performance.
4 CHAIRMAN APOSTOLAKIS: Yes, as it relates to 5 human performance; for example, the length of the cord of 6 the telephone; right? This is really what I have in mind 7 with all of these comments, trying to see at a high level 8 why I'm spending money on each one of these programs.
9 And, also, having this high-level picture, you 10 get feedback from it and say, "Okay. I'm talking here 11 about inspection." What exactly do I mean? Where can I 12 go and inspect? And you may discover that there are areas 13 we are not inspecting right row from this perspective.
(~h v d
\' ~
14 So I don't know if we're using the right 15 words, Dana, but it seems to me that something like that 16 would be very, very useful both to convey to the rest of 17 the world what you are trying to do, but also it would 18 help you focus these programs a little maybe better. I 19 don't know. Maybe a --
20 MEMBER POWERS: I suppose we go into this 21 discussion with the ground rules that they're not done 22 yet.
23 CHAIRMAN APOSTOLAKIS: No. I know. This is 24 just a suggestion. This is not a criticism.
O
( ,) 25 MEMBER POWERS: You just kind of have to guess NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE _, N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 84 1 where they're going. But it seems to me that maybe the 2 way they go is they have these seven top-level functions
', w) x_/
3 that they will now define under each one of those 4 functions another set of functions and then define the 5 requirements.
6 And then I agree with you. If you're going to i l
7 go to a lower tier of functions, having a great 8 integration scheme on why you need those additional tier l 1
9 of functions might be very useful. j 1
10 CHAIRMAN APOSTOLAKIS: I don't have any 11 objections to that. That seems to me even the seven 12 program areas as stated now would benefit from an attempt 13 to place them in this context and also adapt this figure
/ s I
' 14 perhaps to this.
15 Now, I don't know Program 2, for example, like 16 take Program 1, " Conduct operating events analysis and 17 database maintenance to support human performance 18 evaluation and reliability analysis."
19 It seems to me that most of the activity in 20 this area, the existing activity, deals really with human 21 failure events. Most of the evidence comes from there; 22 right?
23 MR. PERSENSKY: Human failure events are 24 really the PRA-based --
,, x_
() 25 MS. RAMEY-SMITH: Can we --
NEAL R. GROS $
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
85 1 CHAIRMAN APOSTOLAKIS: Well, the unsafe
- 2 actions, r i k~ / 3 MS. RAMEY-SMITH: Can we back up for just a 4 second, if I may? First of all, we agree that we need to 5 show this relationship. I'll tell you what my current 6 thoughts are, which are different than my previous 7 thoughts and probably will change. And it hasn't been 8 discussed with other folks.
9 So, with all of those caveats in mind, one way 10 that you can look at this, at this model or at this 11 framework, is I can think -- it requires that you stand 12 way back and look at it.
13 And I can say I have a mission to do (S i
%> 14 something. And I've set up some programs to implement 15 that. And how does that relate, then, to this overall 1
16 vision of human performance and how it fits into the 17 regulatory framework, which is the way I look at this 18 model?
19 And I can say all of my research and analysis 20 activities are supporting my understanding of everything 21 that's in those dotted lines, those boxes that are inside 22 that.
23 And I can think of my review and inspection 24 activities as supporting my risk management decisions.
i ,em l
(v) 25 Now, it doesn't say that they're separate, but you see NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
86 1 that looking at the framework, those research and analysis 2 activities support the review and inspection activities.
' i 3 So from a general perspective, that's where they fit in.
4 Now, if I look at a particular activity -- and 5 let's say it's develop the technical basis to support 6 these things -- it may be that I make an assessment that I 7 understand the impact of these performance-shaping 8 factors. And we got that down.
9 We know about that. What I don't know about 10 is this relationship between the facility conditions and 11 the performance-shaping factors. And so then I can use 12 that as a way to prioritize my future work.
13 Do I want to keep getting smarter and smarter
_) 14 about what I know or look at something different, maybe in 15 a different kind of circumstance? The error mechanism has 16 been shown in events assessment to be really unclear. Why 17 did they make that error? I don't understand why. And so 18 , then we may choose to pursue a look at what's driving 19 certain kinds of errors and so on.
20 But I think what you do is you look at those 21 dotted lines as that's where your uncertainty is, that's 22 where your lack of knowledge is. And we're trying to get 23 smarter in those areas.
24 The output of all of that is how we review,
\
(~h
, , ) 25 how we inspect. And that feeds it back into how those I NEAL R. GROSS j COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N W. 1 (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
87 1 plants are operated. And the process goes on.
7.
2 CHAIRMAN APOSTOLAKIS: And I agree with that.
('~') 3 What I'm suggesting is really contributing to what you 4 just said. All I'm saying is make it a bit more explicit 5 and show how these programs -- and I started saying, fcr 1
6 example, in Program 1, " Conduct operating events )
i 7 analysis." l l
1 8 It seems to me that most of the work right now j l
l 9 in this area deals with unsafe actions. And we have a lot 10 of evidence that the organizational dimensions that 11 contribute to unsafe actions can be identified. For 12 example, organizational learning seems to be a major 13 problem with most facilities.
\
Y/' 14 So now I'm looking at this program. And I'm 15 doing exactly what you said, Ann. I'm saying, " Gee, I'm 16 really doing a lot regarding the unsafe actions, but I'm 17 not doing enough on the performance-unaping factor. Maybe 18 I should increase that activity."
19 MR. HODGES: And, in fact, one change that 20 I've already made as a consequence of looking at it in 21 that perspective is I've said if I'm going to be really 22 looking at error-forcing contexts, I need someone who is a 23 very good systems person to sit in with the projects 24 people to try to understand what's going on and
/O
) 25 particularly if we're going to try to look for precursors NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 t
88 1 to some of this stuff. And that's why we've got Roy Woods 73 2 in the group.
\' ')
'~
3 CHAIRMAN APOSTOLAKIS: Yes, yes, absolutely 4 right. But my suggestion is really to expand this 5 error-forcing context, to expand it for the purposes of 6 this project, not for ATHEANA, because that would allow 7 you to show why these programs are useful and then do the 8 things that Ann mentioned. And that will be a great way 9 of communicating.
10 And then, you know, we can polish the details, 11 but as a first step, I think that would be -- because 12 there is a hell of a lot that's hidden behind this 13 error-forcing context there.
- )
i/ 14 MS, RAMEY-SMITH: So you're saying expand the 15 illustration?
. 16 CHAIRMAN APOSTOLAKIS: The illustration, yes.
17 Oh, I'm sure you guys have thought about it, the 18 illustration.
19 MEMBER SEALE: Well, the operating events 20 activity really provides the truth, if you will, as it 21 applies to the model. It defines. It clarifies the 22 facility design, operation, and maintenance context in 23 which the event occurred, tells you what the facility 24 conditions are and so on.
(n) 25 That first one is clear. I mean, it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3'701 (202) 234-4433
89 1 essentially emplaces your problem on this model because fm 2 you wouldn't know a lot of the details that influence the
\
G 3 event unless you had what's in that first program element.
4 CHAIRMAN APOSTOLAKIS: Yes. Now, again, in 5 the interest of adapting this to the present needs, do we 6 really need to call out the PRA model there? Do you think 7 we need that?
8 MS. RAMEY-SMITH: In the spirit of 9 risk-informed regulation, going back to my previous job, 10 absolutely because, as you know, human reliability is a 11 major source of uncertainty in our decision-making 12 process.
p_
13 CHAIRMAN APOSTOLAKIS: Well, the only thing is
/ \
'd 14 people, -- well, although we don't have to go by that --
15 some people, when they see FRA, they think that's PRA, you 16 know.
17 I know what you mean, and I agree. But this 18 is not really PRA-oriented. I mean, PRA, you have to do 19 what's on the left first in order to do the PRA anyway.
20 Is anybody else around the table bothered by 21 the word " forcing" as much as I am?
22 MEMBER BARTON: Not really.
23 CHAIRMAN APOSTOLAKIS: Maybe it's only me, but 24 I think it's too strong.
,O l,.
) 25 MEMBER SEALE: Well, I know of events where I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
90 1 can truly say that the human that was involved was a 2 victim of the system.
/-~'N
\'" /
3 CHAIRMAN APOSTOLAKIS: That's very true.
4 MEMBER SEALE: And to that extent, I think 5 " forcing" is an absolutely appropriate word.
6 CHAIRMAN APOSTOLAKIS: But there are also 7 contexts where you would say the context has forced him to 8 make a mistake and they acted brilliantly. As I said, 9 Browns Ferry is one. I am sure there are others that 10 don't come to mind right now.
11 So " forcing" has the -- I don't know. Isn't 12 it too forceful? Well, if you are not bothered --
13 MR. PERSENSKY' Is it disproportionately
!n\
k/ 14 forceful? is the question.
15 CHAIRMAN APOSTOLAKIS: That's okey.
16 MEMBER POWERS: No, no, no. Unduly forceful.
17 MR. PERSENSKY: Unduly forceful.
18 CHAIRMAN APOSTOLAKIS: Disproportionately.
19 MS. RAMEY-SMITH: It does, however, get the 20 idea across, I think.
21 CHAIRMAN APOSTOLAKIS: Yes, yes. But you 22 don't want the idea to be --
23 MEMBER SEALE: And it's right on.
24 CHAIRMAN APOSTOLAKIS: -- that if you have ps, 25 that context necessarily, you will have them. I mean,
()
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
91 1 that's --
l 7s 2 MR. CUNNINGHAM: I think in the interactions !
I )
\ /
3 we've had on this, what's been more important is getting 4 in a sense of creating the image that Dr. Seale was 5 alluding to that you're setting up an operator, setting up ,
1 6 somebody. And I think that conveys a very powerful image.
7 In fact, it's very useful.
8 MEMBER SEALE: You could put a parenthesis i 9 after " forcing" and say " depending upon organizational and !
10 management," parenthesis close. j i
11 CHAIRMAN APOSTOLAKIS: In the interest of l 12 brevity; right? i 13 MEMBER SEALE: Right, in the interest of k/ 14 brevity because that's what happens. ;
15 CHAIRMAN APOSTOLAKIS: There must be another 16 word in the English language that conveys the message.
17 MR. PERSENSKY: At this point, as far as 18 ATHEANA is concerned, this is the ATHEANA model. We'll 19 look at it from the standpoint of how we're using it in 20 the framework, but it seems to me that the message comes 21 across using that term very clearly.
22 MEMBER BARTON: Sure does.
23 MR. PERSENSKY: And I think that's what the 24 goal was when it was developed, that it is a setup. If we f s 25 didn't use setup or you guys didn't use setup, it didn't
()
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 92 l
1 ceem appropriate. That's what it's intended to say.
I l
,s 2 (Slide) l i ) l 3 MR. PERSENSKY: I understand looking at the i
4 packet here, my last slide is slipped in between here.
1 5 What I wanted to show just real briefly here are a couple 6 of examples of as the plan is being developed, where we're 7 going with this. This table is very similar to the kind 8 of thing --
9 CHAIRMAN APOSTOLAKIS: One thing before you l 10 start on that one. We discussed in our letter to Mr.
11 Taylor this issue of formal and informal culture and that 12 how things get done in a plant is not always what, you 13 know, the books say. If we go back to the previous n
~
14 picture, the figure in ATHEANA, where would that be now?
19 You mentioned pilots. I looked at some 16 accidents involving pilots in the past. Invariably, they 17 skip steps and the checklists and so on. There is always 18 some element of pilot error. And this is part of the I
19 informal culture. You know, they don't want to go through j 1
20 the whole list and so on. So that's a practice that's l
21 everywhere, not just in the nuclear business.
22 Now, where in these boxes would I be forced to 23 think about that? 4 l
24 MR. PERSENSKY: From my interpretation, one
/~T I
( ) 25 would be performance-shaping of the factors, but it's the 1
~-
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234 4433
93 1 relationship between these two and also the facility
, -~s 2 design because that gets into the issue of operations
! \
'~
I 3 manual. This is where the history of the thing --
4 MR. ROSENTHAL: I just want to point out that 5 there was a strong contextual context within ATHEANA. The 6 argument is that you don't take like an ASEP model of 7 human reliability. You don't divorce it from what's going 8 on in the plan.
9 CHAIRMAN APOSTOLAKIS: No. And I agree.
10 MR. ROSENTHAL: And I think that that's --
11 CHAIRMAN APOSTOLAKIS: But that's the scenario 12 there, scenario definition. They mean the PRA scenario.
13 MR. PERSENSKY: Yes.
,/^\,
14 CHAIRMAN APOSTOLAKIS: They don't mean the 15 actual reality in the --
16 MR. ROSENTHAL: But in an event, I mean, it's 17 how fast things are going and what else is going on does 18 influence whatever he is going to do.
19 CHAIRMAN APOSTOLAKIS: That's very true.
20 MR. ROSENTHAL: In trying to take apart, 21 analyze why did something happen, you have to know the 22 physical plan context in order to do a good job that 23 explains --
24 CHAIRMAN APOSTOLAKIS: But people taking
(~'N,
(,) 25 shortcuts, for example, where would that be there?
l NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 l
1 94 1 MEMBER SEALE: And it seems to me --
f- s 2 CHAIRMAN APOSTOLAKIS: Unsafe actions? ,
i \ \
%-l j 3 MEMBER SEALE: -- you've got to find what goes
)
4 under error mechanisms. That's the mechanism by which the ;
1 5 error was perpetrated.
6 CHAIRMAN APOSTOLAKIS: Well, I don't know.
7 MEMBER BARTON: What do you say?
8 MS. RAMEY-SMITH: I'm hearing a couple of j l
9 different questions. And every time you ask it a little 10 bit different, then I'm pointing at a different part of :
1 1
11 the box. l l
12 If I want to understand why somebody did 13 something, then obviously the part -- well, not obviously, 7._
(' ' '
)
14 but the part that I'm looking at is I'm looking at the i 15 error-forcing context in the human error portion of this.
16 If I'm trying to decide, if I'm just trying to 17 name it, the operator did this, then I'm going to -- yes, 18 I'm going to be looking at that unsafe act. And then, in 19 actuality, the way I'm going to apply this, I'm going to 20 go backwards and go, "Okay. This happened. Now, why the 21 heck did that happen?"
22 And that's where your human factors expert --
23 you know, they have much of this knowledge available to 24 them. So they know people often, for instance, make these f"N
( ,) 25 kinds of mistakes, such as missing a step, because they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
95 1 lost their space, there was poor ergonomic design of the
,- s 2 procedures themselves or whatever, and they can start i !
'~
3 going back and understand the error-forcing context 4 authority.
5 MR. THOMAS: Maybe, say --
l 6 CHAIRMAN APOSTOLAKIS: It ultimately comes to 7 the error-forcing context, yes.
8 MR. THOMAS: Maybe to use an example that we 9 see in NRR quite frequently, it involves operators messing 10 up and using procedures. And we would look at procedures 11 as part of facility operation, design and operation.
12 The procedures themselves are something that's 13 tangible, something that's fixed. And it's, for example, S- 14 how they're used, how it enters into the model. And I
)
15 would suggest maybe a couple of places. i i
16 One of the things we look for when we go out j 17 in the field is whether or not management has clearly 18 articulated its expectations on procedure usage, for 19 example.
20 Are you expected to strictly adhere to the 21 procedures or has the culture relaxed to the point that 22 work-arounds are common, that operators -- not strictly 23 following procedures is condoned and that sort of thing?
24 And that would be more of an environment in the
'(d )j 25 error-forcing box.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON D C. 20005 3701 (202) 234-4433
96 1 So I would suggest, for example -- and this is 7- 2 errors resulting from usage of procedures -- several boxes
'~
3 could be implicated. You need to really break the problem 4 down. It's largely event-specific. l l
5 CHAIRMAN APOSTOLAKIS: Yes. You're absolutely ,
i 6 right. But at some point, don't we have to decide, 7 though, whose fault is it, what is it that we should fix:
8 the procedures or the operators? Is that part of the risk 9 management decision?
I 10 MR. PERSENSKY: Yes, it is, but I think we're l 11 trying to stay away from " fault," that that's --
]
12 CHAIRMAN APOSTOLAKIS: Well, I mean, if you're 13 fixing something, that means it's not ideal.
r~%
! \
\/ 14 MR. PERSENSKY: Yes.
15 CHAIRMAN APOSTOLAKIS: Yes. It's not ideal.
16 MR. PERSENSKY: It was the operator's fault.
17 It was because he didn't have sufficient raining, he 18 didn't have sufficient support, the procedures.
19 CHAIRMAN APOSTOLAKIS: Yes.
20 MR. PERSENSKY: You have to look at that both 21 in terms of whether or not --
22 CHAIRMAN APOSTOLAKIS: Should the goal here be 23 not to have an informal culture, to eventually have one 24 culture?
i
/"'1 / 25 MR. HODGES: No.
N./
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344 433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
97 1 MR. PERSENSKY: I don't think that's --
,_. 2 CHAIRMAN APOSTOLAKIS: Is there something i
/
3 wrong with it? I mean, why do people take shortcuts?
4 MEMBER BARTON: Many reasons, George.
5 CHAIRMAN APOSTOLAKIS: Okay. And these 6 reasons cannot be eliminated?
7 MEMBER BARTON: I don't think you can entirely l
8 eliminate them. I mean, there are times when the '
9 procedure is right and the operator has been trained and 10 he's a good operator. On a certain day, he makes a ,
1 11 mistake using that same procedure that he's used over and l i
1 12 over. You know, how do you explain that one? I 13 CHAIRMAN APOSTOLAKIS: But that's not an l 14 intentional thing, though, if you made a mistake. Taking 15 a shortcut is intentional. There's a big difference 16 there.
17 MS. RAMEY-SMITH: And in that example, what 18 you're doing is you're starting to think about those error 19 mechanisms. In one case, if you make a slip, you just 20 missed it and you knew. That's a different kind of a 21 problem to fix than if it was something that I decided 22 today and yesterday and the day before that I'm not going 23 to do this step because I don't think it's important, 24 different error mechanism.
7.-
() 25 MEMBER BARTON: Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
I (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
, 98 1 CHAIRMAN APOSTOLAKIS: Well, I mean, I 1 3 2 understand that you cannot really eliminate completely I )
NJ' l 3 informal actions people take, but at least maybe we should 4 try to eliminate the major reasons for the discrepancies, 5 like your own work.
6 I mean, there is some work that has been done 7 at Westinghouse for the simulator exercises. And the 1
8 operators who violated the procedures were, in fact, right ;
1 9 for the particular sequence reaction. l 10 MR. PERSENSKY: The procedure didn't cover 11 that particular scenario.
12 CHAIRMAN APOSTOLAKIS: Yes. But, I mean, how 13 do you fix all of that stuff?
-s 14 MR. THOMAS: We're talking about individual l
15 errors and individual events and so on, but I think the l 16 way we regulate, we look more broadly. For example, prior 17 to TMI, there were no guidelines really used uniformly j l
18 throughout the industry on development of emergency l 19 operating procedures. And subsequent to the TMI, we came 20 out with --
l 21 MEMBER BARTON: Volumes.
22 MR. THOMAS: Yes, very hopefully 23 well-thought-out human-factored ways to approach emergency 24 operating procedures, which we felt were the most
/~'T l
(,/ 25 important. And I think we minimized a class of human ,
i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
99 1 error by that effort.
2 And I think that's the approach that we take, i
, w) 3 We monitor performance through reviewing LERs and l 4 inspection reports to see where the problem areas are, not i 1
5 individual errors but where the problem areas are and try l l
6 to see what needs to be done or what can be done to try to 7 deal with them as a whole. At least that's the approach 8 we try to take.
9 ' MEMBER BARTON: Because it's different, l
l 10 because sometimes procedures, sometimes training, 11 sometimes various and sundry -- l 12 MR. PERSENSKY: What we've included in the
.s 13 packet are some examples of a chart that will be included l
/ \
' 14 in the plan which has each program area, the objectives of 15 that program area that gets to that next level that Dana 16 mentioned and then, finally, the individual activities, 17 the things that are going on or will take place in order 18 to meet those objectives and to supply the information to 19 carry out the function.
20 MEMBER POWERS: I see this as a definition of 21 a lower tier set of functions.
22 MR. PERSENSKY: Right. That's what I said. I 23 think this is where it comes into that next level.
l 24 MEMBER POWERS: Aren't you jumping just a g
( ,) 25 little bit ahead of the game? It's not clear to me that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
l 100 1 the upper tier -- there are a lot of things that are not
,m 2 clear about the upper tier.
\
xj 3 It is not clear that they are a necessary and 4 sufficient nature, the elements of your strategy. But 5 I'll bet they are. I'll bet you can probably persuade me, 6 but you haven't yet. Okay?
7 It is not clear that the functions are 8 orthogonal. That is, it seems likely and indeed 9 guaranteed that there are interfaces among those 10 functions. And you have not identified those interfaces, 11 nor have you set requirements for those interfaces.
12 So now without knowing those interfaces among 13 functions, how do you go about setting up lower tier
( )
V 14 functions without ending up being duplicative?
l 15 MR. PERSENSKY: I think when you get to that 16 lower tier since they are all going to be in one place and i 17 they're going to be related back to the upper tier, it 1
18 becomes the catalog in a sense. But it will have the i l
l 19 direct tie back to these original functions.
20 MEMBER POWERS: I think I know where you're 21 going. And, now, you can do it that way, but I'd be real, 22 real careful because of the creative use of language.
1 4 l 23 I think I know what you're going to do. l l
24 You're going to tell me I'm going to lie down the
,o
) 25 functions, sub-tier functions for each one of my upper NEAL R. GROSS COURT REPORTERS AUD TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
101 l
1 tier functions, and I'm going to find out that I have an
,~, 2 interface and an overlap. And then by iteration, I'll get
'~'#
3 rid of that sort of stuff.
4 MR. PERSENSKY: Get rid of it, yes.
5 MEMBER POWERS: Be real careful about the 6 creative use of language. A function that is, in fact, 7 identical can be labeled so that it looks like it's 8 completely different.
9 MR. HODGES: I think you've probably hit upon 10 those parts of the plan that we said are missing right 11 now. And we're not there yet. And that's why I said 12 we're not there.
13 But, I mean, right now what you see under f~h K- 14 activities, those are things that are ongoing. And so 15 they're listed. And right now it's a catalog. There are i 16 some steps between there and these higher-level things 17 that are not there yet. And that's what I meant when I i 18 said we're not done.
19 MEMBER POWERS: You have to iterate. In fact, 20 usually what happens in the textbook exercises, you don't 21 tell you how often you have to iterate.
22 MR. HODGES: Yes.
23 MEMBER POWERS: It all looks very clear once I
24 it's done, but it never starts off that way. This is --
s
( ,) 25 MR. PERSENSKY: We're going to do our best to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
i 102 1 bounce these off of ATHEANA as well, the model --
l m 2 MEMBER POWERS: Oh, yes.
i )
! \ #
/
l 3 MR. PERSENSKY: So that all of it has to fit 1
4 in. And, as Wayne said at the beginning here, there are 5 some things that are ongoing right now that may fall by 6 the wayside, either get lower priority or be eliminated 7 altogether.
8 MEMBER POWERS: I see ATHEANA as a very useful 9 tool for quantitatively defining your requirements for the 10 functions. You can go through, and you can say, "Okay.
11 This function has to be able to do it this well or it I
12 doesn't meet the needs of whatever it interfaces with."
i 13 Then you can get some idea on that quantification through i r'x i I
(> - 14 models.
15 MEMBER SEALE: Before you go on, on that I 16 previous chart, the last two columns identified target 17 dates and lead office. Am I overly suspicious when I note .
18 that the only target dates are opposite lead office j 19 assignments to research?
20 MS. RAMEY-SMITH: It's a coincidence of where 1
21 we are right now in the process. Yesterday evening I 22 received input from at least NRR, where they have gone 23 through the process of providing me with the target dates l
24 and so on.
l t'~'s I
(
v) 25 In this example, since I didn't --
NEAI. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344433
103 1 MEMBER SEALE: Predates that.
,x 2 MS. RAMEY-SMITH: Yes, it predates that. And, f )
LJ 3 rather than get myself in a whole lot of hot water by 4 putting the date down for them, I thought that I would 5 just put "To be determined."
6 MEMBER SEALE: I understand. I understand.
1 7 No. This happens to be Number 4. And I'm sure if it was I
8 Number 1, AEOD would be on the griddle, for example. But 9 just curious.
i 10 CHAIRMAN APOSTOLAKIS: If the lead office is l
11 NRR, --
12 MR. PERSENSKY: Yes. I 13 CHAIRMAN APOSTOLAKIS: -- what does it mean
/
i
\ ') 14 that Wayne has the ownership on the plant?
15 MR. PERSENSKY: Wayne has the --
16 CHAIRMAN APOSTOLAKIS: Yes. What can Wayne do 17 to make them deliver? That's not part of being owner?
18 MR. HODGES: That means that I go to their 19 management if things are not going the way it should be 20 and try to say, " Hey, we've got to get on the stick with 21 it."
22 CHAIRMAN APOSTOLAKIS: Okay.
23 MR. HODGES: Clearly they don't work for me.
24 MEMBER POWERS: But NRC is just operating in a
/
( ,)\ 25 more matrix fashion.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
104 1 MEMBER BARTON: They're a matrixed
, e^3
, 2 organization, right.
3 MEMBER POWERS: Yes, the same problem with 4 have with any matrixed organization.
5 MR. THOMAS: And I assure you Wayne has done 6 that already.
7 MEMBER SEALE: You mean he knows how. Very 8 good.
9 MEMBER POWERS: The one thing you learn about 10 a matrixed organization is that praise from outside 11 organizations is worth twice as much as within an l
12 organization. And criticism is worth ten times as much l 13 when it comes from outside than inside.
/,,T i
\ 14 MR. PERSENSKY: This is just an example of i 1
15 some of the planning sheets, essentially, that we have j 16 done that led to the table. Actually, these came first in 17 many cases. It's just an idea of having an individual PM 18 sit down and explain in a short way what their activity is 19 and how it fits within this overall --
20 MEMBER POWERS: That's just like your 21 brainstorming precursor to --
l 22 MR. PERSENSKY: Right. So it's a thought 23 paper. We probably won't even include these in the public l
l 24 version of the plan. It will be something that we keep
- ,m l (' ,,,) 25 within the NRC to go back to look at this level of detail.
l l
NEAL R. GROSS j
~
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 a
105 1 This level of detail if it goes too far becomes a
. 2 micromanagement tool for outside groups, even advisory
( i
~'
3 committees occasionally.
4 CHAIRMAN APOSTOLAKIS: Are the criteria for 5 success, criteria for closure? Is that what you mean, 6 really?
7 MR. PERSENSKY: Pretty much, yes.
8 CHAIRMAN APOSTOLAKIS: Because that's what --
9 MR. PERSENSKY: This is when we know we have 10 done what we intended to do. We structured these with 11 RES. And they seem to fit well with what we do. It 12 doesn't fit that well with what NRR and NMSS do. I'm not 13 sure about AEOD. AEOD had some problems again filling in 7 s)
/ 14 these things.
15 And, again, we've iterated several times, 16 started off with about eight different bullets, as opposed 17 to the four that are here. And we have brought it 18 together again. This is a document, more of an internal 19 document so that we can keep up, we as an agency, 20 everybody, not just Wayne, but to know where we are on 21 these various things and how it fits within the overall 22 plan.
23 What we're going to put out are more like 24 these tables, which is the same kind of thing that they
) 25 did with the PRA implementation plan. That's the model NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
106 1 for those tables.
,s 2 So I just wanted to explain why these are in
/ i i 1 3 the things to give you examples.
4 MS. RAMEY-SMITH: Can I mention one other use 5 of these?
6 MR. PERSENSKY: Do I have a choice in the 7 matter?
8 MS. RAMEY-SMITH: Yes. As Jay mentioned, this 9 information that's summarized on these sheets was vitally 10 important to support development of the charts, but they 11 will have another purpose as we go forward, you know, 12 because we are in the process of developing a plan.
13 So we have a mission statement that we're m
I \
k.) 14 looking at. We have proposed programs for those or 15 functions. We are developing subfunction requirements, if l 16 you will, or objectives for those programs.
l 17 And then, lastly, what you need to do is look l
18 at how do our activities support those things. Well, this l 19 is a process that's involving many different l
l 20 organizational units. These sheets become a communication 21 device so that we can talk about the relative priorities 22 of things and we can understand the rationale that went 23,I into them and so on.
.; So there was a little bit of grumbling among
,a the staff, "Well, why do I have to do this if it's not
!% )) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 1
107 1 going forward?" But I think it will become a tremendous i f ,s 2 planning help to us.
( \
(Slide) 3 4 MR. PERSENSKY: The final slide really just 5 'gets into some of the questions that we have been trying 6 to address as we have gone along in here. And some of 7 them came up as we talked today; the question of the 8 adequacy of the regulatory framework, for instance. You 9 brought up the whole issue of: Where does M&O fit within 10 our regulatory framework?
11 There are a lot of different elements of that.
12 There are the GDCs. We don't have a GDC that addresses 13 human factors. There are a number of different rules, 73
\
k -) 14 actually, not all that much, the formal rules, 10 CFR, 1
15 that are human factors-oriented. ;
i 16 The SRP and the branch technical positions I 17 think are much more used from a regulatory perspective, 18 and we're always working on reg guides and NUREGs for more l 19 specific issues. But is this the best framework to get 20 into the human factors and human performance elements?
21 The other question is adequacy of data. There 22 is a rule out there. Utilities provide LERs. Every time 23 we try to get to LERs we find that there's limited
- 24 information in that from the standpoint of human
/3 25 performance.
()
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
108 1 There is some information. Jack's people have g~ 2 been doing an effort to go out and compare other types of
( '
)
3 in-detail event reviews against the LER database to see 4 what things can be done. We have had a task force in the
- 5 past look at improving the NUREG-1022, which would include 6 more information, and mention HPIP, the human performance 7 investigation protocol, a lot of those elements in the 8 LERs that didn't go forward.
9 So we need to have -- right now we feel that l 10 there's not adequate data to do, in fact, a lot of what's 11 in that Program Area 1. It also doesn't support what's 12 needed for the HRA.
13 We have been talking about a database, s
i
\--) 14 consolidated database. NRR has their HFIS system, which 15 they find to be very useful. They use it for their needs. l l
16 We have the LERs. We have different databases that AEOD 17 has been working on.
18 We tend to believe that there is a need for 4 1
19 one consolidated database but in a form that people can l 20 pull off their needs, rather than having a bunch of 21 scattered databases. That's something that they were 1
22 working on.
23 We have a specific task force looking at that 24 issue as we're moving ahead with the plan. Roy Woods has
,9
( ,) 25 taken a lead on that task force.
NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C 20005-3701 (202) 234-4433 l l
109 1 MEMBER SEALE: I hate to beat a dead horse, 1
m 2 and I hope he's not really dead. But this issue has come
( '/
3 up repeatedly in a lot of other areas as well. And it's 4 only with I'd say considerable soul-searching and a 5 recognition of the overriding importance of the common 6 interest that the Commission and the INPO people have 7 managed to resolve a lot of issues with regard to 8 equipment data.
9 I guess my feeling is that if you approach 10 INPO today you may very well find that they would have all 11 of those difficulties that we talked about earlier 12 regarding interference and so on. And, yet, I would also 13 feel that ultimately people are going to see the wisdom of I,_,'i
\~/ 14 sharing that information.
15 And I just wonder if it would be the 16 statesmanlike thing to do to try to put together that kind 17 of rapprochement early, rather than later, and just let's 18 be grownup about this, let's behave in our best common 19 interest.
20 MR. HODGES: Yes.
l I
21 MEMBER SEALE: As I say, maybe there's l 22 somebody there that's grownup, too.
23 MR. PERSENSKY: We have attempted this many 24 times in the past. There's some question as to what data I fm
( ,/ 25 they have or --
NEAL R. GRC SS COURT REPORTERS AND THANGCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 j
110 1 MEMBER SEALE: That's always the first step.
7s 2 MR. PERSENSKY: That's always the first step. 1
( ) !
3 And we will do it again.
4 MEMBER SEALE: Deja vu all over again, you 5 know. I don't know. Well, anyway.
6 MR. HODGES: Also recognize that this data are .
l 7 not quite as straightforward as the data you were talking 8 about.
9 MEMBER SEALE: Oh, you're right.
10 MR. PERSENSKY: If we were to change the LER 11 rule, which might be necessary -- we're going to have to 12 go through the whole backfit process, all kinds of things )
1 13 that will go on with --
(')I 1
's- 14 MEMBER SEALE: I understand.
15 MR. PERSENSKY: And this is talking primarily 16 to the power plant. The data that NMSS has is a whole 17 other set of issues. i I
18 MEMBER BARTON: Why isn't one of your data !
4 1
19 points NRC inspection reports? I've read a lot of i 20 inspection reports.
I 21 MR. PERSENSKY: They are. They are.
1 22 MEMBER BARTON: All right. They are.
l 23 MR. ROSENTHAL: And, for example, I know that 24 NRR sent somebody on AIT at Zion. I sent somebody at an
/)s 25 AIT at --
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. ,
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
111 j 1 MEMBER BARTON: You see a lot of work there s 2 about a mechanism didn't follow procedure -- ;
.'^'
]
3 MR. RCSENTHAL: Right.
4 MEMBER BARTON: -- and there's a pump problem, 5 an operator -- l 6 MR. ROSENTHAL: And there are systematically 7 in charters for AITs, IITs, special investigations up 8 front, where tasking the inspection entity to ascertain 9 the human performance aspects of the events.
10 MR. THOMAS: And our database effort includes 11 reviewing all inspection reports, too, for human 12 performance data, which we find to be richer than LERs, as 13 you would expect.
,y
! )
\/ 14 MEMBER BARTON: I believe it.
15 MEMBER POWERS: AITs really I've enjoyed 16 reading their -- how far down they pursue the track and 17 try to give you the shaping factors and things. They did 18 a pretty good job on that.
19 MR. PERSENSKY: We also intend to look outside 20 the agency in terms of other places where there might be 21 data. Well, PWG 1 has got a new effort going on. But 22 even outsido the nuclear industry, there are some things 23 that might help us from the standpoint of identifying 24 issues, especially issues that we may not have faced yet g)
! 25 but are being faced in the aerospace community or the 1
l NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4435 l
112 I
1 petrochemical area. l 1
,, ~s 2 MR. THOMAS: Interestingly enough, we're we i t I i'"/
3 think making some inroads internationally to hopefully 1
4 some of the reporting requirements. International INES I 5 reporting will pick up the human performance information 6 in the near future. And perhaps that will pave the way !
I 7 for us. !
l 8 MEMBER BARTON: Good.
9 CHAIRMAN APOSTOLAKIS: We're going to have to 10 end this in about 16 minutes. So I'd like to have a ,
l 11 general discussion if possible. Are you done, Jay, with )
12 basically what --
l 13 MR. PERSENSKY: The only bullet that's left is ;
)
V 14 to say that we don't have enough money or people.
15 CHAIRMAN APOSTOLAKIS: Yes. We know that.
16 MR. PERSENSKY: I won't even bother saying it.
l 17 CHAIRMAN APOSTOLAKIS: Have you heard do more 18 with less?
19 MR. PERSENSKY: We have been asked to do that 20 all the time.
21 CHAIRMAN APOSTOLAKIS: So maybe it would be 22 most useful to the staff if we went around the table and 23 asked the members whether they had any comments.
24 But before we do that, do you, Wayne, expect
() 25 us to write a letter at some point?
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
113 1 MR. HODGES: Not for now. I think at this
,m 2 point we're looking at just any feedback we can get --
t i
3 CHAIRMAN APOSTOLAKIS: I understand.
4 MR. HODGES: -- from you. Probably early fall 1
5 or some time frame like that it might be appropriate to do 6 that, but we'll show you the end of this month what we 7 have as far as the development of the plan. But there are 8 going to be iterations. So I don't see a letter for a 9 little while yet.
10 CHAIRMAN APOSTOLAKIS: Okay. So maybe we 11 should plan on doing it in the October meeting or 12 something like that? Yes.
13 MR. HODGES: Probably be something like that.
p}
i
'-' 14 MR. PERSENSKY: I'd expect another 15 Subcommittee meeting on the document when you have it --
16 CHAIRMAN APOSTOLAKIS: Yes.
17 MR. PERSENSKY: -- and then a full Committee 18 on --
19 CHAIRMAN APOSTOLAKIS: And that will be 20 sometime in September you think?
21 MR. PERSENSKY: Probably.
22 CHAIRMAN APOSTOLAKIS: Yes. So maybe 23 October-November.
24 MEMBER POWERS: I'd just point out that our I
{) 25 schedule is awfully packed on --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (2 2) 234 4433
114 1 CHAIRMAN APOSTOLAKIS: Yes. That's why I want g, 2 to plan ahead.
', )
v 3 MEMBER SEALE: Right now it's do we drown in 4 200 feet of water or 300 feet?
l 5 CHAIRMAN APOSTOLAKIS: Yes. .
l 6 MEMBER POWERS: So we may become a drag on l
7 this.
l l
8 CHAIRMAN APOSTOLAKIS: No. But the point is l 1
i 9 it seems we already wrote the letter that was -- l l
10 MEMBER POWERS: Yes. I think -- )
11 CHAIRMAN APOSTOLAKIS: It would be nice to 12 come back and say, "Yes."
13 MEMBER POWERS: Yes. And I think we are p
14 obligated to do this, but I don't think we're in a i 15 critical path.
16 MR. PERSENSKY: Our primary critical path 17 right now is the end of June. I mean, that's the only 18 critical date that we have. Beyond that, it's going to be 19 up to the EDO. Our goal right now is to get him a plan 20 that can he sent out for review.
21 CHAIRMAN APOSTOLAKIS: Okay.
22 MR. PERSENSKY: And we will have other 23 reviewers involved.
24 MR. HODGES: Our aim is to have a plan we can
( ,/ 25 start implementing in the next fiscal year, which is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
115 ,
i 1 October, i
,, 2 CHAIRMAN APOSTOLAKIS: Okay. So who wants to 1 I )
LJ 3 start? Dana?
4 5. COMMITTEE DISCUSSION 5 MEMBER POWERS: Sure. It's clear that the 6 staff has looked at the variety of questions and 1
7 criticisms we've looked at. And they have addressed 8 several of them. Certainly there is a clear line of 1
9 ownership to me in this plan. And it's clear that you j 10 have factored in, brought things together and certainly ]
l 11 have a model that allows you to guide the design of this l i
l 12 plan. I 13 I'm still a little bit concerned about the !
1 V 14 discipline of planning that's going into it. I am 15 concerned that your strategy, what I call your mission 16 need but what you call strategy, is not quantitative.
17 And you don't have a metric that will show you 18 that the work is both necessary and more necessary than ,
1 19 other types of work, that the work is progressing toward a 20 goal. And I don't see how you're going to quantify 21 requirements unless you have a metric for that mission 22 need.
l 23 It is unclear to me how the functions, what 24 you call your seven programs, were defined to assure that p.,
{} 25 these programs are both necessary and sufficient to meet NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
116 1 the mission need. And that's really at the heart of my n 2 problem right now.
( )
(/
3 I look at the programs. They lock like good 4 programs. But I don't see how they show that I will meet 5 the mission need. I'm a little unclear about the mission 6 need clearly, but I don't know that I will get everything 7 there.
8 I think you're going to have to struggle with 9 interfaces among those functions. I don't object to the 10 iterative approach. In fact, I think that's fine. But 11 just recognize you're going to have to struggle with those 12 functions.
13 I think you're going to have to struggle to
(~\
d 14 define requirements for the functions. And how you set 15 out to do that, it 1coks to me like your programs are all 16 going to be sub-tier functions. But you're still going to 17 have -- your activities are all going to be sub-tier 18 functions.
19 You're still going to have to have 20 requirements for those functions. What are they going to 21 do and what are they going to yield that you're going to 22 have to make quantitative? I think that's going to be a I
23 big struggle in the design of this plant.
24 I am heartened because maybe I hear what I O
() 25 want to hear, but I hear people moving toward a more I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
117 1 disciplined planning approach than we saw in the catalog. I i
,_ 2 And I understand that it was just intended to be a catalog l
('~' .I 3 at the time, and now you're trying to set up a plan. And ,
4 that makes me feel good. Now, maybe I'm just hearing what !
i 5 I want to hear.
6 That's my view.
)
7 CHAIRMAN APOSTOLAKIS: Bob?
]
\
8 MEMBER SEALE: Well, I think you certainly hit l l
9 most of the bases, Dana. I think the thing that stands I l
l 10 out to me is that what you have now very definitely does ;
11 have the appearance and the critical elements of a plan. I i
12 It's not just a listing. ;
)
13 It's obvious you've done a lot of hard work to r"N.
)
i
' \ i m/ 14 do that and to get there. And I want to see where you go )
15 with it. But so far I'm impressed with where you've come j 16 since the last time we got together, 17 MEMBER POWERS: Yes.
18 MEMBER SEALE: You've certainly not shown any 19 abhorrence with the idea of wadding up a piece of paper 20 and throwing it away. Of course, that's the ultimate 21 measure of the iterative approach to things.
22 They're thinking tools. But, you know, a 23 piece of paper will just lay there. And you can put 24 anything you want to on it. And you're the one that has
()
,f' \
25 to decide whether or not it stays there or is replaced by NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
118 1 something else.
.x
, 2 And it looks like you've done a lot of (v) 3 thinking along those lines and you've thrown away some 4 ideas and you've captured some other good ideas, too. I'm 5 interested to see where you go from here.
6 MEMBER POWERS: I have to echo what Bob says.
7 It looks like it would be just a ball to plan this thing, 8 of course, given that you have time to do it.
i 9 CHAIRMAN APOSTOLAKIS: They do. They do. l i
l 10 MEMBER POWERS: I think these guys are under i
11 more pressures than they let on sometimes.
12 MEMBER SEALE: Yes. And I just wonder how 13 many -- I mean, you've got an awful lot of bases to touch
[,_ )
i/ 14 in this business. And, unfortunately, it's more than just 15 touching them.
16 There are a lot of things that just aren't 17 there that you're going to have to invent or find ways to 18 quantify and so on. That's a lot of fun, but it's a lot 19 of work. And so you've got your work cut out for you.
20 CHAIRMAN APOSTOLAKIS: That's it?
21 MEMBER SEALE: Yes.
22 CHAIRMAN APOSTOLAKIS: John?
23 MEMBER BARTON: I think we ought to let them j 24 go back to work. I'm encouraged by the new direction.
l 't
) 25 And it's nice to know who we can go beat up on. It's l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
119 ,
1 1 Wayne Hodges. Now I think they've got a lot of work --
fw- 2 MEMBER POWERS: But you have to stand in such r a
\/
3 a long line.
4 MR. HODGES: I used to beat up on him when he i
5 was at Oyster Creek.
l 6 CHAIRMAN APOSTOLAKIS: Yes. I think this has 7 been a very good meeting. I think you guys have made a 8 lot of progress. Two specific comments that I have 9 already made, but !ust to repeat: I would like to see an 10 adaptation of the ATHEANA diagram to this particular issue 11 by showing more boxes feeding into the performance-shaping l 12 factors and then show how the programs or maybe follow 13 Dana's nomenclature, how they fit into this.
)
14 And then perhaps that would be the basis for 15 developing the metrics that Dana mentioned, although 16 that's something we would probably have to discuss, 17 though, whether you want to be some quantitative if that's 18 what you meant. An expanded diagram I think would go a 19 long way towards conveying pictorially the whole program.
20 I would like to see the words " safety culture" 21 being used more prominently with a definition and so on.
22 I realize this is a human performance program plan, but l
l 23 this is the place where we deal with this thing.
24 And issues such as formal versus informal gy
(,) 25 culture and what are the elements of a good safety
- NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(
l (202) 234-4433 WASHINGTON D C. 20005-370i (202) 234-4433 l
120 l
1 culture, be they organizational or psychological or 7- 2 whatever, I think there is a lot of information already in
( )
3 the literature, starting with NSAC-4, which is already 4 kind of old. I mean, it was published in '90 '91, '90 5 something.
l 6 MEMBER BARTON: Early '90s.
l 7 CHAIRMAN APOSTOLAKIS: Yes. NSAC-3 was in the ;
l 8 mid '80s. And then '89 or '90 we had NSAC-4. And then l 9 there was a conference in Vienna in '95 and so on.
l l
10 So there is a lot of information out there on j l
11 safety culture. And I think it's a good opportunity here 12 to bring some of it as it relates to human performance, of l l
13 course, although I think most of it does, into this and j
,( '~h s I
'k / 14 use the words more often than just when it happens to come 15 up in the discussion.
16 Other than that, I think you're on the right 17 track.
18 MEMBER POWERS: I would just add one just 19 supplemental comment just to reiterate. Again, I 20 adamantly do not believe the NRC is guilty as charged, but 21 I think NRC's program managers and your project managers 22 ought to examine the criticisms that have been leveled at 23 the NRC concerning involvement with the rest of the 24 technical community.
(
(3; 25 I think you're guiltless but you're being l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
121 1 criticized anyway. So it may be a PRA game that you have 2 to play. But in this area because I think you will be O 3 breaking ground and setting standards for the rest of the 4 government, I don't think you want some National Academy 5 committee coming in and criticizing you. I want the 6 National Academy to be able to assemble a committee to 7 examine you because you have such a presence in the field.
8 It's just something to pay attention to 9 because, like I say, I don't think you're guilty as you 10 have been charged.
11 MEMBER SEALE: Well, and in that regard with 12 respect to this travel charge that has been made against 13 you, in particular, I would encourage you to nip that sort 14 of thing in the bud, the kind of thing that Dana is 15 talking about, because you clearly are getting closure 16 between the concept of a human performance plan and the 17 nuts and bolts of actually applying it to some real events 18 and the real industrial problem.
19 And so I would urge you to share what you're 20 doing with the technical community, get their input, get 21 their criticism. So you really need to be -- he says it's PR, but it's PR of a peer review kind. And the last thing 22 23 you want to do is to be criticized for not being in the middle of the hunt in this particular area. So go get 24 25 them.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N.W WASHINGTON, D C. 20005-3701 (202) 234-4433 (202) 234 4433
122 1 MR. PERSENSKY: I'll be making a presentation
,- 2 at the IEEE meeting next week.
( ,
3 MEMBER SEALE: Fine. And make sure it's to 4 the guys who are doing human factors work in other areas.
5 I mean, pick your sessions if you can.
6 MR. PERSENSKY: Well, to the extent that we 7 can get them to come to our meetings and where we can go 8 to their meetings.
9 MEMBER SEALE: Yes, right.
10 MR. PERSENSKY: You also find that when you 11 submit a paper on nuclear activities to an FAA --
12 MEMBER POWERS: You get pigeonholed.
13 MR. PERSENSKY: This doesn't relate to our r3 i a KJ 14 work. So it doesn't necessarily get accepted.
15 MEMBER SEALE: Yes.
16 MR. PERSENSKY: Now, they do occasionally, but 17 --
18 MEMBER POWERS: I am just very distressed by l l
i l 19 the criticisms when I've heard it made before, my most ;
l l 20 recent one. And I don't know what to do about it because l
j 21 I know all the problems you're talking about. I know all l
i 22 the problems of, " Gee, we barely have enough money to do 23 the work," period, let alone go out and talk about it.
24 That's going to kill everybody.
,7
) 25 CHAIRMAN APOSTOLAKIS: Speaking of money, NEAL R. GROSS COURT REPCRTERS AND TRANSCRIBERS 1323 RHL DE ISLAND AVE., N W.
(202) 234-4433 WASHINGT JN, D.C. 20005 3701 (202) 234-4433
123 1 maybe this is not the appropriate forum, but at some point r~s 2 I would like to discuss whether the agency is spending
/ i
~
3 enough on this subject. Is that something the ACRS can 4 comment on?
5 MEMBER POWERS: My problem is I have no 6 expertise in sm '".g what enough is.
7 CHAIRMAN APOSTOLAKIS: But you can compare 8 with what we are spending in other areas.
9 MEMBER SEALE: About which he is equally -- l l
10 MEMBER POWERS: I can compare only whether the 11 risk significance and importance are commensurate between l l
l 12 --
13 CHAIRMAN APOSTOLAKIS: When was the last time i' ') 14 that we had a major incident in this industry because we l
15 didn't have very good thermal hydraulic codes?
16 MEMBER POWERS: Quite often, George. Every 17 time we --
18 CHAIRMAN APOSTOLAKIS: Jack, when was the last 19 time you saw an incident where the lack of a thermal 20 hydraulic code was a major cause of a nuclear incident?
21 MEMBER POWERS: Wolf Creek, frazzle ice, 22 improper calculation of the flow through a warning line.
23 CHAIRMAN APOSTOLAKIS: And that led to what?
24 MEMBER POWERS- Shut the plant down.
x q ,) 25 CHAIRIMN APOSTOLAKIS : Now, we're going to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE If, LAND AVE., N W.
(202) 234-4433 Wf SHINGTO'4, D c. 20005-3701 (202) 234-4433
1 l 124 i
1 say, "Your incident is worse than mine"? I mean, I can l
, ,f~ss 2 give you TMI. I can --
3 MR. ROSENTHAL: I think the frazzle ice, Wolf 4 Creek, was like a 10 conditional core damage probability.
5 MEMBER POWERS: He keeps ratcheting on me. He l
6 asked for the last time I can remember. And then he wants 7 me to put a core damage frequency on --
8 MR. ROSENTHAL: Let me point out that what I 9 call Wolf Creek 1, an event that happened a year before, 10 totally dominated by human performance, was quantified at 11 10~3 and, in fact, was the most significant event since 12 Davis-Besse in the ASP measures.
,- s 13 CHAIRMAN APOSTOLAKIS: So if I use as a 14 criterion the number of incidents that have had some 15 significance --
16 MEMBER POWERS: Now, you know very well in a 17 letter from the EDO, that the criterion has nothing to do 18 with risk.
19 CHAIRMAN APOSTOLAKIS: Well, could you use 20 more money?
21 (Laughter.)
22 MR. PERSENSt:Y: Of course.
l 23 MEMBER SEALE: Dr. Apostolakis likes to leave 24 them laughing, you understand.
,/ ~ \
) 25 CHAIRMAN APOSTOLAKIS: Okay. We appreciate NEAL R. GROSS COURT REPORTERS AND TRANSCR:BERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C, 20005-3701 (202) 234-4433
125 1 very much your presentation today. It was very useful, I
,s 2 think. And we hope that we have given you some comments
)
i
'~'
3 that have helped you.
4 And then we will arrange for another 5 Subcommittee meeting, then, when you feel you are ready, 6 early fall sometime. And then later we'll write a letter l l
7 because I think we need to write a letter given that we '
8 wrote a letter that was of a negative nature.
9 Okay? l l
10 MR. PERSENSKY: Thank you. l 11 CHAIRMAN APOSTOLAKIS: Thank you very much. l l
12 (Whereupon, the foregoing matter was concluded 13 at 3:06 p.m.)
'. )
'w / 14 15 16 17 18 19 20 21 22 23 24 r~ ' 3
'..//
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
f~
b) C E R T I F I C'A T E This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission in the' matter of:
Name of Proceeding: ACRS SUBCOMMITTEE ON HUMAN FACTORS Docket Number: N/A Place-of Proceeding: ROCKVILLE, MARYLAND were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and, thereafter reduced to r
typewriting by me or under the direction of the court reporting company, _ and that the transcript is a true and accurate record of the foregoing proceedings.
r WA '
M)f1BETT iill/ER Official Reporter Neal R. Gross and Co., Inc.
l I
O NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RilODE ISTAND AVENUE, NW (202)234-4433 WAS!!INGTON,D.C. 20005 (202)234-4433 I
/
5 INTRODUCTORY STATEMENT BY THE CHAIRMAN OF THE HUMAN FACTORS SUBCOMMITTEE 11545 ROCKVILLE PIKE, ROOM T-2B3 ROCKVILLE, MARYLAND JUNE 3, 1997 The meeting will now come to order. This is a meeting of the ACRS Subcommittee on Human Factors.
I am George Apostolakis, Chairman of the Subcommittee.
The ACRS Members in attendance are:
John Barton, Dana Powers, and Robert Seale.
The purpose of this meeting is to hold discussions with representatives of the NRC staff regarding the status of the Human Performance and Reliability Implementation Plan. The Subcommittee will gather information, analyze relevant issues and facts, and formulate proposed positions and actions as appropriate, for deliberation by the full Committee.
Noel Dudley is the Cognizant ACRS Staff Engineer for this meeting.
The rules for participation in today's meeting have been announced as part of the notice of this meeting previously published in the Federal Register on May 9, 1997.
A transcript of the meeting is being kept and will be made available as stated in the Federal Register Notice. It is requested that the speakers first identify themselves and speak with sufficient clarity and volume so that they can be readily heard.
We have received no written comments or requests for time to make oral s;atements from members of the public.
[ Chairman's Comments-follow:)
In response to comments by the ACRS concerning the Human Performance Program Plan, the staff has been developing a comprehensive program plan. The Of fice of Nuclear Reactor Research has been leading an agency-wide effort to develop a human reliability assessment and human performance evaluation program plan, which will serve the agency's future needs for. confirmatory research and developmental projects.
\/ We will proceed with the meeting and I call upon Dr. Jay Persensky of the Office of Nuclear Reactor Research to begin.
l
. - = - - _ _ - . _ - . - - - - . - _ .
~?
O .
O O . n.
NRC's Human Performance and Reliability Implementation Plan:
Mission, Framework, and Progra.ms Presentation to ACRS Human Factors Subcommittee June 3,1997
O O O l VISION STATEMENT Human error should not unduly or disproportionately contribute to the risk associated with the use of nuclear materials.
i i
l T
1 l
f I
i.
O O O :
MISSION To ensure effective risk-informed and performance-based regulation and oversight oflicensee personnel performance in the design, operation, maintenance, license renewal, decontamination, and decommissioning of NRC-licensed facilities, as well as the transport, storage, and use of nuclear materials.
1
-- -- -_ - - - - - - - - - - - - - - - - - - - - - - - - - J
I i
HUMA% PERFORMADICE Afe 88 mm ITY NATION PLAN , j (May 1997) -
I i'
Progresa Aree Obfectives AcIlvlues Target Lead seneduto Osmesse) i
- 4. Developthetechnscal 1. Improve the .1._..i., of human femors 1. Evolueto the eseds of advanced cJ._ ? , , TBD RES beeis to ogport human issues and their eGods on human portormance mledece management teoks on human performance podermance evolushon and and human sekstuhly of liconese personnel and and sehebay human M analyse develop appropnote gindence 2. Assess and expend the technscal beeis for RG 1.164 TBD fBES I (ANS/ ANSI-58.8). !
- detersrunang IIInsos for duty. l
human rebelugly ;
- 5. Evolueto the faselbGty of usin0 took network TBD RES I modehne to soport review acevees for meteness bconeses -
- 6. Provato foodhed on event data needed to support Penoecosy RES.
the development of the technecol besos for human
-- _ - evolushon and human sekstutty _._^; x.
- 2. Integrate human ladors pnncipios into risk- 1. Developguagenceforreviounng i informed and persormance4eeed reguiston - advanced alarm systems V98 RES !
- hybnd"controlrooms. TBD RES !
- w;._ , , voorface management 2/99 RES i
- 2. Upgrade the Pp Reviour Model for nnneur of TBD NRR advanced control room doogns and incorporate it as i appropstete into sie SRP.
- 3. Upgrade NUREG-0700. Rev.1. 2/00 RES ;
- 4. Develop root cause invesegemon guagence 12/98 RES
bened emeernent of enees for duty
- 8. Complete sowiesor. to SRP Chapters 13 and 18 to TBD NRR solemnce NUREG#711 and NUREG-0700. Rev.1.
J i
- 3. Condud evaluamons and develop 1. Provide human factors ster support for intemel NRC As requwed AN I secommendemons, as smym, on NRC's in- needs associated weh personnel performance and !
house human performance inoues isnpostent to reheluity seguistory esednenees. eniciency, and 2. Develop pnncsples for the eSedne communscahon of TBD emosannoe RES nok to Ine punte.
- 3. ReGect planned adnnhos to develop technscal beoes Perm RES ;
in a modecohans no niis pion. y i
f 1 ;
i>
. - - - - -_ _ _ _ _ - - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ . - _ _ - - _ _ _ _ - . _ - _ _ _ . _ _ _ _ = _ - - _ - _ - _ - _ _ _ _ _ _ - _ _ _ _ _ _ _ _ - _ _ _
o O O -
~
ISSUES ~
o Adequacy of regulatory framework GDC Rules SRP & BTPs j RGs o Adequacy of data LER rule LER guidance NUREG Convert databases to consolidated system o Resources Staff - number and composition Contractor S June 2,1997 (2.07pm)
G:WP&RPLAN%CRSSLID.602
z kO l s
n N=
a m
8 a s u .
I i
. ---- Error
- :: :. PRA :
Forc.mg ::
Human Error
- Context :: :: Model :
~
- - m- Enor Operatons w
- ' Shaping n M-X.Lw %
Unsafe Actions
- I Human Fadure Events w
- W Risk Factors jj :* w.v,s asarmonance .
.i _
m.a , 6 .
oo .._ ..............
s t . _
. . , u
! : : . scenano -
Defindson Figure 2.1 Medtidisciphnary HRA frasnework
SEVEN PROGRAM AREAS Program 1. Conduct operating events analysis and database maintenance to support human performance evaluation and human reliability analysis Program 2. Provide human performance evaluation and human reliability support to inspection and review activities of nuclear reactors Program 3. Provide human performance evaluation and human -
reliability support to inspection and review activities for materials licensees
...m. . _ _ _ _ _ . _ _ _ _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . -. _ _ . _ _ _ _ _ _ _ - . --
__ - - - ----v
O O O :
l PROGRAMS (continued)
Program 4. Develop the technical basis and guidance to support human performance evaluation and human reliability analysis Program 5. Develop the technical basis and guidance on management and organizational influences in human performance and facility risk Program 6. Develop and update an integrated model of human performance and human reliability
O O O :
PROGRAMS (continued) 1 I
Program 7. Foster national and international dialogue and cooperative efforts on human performance evaluation and human reliability analysis methods and data I
O O O FRAMEWORK FOR AGENCY ACTIVITIES The ATHEANA (A Technique for Human Event Analysis) model .
of human reliability is a suitable framework for demonstrating the relationships among the agency's human performance and human reliability activities.
t
I iO
SUMMARY
OF RESEARCH Environmental Influences on Human Reliability
& ANALYSIS ACTIVITY:
i
) PRIMARY PROGRAM AREA: Program Area 4, Develop the Technical Basis to Support Human i
Performance Evaluation and Human Reliability Analysis, Activity
- 4 3 BACKGROUND: An earlier effort involved the identification, review, critique, and summarization of 1
published literature that described the influences of environmental influences on human performance. The first volume of the resultingNUREG/CR is a handbook for use by NRC inspectors to help them determine the impact of specific environmental conditions on licensee personnel performance; included are the units i used to measure each condition, effects of the condition on task performance, examples of assessment of j each condition in a nuclearpower plant and methods for reducing the effects of exposure to the condition.
j
' The second volume details the specific behavioraleffects ofparticularenvironmentalstressors. For example, information was developed that shows the impact of high ambient temperature on cognitive reasoning ability
- and the impact of floor movement on motor skills.
i RATIONALE AND PROGRAM INTEGRATION: The availability of such human performance information can provide invaluable support in the assessment of human reliabilityduring accident scenarios, i
especially those involving extemalevents (i.e., seismic, flooding, and fire conditions). Rather than making gross calculations of the impact of extemal events on quantified estimates of human reliability, a far more j 1 research informedHRA can be conducted that brings to the analysis knowledge of the performance effects l l of environmentalconditionspostulated to occur during accident scenarios. 'This knowledge can be used to l ensure that appropriate human events are modeled in the extemal events PRAs and the effects of-
! environmental conditions are understood and considered properly.
PLANNED WORK:
- 1. Appropriate HF and IPEEE review staff review insights from NUREG/CR-5680, "The Impact of ;
- Environmental Conditions on Human Performance."
l 2. Evaluate results of the environmental influences project.
- 3. Determine additions and modifications to the " error forcing context" aspect of the ATHEANA 4 model and associated guidance based on work in (2), above. I
- 4. Suggest further research, as appropriate.
, 5. Incorporate final results of this effort into IPEEE review guidance. i l
CRITERIA FOR SUCCESS: Future HRAs will be conducted to include research-infamed knowledge of l the performance effects of environmental conditions that can occur during accident scenarios. Effects of j
- those conditions will be better understood and considered before inclusion in extemal events PRAs.
1 b
e l!
4 iO June 2,1997 (3:58pm) G:\HP&RPLAN\P4AX-DM.425 2
.c
SUMMARY
OF RESEARCH " Effects of Advanced Control Display Interface
& ANALYSIS ACTIVITY: Management Tasks on Crew Workload and Performance"
- PRIMARY PROGRAM Program Area 4, Develop the Technical Basis to Support AREA: Human Performance Evaluation and Human Reliability Analysis, Activity #4 BACKGROUND: As operational experience with advanced human-system interfaces (HSis) in
' complex systems such as aviation, medical and nuclear systems develops, there is increasing evidence that advanced HSis are associated with increases in: crew workload, the time ruluired for operating personnel to configure the interface, and human errors associated with interface management activities.
Interface management tasks associated with advanced HSIs are known as " secondary tasks" because they are not directly related to the operator's primary task of monitoring and controlling the plant. Typical interface management tasks may include: navigation between displays and within display hierarchies; window manipulation and control; dialog with the computer system.
RATIONALE AND PROGRAM INTEGRATION: Interface management tasks can impose a significant additional workload on the operator which can become a safety concern because these secondary tasks could require operator attention which would be better allocated to the primny task.
This concern is applicable to both current and advanced reactor designs. Advanced plants are being designed using digital instrumentation and control and computer-based control rooms (CRs). However, advanced HSis are increasingly being introduced into current plants due to: (1) replacement of existing HSis with computer-based technologies when existing hardware is no longer supported by equipment vendors, and (2) upgrading CRs with new, c*omputer based monitoring and control systems. A lack of d
f,1 human factors engineering guidelines to address interface management issues was noted during the V
development of NUREG-0700, Revision 1, and documented in NUREG/CR-5908. The primary objective of this research is to develop guidance for NRC staff for the review of the interface management aspects of advanced HSis.
' PLANNED WORK: Human factors issues associated with interface management tasks will be identified and prioritized. A methodology for developing review guidance for advanced HS1 interface
- management tasks will be described. Validated design review guidance for interface management aspects of advanced HSis that could adversely affect safe operations at a nuclear power plant will be prepared.
l CRITERIA FOR SUCCESS: Independent peer review will be conducted on the draft review guidance.
The guidance will be revised based on the peer review and NRC comments and published as a final NUREG/CR.
)
4 i
i June 2,1997 (4:05pm) G:\HP&RPl.AN\P4A4-IS.602 i