ML20217Q430
| ML20217Q430 | |
| Person / Time | |
|---|---|
| Issue date: | 03/03/1998 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-3029, NUDOCS 9803120246 | |
| Download: ML20217Q430 (186) | |
Text
W..,; >.'s. ;;.q: ',.:y., (;, ;a,; &};y.;, r:ay
- c ' '
v.
G*
e
..f.. -
- i.... }
.G.,..
1,..
J.
p
, j. -
b.
,.',..,,. A, r
- A
.... yW.f..y,.': * %l Ti-
$l g
. ::'p
."r.p.
,-e e
s.:q x..
,t
.f,
^
r h*,p l.J.'.". 7
- Q.' .7Q'4,(}p
g,,
7 y
1 1
g -
g f-i.7
.[e \\
h :.
.~
T
?.
W (k.a
. I:..
%1 5
9' 3
if'. I.
4 Mi-4.y Z.Ml,,g;..\\ ;. 3,;{-)%
.,.d f(f,9, m N ^7..
- a
-l};;;.:
- g.,: ^ h ',. _ -
Q. 3 -
. ' +. -l t.
f.'.';[f. M.".? o.
, i ' J '.
f f
^
f"'
.k..
fNp
... l. ' f..
f,,'. '. '
1
^,
.,.h.;,.^.f.&.c.rp*f.. :':,fff; !'. :, f ' '; l.h..
v
,f. );
. i:... * &. 4o y g ~ q
.s a
- 9..
.y 4
"W y.
... ;l L.
h.r s
j
,.?
].;.
.; L.'.7_f_%Q;0.{;. _;$(g 7:; '; 7.if:h :;,f.: c:
b ?;.. :.
pf.'l,.W.'l.4hT[f,f;4,4[
ag *s
.. #:L f.
- n
- . g.
- j;5%$
- f M%%.f}hA6.
~>.
}g '.i. iG&h*<p'M gy
'* s j'i '..i.",A..&. ' ;w P." + ',.e,.23. L' [}t.
y.
,. ',.. [ ' :
<. i.. h -
M, 0'
.fY
'"['
-' {, (. - i.
.N.
b
[.
4.. ;. ;. '3>..V..A.....
) :
8
' [,,f w (
i:
.- g [J...
- 1. ; w +].....m..V.
i g
Q:.p:c x.:.l;;3. Q: s s: Q
.T.
. W:& 2 C-J: h
.. 7., $m.,w:f%.9.~,..y.:
e 1.. ;)
- 7...
-:7 s.x;g'. '
Y.'.4*
.; W m-
, r.
.N!.:- Mlf?l *l :/ :
?
- 5..
.h
. /. h l:h[.;,.
.h,(
k,.h.'
b
' b i ~ ' '
L A
,' h
},,[,i :
NI
~
- M.
[
%, _... w n. m. m..w m....,p.ig
.=
4 n
- M.T &;&.Q W r Q 7..
.... 6
'e.. Ci k,,,,..
- h,,;L,: T.{f'..f.]p% qy@g.Q.,Z..
.C; y.....
~
G. ':'.
.t-
>. j r
. ';. M:...... p
.h. :.. : : ' -.:. n " s..
y. :py.:.7
- i? '. ?'.:s. 6;;:.p + 7;..:.;.,gl.T. G g c.Yr%V
' k.
/
7 B -
e y
) '... % jfl(..,. P; }' { '.
} ' y, s
~
.W 9
. ; j ;"J.. y 4...e. :.9.3:.y
- f' ;f.
. - l,:
- y;%.. #,. Le,9.pt.W).,,p+ d, ' 5.t
.. }.,.,.... ' '(f.% y c. V. m, p, n 4.;.g '..y g..: ~s
~
- w
\\
~
h y>
y, v,?. r ' &.
?
p.n,
'..e.e.. M.. },
.Y-
.n.,
- e c
. - i, '..,.s.,y, '
c
',[.
.h 4 D y p.: q d :
'p 0(; '
)
- 's
""~
^
l..%.k.Q,:.L's?; ls..'
, g%.
t\\
~
-. :.5l }.:
.. ;[;,.:, pgagn#9
-i i i..... b.
f' Y.'.
- ~
e;
.. 5..,. :.I.,T N.(,M. u.W;.:.M.a( %, g:u, e@. '
. ' "..I.'
E w,4
.e
- ;i
- -
a'
,. A e
- =
4+.
f 'Qg..
- Q.s.
4,;
s.
-is 1.
'p 7
Rh.l
- ye
- 7.',A p L [$.
- W '. : lJ: e..,9 ' ::
i y,
..~~.i y:- F' A f
l
,t
..q ;.
_ A
..y,
,~~g._.,.,;.,,g.:,.. ;..q,
...~..e
' :..v
- 7
- 7. ';.(l,. '
... ;;, jp..
(,
_...*';,V l
a.
s' k
( - e
- ' ',;; ',7.'. ~.. %, );$l.%(.y: hi' Qh;; q s
[.
- 'y;t *
- i.' ;, ', '. 's. ;..V.:.L i
~~
b.: ':'
- M. n,.-u',.
4.
s -s.
a l l?l /N. %,d. :;g.:
.u.:
.n
.v f.:7 s
.r, f
- ..:c w ?.... '
o 3-c.
.: ~ ;pp r L
- d. y'e:.::: p% W ' s,
y
.w
+
_1 y l.
-! [fY
^
- r. '.fl 5$.q.f;..:.:Q M f'
~
}E,.. -,.
..c.
i
- r e Y,.$gu9hp u
u.
.. r.... i.
f..,
4.
' w..;-)..f.Q:Xl:
n
' ~
- t..
e.
t.
..:... ql.,g -
\\
1.'-
f s
,; }
- ...v...c.c((h.r, w#g $;;
r s
. < m.
- ... v,
?e w
w
/n..,
s.
"~
L.l:f.hf k I?
~
h -).
\\
/
'f. *
{
ng -
N.....h.', ' ;c.. '.'f,
. ~.
A Ak.
e.
y e
.[ f;'. y
. f s
9 A:
M'..F.h 1-t 30,j.:l}j:Y.h+,a&&b$,'
c' ~ ' ' ' ' ' '
j.: Wj.,
"'f t*
m.;jf'iW;m^ h:.
o
+ w e
- h.
.L:.)A :h
+,..:,..y >
C.,.
>, s. y = 3 -, 9.+, e..
y e.
.T:
j. ~. ~ &
.i-
.. ?g.K..c.s.4,.e.
,.,u
.:m.
,s
.:.[& w.c. n: 3..
. 9 y
.#.a
-. :.~.f.yg ;g: : ),. 2..
4:
<,...g u,..-.
e-e og.
- e o
)
.'.,d j
't
't h
s u
t f
4 m.
.f y
~
%.....; yo;. y +...~. gh.n,9..,. n.
,p 1 -
3 e-
- ..,. j/..
4-m;.
e
- . v. 4...
g
. ".h,k l<,[ : '_
. l Y
['
n M G:\\ AL ACR37~ 3029 OFFICIAL TRANSCRIPT OF' PROCEEDINGS C'
'f.
l
- NUCLEAR' REGULATORY COMMISSION ADVISORY l COMMITTEE ON REACTOR SAFEGUARDS
{
1
Title:
.449TH ADVISORY COMMITTEE ON REACTOR SAFEGUARDS j
i TR14 ATFs R ER'R N 2.R : 3 : N A L
~
Ev"4H I T E M.S i
n.. r~
3 ocket No.:'
D pg 4:< 7: -
.z j -
THAxxs.
v b"
. Work Order No.:
ASB-300-166 i
Yj c'.
11 ll 1111Illl Il ll LOCATION;.
Rockville, Maryland DATE: ~
Tuesday, March 3,1998 PAGES:230 - 393 g 71 e % 0'303 3 xm rop ANN RILEY & ASSOCIATES, LTD.
1250 I Street, NW, Suite 300 Washington, D.C. 20005 A2S 0"ce 3hh Reta' foQelEe oithe Comittee
- W
l
. n) l
\\_
DISCLAIMER I
1 UNITED STATES NUCLEAR REGULATORY COMMISSION'S ADVISORY COMMITTEE ON REACTOR SAFEGUARDS MARCH 3, 1998 The contents of this transcript of the proceeding of the United States Nuclear Regulatory Commission Advisory A(,)
Committee on Reactor Safeguards, taken on March 3, 1998, as reported herein, is a record of the discussions recorded at the meeting held ec the above date.
This transcript had not been reviewed, corrected and edited and it may contain inaccuracies, i
l D>
u-J
1 230 1
UNITED STATES NUCLEAR REGULATORY COMMISSION
[O 2
ADVISORY COMMITTEE ON REACTOR SAFEGUAT.OS 3
4 449TH ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5
6 U.S.
Nuclear Regulatory Commission a
7 Two White Flint North, Room 2B3 8
11545 Rockville Pike 9
Rockville, Maryland 20852-2738 10 11 Tuesday, March 3, 1998 12 13 The Committee met pursuant to notice at 8:30 a.m.
14
)
15 MEMBERS PRESENT:
16 ROBERT SEALE, Chairman, ACRS 17 MARIO H.
FONTANA, Member, ACRS 18 GEORGE APOSTOLAKIS, Member, ACRS 19 JOHN BARTON, Member, ACRS 20 THOMAS KRESS, Member, ACRS 21 DON MILLER, Member, ACRS 22 DANA POWERS, Member, ACRS 23 WILLIAM SHACK, Member, ACRS 24 ROBERT UHRIG, Member, ACRS 25 DR. WALLIS, Member, ACRS
[\\
ANN RILEY & ASSOCIATES, LTD.
\\>
Couri Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
231 1.
PROCEEDINGS 2
[8:30 a.m.]
3' CHAIRMAN SEALE:
Good morning.
The meeting will 4
come to order.
I'm pleased to note that our two colleagues 5
joined us today, so we are ready to sally forth.
6 This is the second day of the 449th Meeting of the l
7 Advisory Committee on Reactor Safeguards.
During today's 8
meeting, the committee will consider the following:
9 Evaluation of -- elevation, I'm sorry, of CDF to a l
10 fundamental safety goal, and possible revision to the safety 11 goal policy statement; proposed final SRP sections and 12 regulatory guides for risk-informed performance-based-13 regulation and related matters; report of the Subcommittee l
14 on Reliability and Probabilistic Risk Assessment; discussion 15 of Commission members' paper on risk-informed 16 performance -based regulation; and proposed ACRS reports.
17 This meeting is being conducted in accordance with 18 the provisions of the Federal Advisory Committee Act.
Dr.
19 John Larks is the designated federal official for the L
20 initial portion of the meeting.
21 We received no oral -- pardon me.
We received no 22 written statements or requests for time to make oral 23 statements from members of the public regarding today's i
I 24 sessions.
A transcript of portions of the meeting is being j
25 kept, and it is requested that the speakers use one of the O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034 l
1 L
232 1
microphones, identify themselves, and speak with sufficient
()
2 clarity and volume so that they can be readily heard.
3 Do we have any special comments, any issues we 4
need to alert members on before we get started this morning?
5 Anyone?
John?
Gail?
Okay.
6 Any of the members have any issues?
7 Okay.
Well, our first topic is the elevation of 8
CDF to a fundamental safety goal, and possible revisions to 9
the safety goal policy statement.
Dr. Apostolakis is the 10 chairman of that subcommittae effort, and I will turn it 11 over to you, George.
12 DR. APOSTOLAKIS:
Thank you, Mr. Chairman.
13 The Subcommittee on Reliability and Probabilistic 14 Risk Assessment met with the Staff on February 20th, and we
( ~\\
(,/
15 discussed the issues that are involved in such an endeavor.
j l
16 We were told that the Staff is inclined, although they were i
17 not sure at that time, to recommend to the Commission that 18 such an elevation of CDF to fundamental level be postponed.
19 So today we are going to hear again from the Staff and they 20 will tell us how their thoughts are evolving, and then we 21 have to write a letter, I believe, this time, and you have a 22 draft of the letter already in your hands.
I believe Mike 23 Markley has handed it out.
And without further ado, Mr. Tom 24 King of the Office of Research.
25 MR. KING:
My name is Tom King.
I am with the ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
233 1
Office of.Research, and with me at the table is Scott i(
)
2
'Newberry from.NRR.
Scott is Gary Holahan's deputy.. Gary j
3 couldn't make it today.
4 What I want to do is go back'and cover a little f
5 bit of the background we covered at the subcommittee meeting i
6 on the 20th, and then provide an up-to-date status of where 7
we stand on lod.ing at the issues, and what our -- at this 8
point what our intent is in terms of a recommendation to the 9
Commission.
10 As Dr. Apostolakis said, basically the bottom line 11 is these are tough issues, they are not anything that has at l
12 this point an immediate need to jump into it and modify the
-13 statement, and we.think it is probably worthwhile taking 14 some more time and carefully looking at these.
But we'll 15 talk about that some more.
16 What I wanted to go through then today mainly was 17 our approach, plan and approach for assessing the need to 18 revise the' safety goal policy.
We will talk about the --
19 there's eight issues now that are candidate areas for 20 update.
We will talk about those into whatever level of 21 detail you would like to get into; talk about what we think 22 needs to be considered hefore we recommend whether 23 proceeding with the revision; and we would like a letter, at 24 least on our plan and approach -- we are not asking for a j
25 position on any of the eight candidate areas at this time, l
k O-ADN RILEY & ASSOCIATES, LTD.
Court Reporters i
1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 i
-l
234 1
but at least our approach for dealing with those, we would
()
2 like the Committee to write a letter on that.
3 By way of background, as you recall, I think the i
4 idea of revising the safety goal policy was put on the front 5
burner back in '86 by ACRS when they sent a letter to the 6
Commission recommending that the CDF be elevated to a
{
7 fundamental safety goal.
8 That was followed in '87 by two more letters from 9
ACRS.
One suggested a way to define the term adequate 10 protection, using the safety goals; and another one 11 suggested a policy statement on defenpe-in-depth.
We had a 12 subcommittee meeting on that last year, and at that time it 13 was the Staff's view that if we are going to do something on 14 defense-in-depth in terms of a policy statement, it makes
()
15 sense to do that in the context of the safety goal policy, 16 not a separate policy statement.
We still -- the Staff 17 still believes that that is the case.
18 Following the ACRS letter on elevating CDF, the 19 Chairman asked the Staff to consider that, and we providec' a 20 paper in September '97, SECY 97-208, to the Commission that 21 gave our plan and schedule for doing that consideration; i
22 gave some pros and cons of elevating CDF; it also identified 23 a number of other issues that we felt if we were going to
. 24 get into the safety goal policy, it was worth looking at 25 some other areas.
We identified eight candidate areas for OA ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 2
A
235 l'
potential update, and we said to the Commission we would be
()
2 back'to the end of March with a recommendation.
3 Subsequent to that, we also had the NRC strategic 4
plan issued in September, which had some items in it related 5
to things that are covered by the safety goal.
It talked 6
about having no core damage accidents.
It talked about a 7
conditional containment core damage probability.
It also 8
talked about radiation exposure to the public.
So there's 9
some overlap in coordination and consistency that needs to 10 be looked at with that document.
11 We also completed work on our general reg guide 12 and SRP on risk-informed regulation which used the safety 13 goal to derive some criteria, some guidelines that were in 14 that document.
()
15 And then just very recently the Commission has a i
.1 16 draft White Paper circulating for review on risk-informed 17~
regulation.
The Committee will be talking about that later 18 today, but that has some overlap as well.
19 So a number of things have happened that we feel 20 could impact the safety goal policy, certainly need to be 21 coordinated with it.
All of these things need to be 22 carefully considered if we are going to get into revising 23 the safety goal policy.
24 Just quickly, to refresh your memory on the eight 25 candidate areas that were identified in SECY 97-208 as ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
r l
236 1
potential things that might be revised if we go into the
(
2 safety goal policy, certainly elevating the CDF was at the g ;
3 top of the list.
There were three items that came out of 4
our work on the.isk-informed reg guides and SRP, that was 5
use of the goals in plant-specific regulatory activities.
6 The original policy statement said they are supposed to be 7
used for generic type activities.
We derived some 8
plant-specific guidelines from the goals, and have applied 9
those in the reg guide.
So we feel it might be worth some 10 revision there.
)
11 We have developed a criteria on large early 12 release frequency which might make sense to put into the 13 safety goal policies, particularly since the policy 14 statement now has what's called a general performance mk,)
15 guideline which the Staff looked at and said wasn't 16 practical to implement because it was more restrictive than i
17 the OHOs, so it might make sense to remove that and put in i
18 something like LERF.
19 We have developed a more comprehensive set of 20 guidance for how to treat uncertainties.
The safety goal 21 policy talks about the treatment of uncertainties.
It might 22 be worth updating that to reflect our current thinking 23 there.
24 DR. APOSTOLAKIS:
Tom, just to clarify things a 25 little bit.
If we are to revise the safety goal policy
(
ANN RILEY & ASSOCIATES, LTD.
\\
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 237 1
statement, we don't necessarily have to address all of the (o) 2 issues that are being identified here.
We could just x_-
3 elevate the CDF.
4 MR. KING:
That's correct.
That's correct.
i 5
DR. APOSTOLAKIS:
So we don't have to worry about 6
all-or-nothing?
7 MR. KING:
Right.
This is not an all-or-nothing.
8 DR. KRESS:
But it makes sense, if you are going 9
to work on the safety goal policy statement, to fix all the 10 things you think need fixing in it while you're at it; 11 right?
12 MR. KING:
As we talked about at the subcommittee, 13 I think some of these you have to treat as a package.
It's 14 not a package of eight, but I think there are -- and there O'i
(,,/
15 will be a slide that talks about coupling some of these i
16 together.
17 CHAIRMAN SEALE:
These are just issues where there 18 is overlap and you have to consider them, or you run a very 19 real risk of finding yourself in a logical cul-de-sac.
If 20 you approach a particular issue from one direction, then try 21 to approach it from a different direction, you may not be 22 able to get yourself a consistent answer, unless you look at 23 them as a pacP. age.
24 MR. KING:
Correct.
25 DR. APOSTOLAKIS:
But my point is that if certain I
)
k- /
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 238 1
things need fixing and we are aware of them, that's fine.
l 2
The point is that, you know, it took about what, six years, N
3 seven years last time to issue a statement.
They started a 4
debate in '79 or
'80, and a statement was issued in
'86.
I 5
would have to have to go through that process again.
You 6
know, to have your statement in the year 2006.
So I think 7
that is an important consideration.
So we have to weigh the 8
relative importance, I think, of the various issues and 9
decide what is the appropriate course of action.
10 MR. KING:
Yes, but at that point they were 11 starting from point zero.
12 DR. APOSTOLAKIS:
I know.
13 MR. KING:
And we have got all that behind us.
14 DR. POWERS:
George, I mean there is another way O)
(,
15 of looking at it, and the other way of looking at it is no i
16 matter what change I take, it's going to take six years, so 17 I might as well change everything all at once.
18 MR. KING:
That's another way to look at it.
19 DR. APOSTOLAKIS:
If you convince me that it will 20 take six years anyway, then I'm with you.
But I wouldn't 21
--I'm reluctant.
I don't want to think that it will take 22 six years, because, you know, we have other things to do as 23 well.
24 So your answer is that no, we don't have to do 25 everything we're discussing here, you are just presenting
[)
ANN RILEY & ASSOCIATES, LTD.
\\- '
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
3 239 1
thoughts and ideas?
(' J 2
MR. KING:
This is the list of areas we are 3
considering for update.
That doesn't mean you have to take 4
them all.
You can pick one or, you know, some grouping or 5
pick them all.
6 DR. APOSTOLAKIS:
Fine.
7 MR. KING:
But you're right, some are more 8
important than others, in my mind, and part of the task 9
would be te figure out which those are.
The benefits, 10 impacts, and so forth.
11 Just to complete the list, the remaining four 12 items were things that came up subsequent to issuing the 13 policy statement through implementation and use of it.
If 14 you read the policy statement, it has a quantitative goal O) 15 and societal risk that when you get to -- the qualitative
(
16 goal and societal risk, but when you get to the quantitative 17 measures, the QHOs and how they are calculated, they are all 18 calculated based upon individual risk.
So it makes sense to 19 go back and put something more quantitative in on societal 20 risk.
The policy statement is silent on land contamination 21 or other environmental impacts.
Should it say something 22 about that is an issue.
23 DR. APOSTOLAKIS:
I'm a little confused here, and 24 I am looking for my policy statement, but of course I cannot 25 find it.
[
ANN RILEY & ASSOCIATES, LTD.
\\-
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
240 1
I know that we have been saying this for a long 7( )
2-time, but the policy statement does not address societal 3
issues.
4 MR. KING:
It does qualitatively.
5 DR. APOSTOLAKIS:
Well, even quantitatively it 6
says you should go out 10 miles --
7 MR. KING:
What is it, Mike, seven?
8 DR. BARTON:
Eight.
Tab eight.
9 MR. KING:
Page 26.
10 DR. POWERS:
It is my perception that a societal 11
' risk measure attempts to recognize that within any group of 12 people, you will have some more susceptible to radiological 13 damage than others, and that the societal risk measures 14 differ from individual risk measures, and that you try to D)
(,
15 account for that.
Is that your feeling, that the
~
16 qualitative health objectives do that?
17 MR. KING:
It's true within a given population 18 some people are more susceptible to radiation effects than 19 others.
The way it's calculated, that differentiation isn't 20 made, it's calculated based upon an average individual.
21 DR. POWERS:
What I'm distinguishing between is 22 the calculation which you concede is an individual risk 23 measure.
i 24 MR. KING:
Yes.
25 DR. POWERS:
And the qualitative, whien you are
/
ANN RILEY & ASSOCIATES, LTD.
\\-
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
241 1
arguing is a'socittal measure.
(
-2 MR. KING:
Well, if you read the -- under the
-3 safety goal policy, right up front it talks about 41 qualitative safety goals.
The first one says individual 5
members of the public should be provided a level of 6
protection.
And the second one says societal risks to life 7
and health should be comparable to or less.
8 DR. APOSTOLAKIS:
But it also gives a number.
It 9'
says the risk to the population in the area near a nuclear 10' power plant of cancer fatalities that might result from 11 nuclear power plant operations should not exceed one-tenth 12 of 1 percent of the sum of cancer fatality risks resulting 13 from all other causes.
14
_And later on, it says that the Commission has
()
15 defined the population generally considered subject to i
16 significant risk as the population within 10 miles of the
-l l
17 plant site.
18 MR. KING:
Yes.
{
19 DR. APOSTOLAKIS:
Well, why isn't that societal 20 risk?
I mean you have to take into account the number of 2.1 people living within 10 miles of the site.
22 DR. KRESS:
But then they give you a guidance on 23 how to calculate that, and they say divide by the population 24 in that region.
25 DR. APOSTOLAKIS:
Not in the statement.
I didn't O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
242 1
4
-see that.
Maybe that's the way it was implemented, but --
2 let me give you another -- ask you a question.
3-Would evacuation plans have an impact on what this 4
statement calls societal risk?
5 MR. KING:
Yes.
6 DR. APOSTOLAKIS:
Yes.
So it's societal risk.
7 You calculate in the number of people that are affected.
8 MR. KING:
Yes.
9 DR. APOSTOLAKIS:
Whereas evacuation would have no 10 impact on individual risk because the Commission says take 11 the individual that's within a mile, and if you don't find 12 anybody, postulate one.
So there is no question of 13 evacuation'there.
I mean you have one guy there.
14 MR. KING:
As a minimum.
15 DR. APOSTOLAKIS:
As a minimum.
16 MR. KING:
But the way.it's calculated, you do 17 take credit for evacuation plans.
18 DR. APOSTOLAKIS:
Even for individual risk?
19 MR. KING:
For --
20 DR. APOSTOLAKIS:
That doesn't make sense, because 21 you have to assume that he's there.
22 DR. POWERS:
What he's saying is that in practice, 23 they give credit for evacuations.
i 24 DR. APOSTOLAKIS:
In calculating what?
The 25 societal part?
i ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034 U
243 1
DR. POWERS:
No.
In calculating the QHOs, whether
()
2 it's prompt or latent, credit is given for evacuation.
3 DR. APOSTOLAKIS:
.The Commission says very clearly 4
that you have to look at the people living within a mile of 5
the. site boundary, okay?
And if you don't find anybody, 6
then you assume that there is_somebody there.
How can 7_
evacuation affect that?
8 MR. KING:
It can affect, if you've got more than 9
one person there, you give' credit for moving those people 10 out of the way, but within the constraints of the evacuation 11 plan.
So if you've got a thousand people within a mile, you 12 give credit for those people moving.
13 DR. APOSTOLAKIS:
Well, it seems to me that if I 14 didn't know how these things were calculated, and all I read
()
15 was the Commission's policy statement, I would say that is 16 an individual risk and a societal risk.
Now it may be 17 implemented in the wrong way, but that's what the Commission 18 says.
19 DR. BARTON:
George, the way it's implemented, you 20 don't worry about the one individual within a mile, because 21 you usually use a keyhole effect and you evacuate everybody 22 within two miles, and then you go out, depending from there 23 on which way the wind blows.
So you've taken care of 24 everybody within that mile by the way you implement the 25 evacuation plan, the emergency plan.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
244 1
DR. APOSTOLAKIS:
Okay, let's say the calculation I
2' for the individual is the same.
Now let's say that I have D)
-3 900' people or a thousand people within 10 miles.
Wouldn't
{
4 this statement require that I calculate the number of people 5
out of these thousand who will contract cancer as a result l
6 of the accident?
7 MR. KING:
The way the calculation is done, it's a probabilistic calculation that looks at-the various accident o
9 scenarios and how fast the release ~ occurs, and so forth.
10 And it samples the weather, and it basically, in doing that, 11-all of these give credit for evacuation.
The way they give 12 credit for evacuation on -- in an average. sense is they 13 assume 99.5 percent of the people will evacuate, and they 14 will start to move within two hours or some reasonable 15 amount of time after the beginning of the accident, and they 16 will move at a certain speed radially outward, end once they 17 reach the 10 miles or the one mile, depending cn what you 18 are calculating, they are essentially out of t.he plume.
19 DR. APOSTOLAKIS:
Right.
You remember the 20 infamous risk curves from the reactor safety study in the 21 executive summary?
It's frequency per year, number of 22 people, right?
23 MR. KING:
Yeah.
1 24 DR. APOSTOLAKIS:
That's a measure of societal 25 risk?
ANN RILEY & ASSOCIATES, LTD.
/
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
245 1
DR. WALLIS:
Tom, when you say these things are
()
2 assumed, is this simply an assumption, or is it based on 3
experience of some sort?
Or how do you know it's a 4
reasonable-assumption?
You have a whole string of 5
assumptions here.
6 MR. KING:
Yes, it's based upon an analyst's 7
estimate of looking at a site, an actual site evacuation 8
plan, what a reasonable estimate would be for how fast you'd 9
get people moved and how fast they move.
10 DR. WALLIS:
I don't know what reasonable means.
11 Is it based on experience of evacuation with comparable 12 events somewhere else?
13 MR. KING:
It's based upon experience.
Back when 14 NUREG-11
)
15 DR. WALLIS:
What sort of experience?
16 DR. APOSTOLAKIS:
I think chemical.
17 DR. WALLIS:
Chemical?
18 MR. KING:
Yes.
Yes.
At one time I looked this 19 up when we were doing some work, and there is -- I could get 20 you some references, but it's not, nuclear experience.
It's 21 based upon industrial type accidents.
22 DR. WALLIS:
Okay.
It would be nicer to say then 23 it's based on experience.
When you use words like assume 24 and reasonable, that sounds very suspicious to me.
25 MR. KING:
It's based upon experience.
)
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
i 246 1-DR. WALLIS:
Okay.
Thank you.
\\s.)/
2 MR. KING:
And actually, as I recall, they feel q
l 3
these are fairly conservative assumptions in terms of how 4
fast people move.
But there is an assumption that some 5
people don't. move.
6 DR. APOSTOLAKIS:
Okay.
But coming back to the 7
fundamental question, the reactor safety study did produce 8
risk curves for societal risk in the sense that I can go 1
9 there and look at the figure, and I can figure out the I
10 frequency per year of harming X or greater people.
That's a definition of societal risk.
You know, there are major 12 problems because the uncertainties were not displayed, but 1
13 the essence of the argument was there.
i 14 I have seen many full scope PRAs that.still do the
()
15 same thing.
The Seabrook PRA comes to mind; Indian Point, 16 Zion and so on.
So I don't know why we say that the 17 statement, the policy statement is individual risk only.
18 Now the calculation is something'else, and there may be a 19 lot of commonalities in how you calculate these things, but 20 if I -- if I take two sites and in one I have a million i
21 people within 10 miles, and in another I have a thousand, 22 even though the calculation may have similarities, the 23 societal risks must be different, simply because I have many 24 more people.
25 MR. KING:
Exactly.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
247 1-DR. APOSTOLAKIS:
So this statement, then,
[)
2 includes individual risk criteria and societal risk
'u 3
criteria,. independently of how they are calculated?
4 MR. KING:
Clearly the policy statement addresses
'S those in a qualitative sense.
6' DR. APOSTOLAKIS:
And quantitative.
The risk to 7
the population in the area should not exceed one-tenth of 1 8
percent of the sum of cancer fatality risks resulting from 9
all other causes 10 MR. KING:
Yes, I think what you're getting at is 11 maybe the words in the policy statement don't need to be 12 changed, but maybe the way it's implemented needs to be 1
13 changed.
14 DR. APOSTOLAKIS:
Maybe.
Maybe so.
But I mean 15' the reason why I insist on this is because in a few past 16 meetings, we're saying, you know, the person standing on his 17 own individual risk and everybody was not in --
18 DR. SHACK:
But that sum just scales, doesn't it?
19 As a matter of, you know, the risk from all other sources 20 just scales with the population, so that you end up dividing 21 by the population again.
So there's no absolute number 22 there.
I mean it still comes down to -- it's a per-person 23 kind of risk.
24 DR. APOSTOLAKIS:
Well, I don't know.
For I
25 example,.if you have to evacuate a million people versus a
()
ANN RILEY & ASSOCIATES, LTD.
Court Reporters i
1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 248
.1 thousand, you may have a problem there.
Maybe the scaling r
21 is not, you know --
. (.,
3 DR. SHACK:
Okay, the scaling, but would it be 4
societal only to the point that it didn't scale?
5 DR. APOSTOLAKIS:
That's right.
Exactly.
Simply 6
because it's calculated that way, yes.
Now I don't know the' 7
details of the calculation.
-8 Do you want to say something, Rick?
You're an 9
expert on this.
10 MR. SHERRY:
The --
11 CHAIRMAN SEALE:
Identify yourself, Rick.
12 DR. APOSTOLAKIS:
Who are you?
13 MR. SHERRY:
Rick Sherry, Senior Fellow.
14 Basically I think we are getting lost or being
) )-
15 fooled by the terminology societal and individual.
The way j
16 that it's implemented is basically you calculate the
{
17 consequences within a certain region, and then you divide by 18 the population within that region.
l t19 DR. KRESS:
Where does it tell how to implement 20 it?
That's -- is there a reg guide or --
21 DR. APOSTOLAKIS:
Exactly.
22 MR. SHERRY:
Well, there's no reg guide, but it's 23 based on the way that the safety goal policy statement is 24 worded, as a fraction of the existing risk --
25 DR. KRESS:
You're saying it is in the policy ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
L-l 249
'l statement?
(
)
2 hUl. SHERRY:
The statement that Tom King 3
mentioned,.the one-tenth --
4 DR. APOSTOLAKIS:
I mentioned that,'yes, the 5
Lone-tenth of 1 percent of all cancers.
l
.6 MR. SHERRY:
One-tenth of 1 percent of the 1
7 existing fatality.or latent cancer fatality risk.
8 DR. APOSTOLAKIS:
Yes.
9 MR. SHERRY:
Which implies that you multiply that 10 by the population exposed.
11 DR. APOSTOLAKIS:
But isn't there an-assumption 12 there, Rick, I mean if you are saying that even for 13 individual risk, evacuation is a factor, isn't that 14 assumption that the ease with which I can evacuate one
)
15 person is the same as evacuating a million persons?
16 Now maybe that's the way it's calculated, but in 17 terms of actual, you know, what it's supposed to be, it 18 doasn't-sound right to me.
19 MR. KING:
No, that's not the way it's calculated.
20 In 1150, there were different assumptions on how fast people 21 moved because the sites were different.
22 DR. APOSTOLAKIS:
Different in the sense of how 23 many people were around?
24 MR. KING:
Different on how many are around, 25 different on how fast they started to move, and how fast i
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 l
250 1
they moved, (m}
2 DR. APOSTOLAKIS:
Okay.
So it seemsEto me, then, 3
that those guys were calculating societal risk, then they 4
divided by the population and calculated something that they 5
called individual risk, but maybe it's not quite individual 6
risk.
You see, it's a normalized societal risk, and I 7
question whether it's the same as individual risk, the way 8
the Commission defines it.
9 The Commission says you will take people within a 10 mile to calculate, you will take an individual living within 11 a mile, and if you can't find anybody, postulate one.
They 1
12 say nothing about postulating populations.
Now, maybe for i
13 convenience, and from habit, people are taking the overall 14 population, they doing whatever calculations they are doing,
()
15 then they divide by the original number and say, well, this i
16 is in individual risk.
17 Well, I am not that that is individual risk the 18 way the Commission intended it to be.
Now, the difference 19 may be small, but the question is, is it a fair statement tc 20 say that the Commission's policy statement ignores societal 21 risk?
I don't think so.
I read it, and I don't think so.
22 Maybe the calculations are messed up, but the Commission's 23-intent was very clear.
24 MR. SHERRY:
My opinion is that, since these goals 25 do not scale with population density, then, in my mind, they O>
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
251 1
don't' reflect societal risk.
] )
2 DR. APOSTOLAKIS:
Why do you say that they don't 3
scale?
4 MR. SHEkRY: _If you had two sites, the only 5
difference being the population density, then they would 6
come out with equivalent individual risk values, even though 7
the total number of -- the total consequences would be much 8
greater in one case than in the other.
9 DR. APOSTOLAKIS:
Well, but, again, that may be 10 the way that it is calculated, because, I mean, one obvious 11 example is wouldn't there be a different impact from 12 evacuation if you had a million people?
13-MR. SHERRY:
It's hard to say which direction it 14 would be because it depends on other site-specific factors, 15 which are --
16 DR. APOSTOLAKIS:
Sure.
17 MR. SHERRY:
Which impact evacuation.
'18 DR. APOSTOLAKIS:
Which are also ignored.
19 MR. SHERRY:
No, they are not ignored.
They are 20 considered in coming up with the evacuation time estimates.
21
'DR. APOSTOLAKIS:
Would they consider what kind of 22 people you would have in those million people?
I mean the 23 distribution.
You would have more babies.
You would have
'24
-- I mean if you really want to implement the statement as 25 stated, then you would really have to know, for example, ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
252 1
whether this is in Florida or in Idaho.
Wouldn't you?
C\\
(
)
2 But.naybe the calculations don't do that.
That's 3
the fault of the calculations, not of the Commission.
4 MR. SHERRY:
Actually, the way it is implemented 5
is not that sophisticated.
They don't actually locate the 6
people within the 10 miles, exactly where they are, it is 7
based on, you know, much grosser data that that.
And, no,
8 they don't take into account differences in the 9
demographics.
10 MR. KING:
No, they do look at the population 11
' distribution around the site.
If there is a city or a small 12 town located in one sector, that is modeled.
13 DR. APOSTOLAKIS:
Okay.
14 MR. KING:
And if the prevailing winds are always A(,)
15 in that sector, that's modeled.
16 DR. APOSTOLAKIS:
But it is modeled as if they 17 were typical individuals that could get up and leave when 18 told and so on, but it is not modeled in the sense that, 19 gee, in this particular town, we have a larger fraction of 20 senior citizens, for example.
21 MR. KING:
No, it doesn't get into that level of 22 detail.
23 DR. APOSTOLAKIS:
But the policy statement doesn't 24 say that you should do things like that.
25 MR. KING:
No.
[\\
ANN RILEY & ASSOCIATES, LTD.
\\'"
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
253
'l DR. APOSTOLAKIS:
So the real issue, in my mind,
(
2' is that'the calculation has simplified the policy of the 3
Commission-to the point where there is no distinction 4
between. individual risk estimates and societal risk 5
estimates.
All you have to do is take the individual risk 6
and multiply by the number of people around.
7 The fact that you have a million-people, though, 8
may introduce additional factors, so scaling may not be 9
appropriate here.
10 DR. SHACK:
Yeah, but I mean I would think a l
11 society goal would be one where, you know, we don't worry 1
12 about non-linearity, you know, deviations from linearity in 13 the scaling, but one that is definitely steted as the 14 probability that you will have 100 deaths is -- bing --
()
15 which would make it, you know, very definitive that, you 16 know, it makes a difference just how many people are 17 affected -
i l
l 18 I mean I agree there's some -- you know, it 1
i 19 doesn't scale perfectly, but, you know, it scales pretty 20-close.
l 21 DR. APOSTOLAKIS:
Oh, it might.
But I think the 22 fundamental issue here is -- are we right in saying that the 23 existing statement does not address societal risk?
I don't 24 think so.
It is very clear on the first page.
Say we are l
25 going to look at two things, individual risk and society b
ANN RILEY & ASSOCIATES, LTD.,
N
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
254 1
risk.
Then, two lines down, quantitative goals.
The risk h) 2 to an average individual in the vicinity of plant, of plant q.
3 fatalities, and then the risk to-the population from 4
cancers.
So, actually, it is delayed, right.
5 So -- and then they address the issue of societal 6
risks to life and health from nuclear power plant operation 7
should be comparable to or less than the risks of generating 8
electricity by viable competing technologies and so on.
9 They bother to tell you how far from the site boundary you 10 should go.
i 11 Clearly, these are not people who intend to have 12 those two things being identical.
They very clearly say 13 that for individual plant fatalities you have to go only one I
14 mile.
For the societal, you have to go 10 miles from the qQ 15 plant.
And beyond would generally be -- yeah, okay.
16 So, leaving the calculation aside, it seems to me 17 this statement addresses individual and society risk.
18 MR. KING:
I agree.
The policy statement tries to 19 distinguish between the two.
20 DR. APOSTOLAKIS:
Okay.
21 MR. KING:
And I think what we have been talking 22 about is the way it is calculated, does it really accomplish 23 the intent of the policy statement?
And the discussions we 24 have been having here are the same kinds of discussions we 25 have been having in trying to prepare this paper, and we h
ANN RILEY & ASSOCIATES, LTD.
\\/
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
255 1
haven't reached a conclusion on it.
~
[\\s,/\\
2 DR. KRESS:
if you read the clarity of the safety 3
goal policy statement, along with their discussions, you 4
arrive h a conclusion that they meant for you to calculate 5
it by. dividing by the population.
Because, especially in 6
the parts where they are comparing the result to the prompt-7 fatality of the latest cancer and how they might, which one 8
might be controlling and so forth.
They meant for you to 9
make that division.
10 So it is fairly clear when you read the whole 11 thing, although it doesn't explicitly say that anyvhere.
12 DR. APOSTOLAKIS:
Yeah.
13 DR. KRESS:
krid in the discussion, they meant for 14 you to do the division by the population, and make it an Ot-(,,/
15 individual risk.
16 MR. KING:
I think you could certainly interpret 17 it that.
I am not sure everybody would interpret it that 18 way.
19 DR. KRESS:
I think it has to be an interpretation 20 because they don't I can't find it explicitly stated.
21 MR. KING:
And if you look at what we do in 22 regulatory analysis, we don't calculate individual risk, we 23 calculate a population dose, a person / rem.
To me, that is 24 more a societal measure of what is the society getting from 25 this particular accident.
And then that total measure then ANN RILEY & ASSOCIATES, LTD.
s Court Reporters 11250 I Street, N.N.
Suite 300 Washington, D.C.
20005 (202) 842-0034
256 1
factors into the cost benefit equation.
()
2 DR. KRESS:
Then you would go back to the 50 mile.
3 MR. KING:
They also use 50 miles, which is 4
different than the policy statement.
5 Dk. KRESS:
But it is a better reflection of the 6
total societal --
7 MR. KING:
More people, more person / rem.
8 DR. APOSTOLAKIS:
Does evacuation affect both 9
prompt,and latent?
10 MR. KING:
Yes.
11 DR. KRESS:
See, in there, they talk about moving 12 from 50 to 10 because that gives you a worst situation -- a 13 worser situation, and the reason is because you are dividing 14 through by the population.
And they are getting a higher
)
15 dose.
So it is clear from that statement they meant for you 16 to divide through by the population.
That was the intent of 1
17 how they calculated it.
18 DR. WALLIS:
While we are talking about distance, 19 I was asked by our local radio station some time about this 20 10 miles and 50 mile stuff and this was not too long after i
21 Chernobyl.
And it was obvious that, in the case of 22 Chernobyl, we were talking about thousands of miles.
I am 23 sure this isn't a suitable topic for this discussion today, 24 but it is always mysterious to me how they can legally sort 25 of decide that no radiation goes beyond 10 miles, if that is ANN RILEY & ASSOCIATES, LTD.
Court Esporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
257 1
the intent.
When, in a major accident, it may well be that
).
2 the plume dynamics are such that the danger is much greater 3
'at.a greater. distance.
J 4
MR. KING:
Depending on the accident and the 5
weather conditions at the time, that's right.
People within 6
10 miles may not get much, that's true.
But, again, this is
-7 not looking at specific accidents, it is looking at an 1
8 average of a spectrum of accidents and a spectrum of weather 9-conditions that, you know, on the average takes care of 10-those individual specifics that may throw you off.
11' DR. POWERS:
When you say it could be that you get 12 these Chernobyl-like energetic releases, you are saying that 13.
in.the sense of, yes, it is physically possible, not that it 14 is-probable?
O)
\\,j 15 MR. KING:
Yes, it is physical possible.
And when 16 you do the calculaff.on, you look at all the accident 17 scenarios.
You know, you look at the early -- you know, the 18 early contain aent f ailures with an energetic release, just 19 as well as you do with the containment bypasses, and --
20 DR. KRESS:
None of those have the energy --
21 DR. POWERS:
Yeah.
Even when we --
22 DR. KRESS:
-- to loft it that the Chernobyl did 23 it.
None of those sequences have that kind of energy in it.
24 DR. POWERS:
I mean I have certainly heard --
25 DR. KRESS:
Because it has been already relegated i
ANN RILEY & ASSOCIATES, LTD.
Court. Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 i
(202) 842-0034 i
258 1
to such a low probability for these reactors that it is not 2
even part of the PRA.
v 3
DR. POWERS:
I have certainly heard the words said 4
that it is physically impossible to get that-kind of plume 5
lofting.
6 DR. KRESS:
I have heard that, but I don't quite 7'
believe that either.
I just think the probabilities are low 8
enough that it doesn't show up in PRA space as being very 9
important.
10 MR. KING:
The probabilities of energetic releases l
11 are low, but they are not ignored.
I mean we have 12 calculated the -- looked at the energy. content of the j
13 release and its probability in doing these -- a number of 14 calculations where we are revising Part 100, for example.
15 DR. POWERS:
And it is my impression, or I have 16 certainly heard people say, that when_you did that, you 17 couldn't find anything that even approached, within an order i
18 of magnitude, the energies that were involved in the l
19 Chernobyl accident.
i 20 MR. KING:
I never heard anybody say that, but it 21 wouldn't rurprise me if that is true.
22 DR. POWERS:
I think -- I think what makes the 23 statement very feasible is that the Chernobyl accident had a 24 driver that continued the lofting for a substantial period 25 of time.
And when you integrate over that energy, it is a O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
259 1
lot of ene)-
2 CHAIRMAN SEALE:
Big coal fire.
~3 DR. APOSTOLAKIS:
So what is the conclusion of all 4
this?
Can we say again that the policy statement does not i
5 address societal risk, or that is a false statement?
I 6
DR. KRESS:
It does and it doesn't.
7 DR. APOSTOLAKIS:
Well, the Commissioners admit 8
that if one meets the prompt fatality objective for an 9
individual, it is very, very likely that the quantitative 10 objective for delayed death will be met as well.
11 DR. KRESS:
That makes it very clear that they are 12 both individual risk and that you divide through the 13 population there.
Because that is the only way that 14 statement can be true.
O
\\,)
15 So my feeling is it doesn't really address, 16 although it says the words, it doesn't really do --
17 DR. APOSTOLAKIS:
What else could they say to 18 leally address it, the societal issue?
19 DR. KRESS:
Well, you could talk about total 20.
deaths or land contamination.
21 MR. FULFORD:
The societal risk goal would say 500 22 person / rems, that would be a load on society.
While I do; t 23 agree with the linear hypothesis, that would be the societal 24 goal.
And that would differ as to whether there is a 25 thousand people or a million people.
That would come up O
ANN RILEY ~ ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
i 260 1
with a different answer whether there was a million people
/
2 or.a thousand people within 10 miles.
3 DR. APOSTOLAKIS:
Socif they wanted to put it in 4
terms of deaths, they should say, what, no more than 1/10th 5
of_1 percent of the total number of deaths you expect from 6
all other --
7 MR. FULFORD:
No.
An absolute number in 8
person / rems.
9 DR. KRESS:
You're right.
10 DR. APOSTOLAKIS:
No, because they are giving 11 everything as a fraction.
12 DR. KRESS:
Well, you could back --
13 DR. APOSTOLAKIS:
So you can say I take this city
-14 of a million people.
()
15 DR. KRESS:
Yeah.
16 DR. APOSTOLAKIS:
How many counts of deaths do I 17 have a year?
18 MR. FULFORD:
X.
19 DR. APOSTOLAKIS:
Well, 1/1,000th of that X is my 20 goal.
21 DR. KRESS:
Yeah, would could do that, that would 22 be a way to do it.
23 DR. APOSTOLAKIS:
- Yeah, s
24 DR. KRESS:
And you could further express that in 25 terms of person / rems.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 j
i
I 261 1
DR. APOSTOLAKIS:
And why.aren't they saying that
(
'2 now?
I.think'they_are-saying that.
3 DR. KRESS:
No, I don't think so.
)
4 DR.-APOSTOLAKIS: 'The risk to the population 5
should not exceed 1/10th of 1 percent of the sum of cancer-6-
. fatality risks resulting from all other causes.
Why not?
7 It's the total number.
On.page 27, at the very top.
8 DR. KRESS:
Yeah, but you can't just read one part
~
9 of this goal and then --
i 10 MR. KING:
Then it is different from site to site.
'11 DR. APOSTOLAKIS:
Of course it is different from 12 site to site.
But that's how I interpret this.
From all 13 other causes.
So, you know, I have a million people, 3,000 14 of them die a year of cancer or other causes, so the nuclear
)
15 plant should contribute at most three to those 3,000.
16 That's what the Commissioners are saying.
1/1,000th --
1 17 DR. MILLER:
It doesn't matter then how many 18 people you have living near the site.
That's the way I read 19 that paragraph.
20 DR. APOSTOLAKIS:
It does.
It doesn't matter now 21 because of the way it is calculated.
But I want to separate 22 the calculation from the statement and the Commission's 23 intent.
Because we cannot say the Commission did not
'24' address societal risk,'when I read it and I interpret it as 25 if they did.
j O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250-I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
262 1
DR. KRESS:
But Bill is right, it cancels out.
2 You know,'if the --
3 DR. APOSTOLAKIS:
That's a separate issue.
4 DR. KRESS:
No, no.
The cancer rate is given 5
something, some value, multiply it by the population, you 6
get the total number of deaths.
You could just as well use 7
the --
8 DR. APOSTOLAKIS:
For prompt fatalities for 9
individual risk, you say that evacuation plays a role?
10 MR. KING:
Yes.
11 DR. APOSTOLAKIS:
Now, the statement doesn't say 12 anything like that.
13 MR. KING:
The policy statement in either case 14 doesn't talk about evacuation playing a role.
()
15 DR. APOSTOLAKIS:
Okay.
So the Commission says 16 that, in applying the objective for individual risk of 17 prompt fatality, the Commission has defined the vicinity as 18 the area within one mile of the plant site boundary.
Okay.
19 If there are no individuals residing within a 20 mile, an individual should, for evaluation purposes, be 21 assumed to reside one mile from the site boundary.
22 Now, doe that imply that at some point that 23 individual starts running?
24 MR. KING:
Let me ask --
25 DR. APOSTOLAKIS:
I don't understand.
How long
()
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
263 1
does it.take for the thing to be --
()
2 DR. POWERS:
I-wonder if, in order -- since this 3'
discussion is not coming to any resolution, I wonder if we 4
.oughtn't move on, if we are going to get through this 5
presentation.
6 CHAIRMAN SEALE:
Yeah.
7 DR. APOSTOLAKIS:
But it does affect what we 8
write.
9 DR. POWERS:
Well, I have a feeling that if the 10 difficult is such as it is, we are not going to be able to 11 produce anything.
12 MR. KING:
We are not asking for a position on any 13 of these eight issues at this point.
14 DR. POWERS:
Well, unfortunately, other people 15 are, so.
16 DR. APOSTOLAKIS:
I am bothered by the fact that 17 in past meetings we sort of all agreed that the Commission 18 statement does not address societal risk, and it seems to 19 me, if you read it, at least from one perspective, it does.
20
.And you should not confuse the statement and its intent with 21 the calculation.
22 DR. POWERS:
Well,.I meani you have said that now.
23 Maybe we ought to move on, because --
24 DR. APOSTOLAKIS:
Well, I know we have to move on.
25 But we are going to skip a lot of this stuff.
ANN RILEY & ASSOCIATES, LTD.
\\/
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
264 1
Okay, Tom.
(
2 MR. KING:
The last two issues were -- right now 3
the policy statement is silent on temporary plant 4
configurations.
5 DR. APOSTOLAKIS:
So the first bullet there, the 6
second bullet, I sort of disagree.
Okay.
Neither present 7
QHOs provides a quantitative measure of societal risk.
I am 8
not sure that is correct.
That's the result of all this 9
discussion.
10 MR. KING:
It depends how you interpret the policy 11 statement.
Clearly, the way the QHOs are calculated, it is 12 an individual risk basis.
13 DR. APOSTOLAKIS:
Okay.
14 MR. KING:
They clearly don't address land
()
15 contamination or any other environment effectc They do not 16 address temporary plant configurations.
We have a strategic 17 plan out now that has a numerical goal that could be 18 interpreted as dealing with a temporary condition, should we 19 do something in the policy statement to address that 20 condition.
And then the policy statement talks about 21 defense in-depth.
It talks about use of PRA.
But I think 22 it could be clarified in terms of the relationship of the 23 QHOs to defense in-depth, ?,o adequate protection.
24 We have an SRM from the Commission that tells us 25 how to use the goals in terms of defining how safe is safe O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters p
1250 I Street, N.W., Suite.300 Washington, D.C.
20005 (202) 842-0034
l 265 1
enough, but the policy statement doesn't -- doesn't describe
()
2 that.
So there are some clarifying words for consistency 3
and coherence that we thin could improve the statement.
4 That's the intent of the last bullet.
5 Anyway, those are the eight areas that we have 6
been looking at.
And the question is, what are we going to 7
do?
At this point, we haven't been able to reach a 8
conclusion as to which of these we feel we should go forward i
9 with and propose a change and which of these we should not.
10 Basically, the bottom line is we believe we need some i
11 additional time to look at the benefits and the impacts, 12 that this is not something we should rush into.
There is no 13
-- no urgency to rush into it.
There is no safety issues, 14 you know, sitting there unrescived, waiting for this to
)
15 happen.
16 There's a lot of things to think about.
What I 17 have tried to do on the next few viewgraphs is to talk about 3
18
.the things that we feel we need to do before we actually 19 come back to the Commission with a recommendation.
20 DR. IPOSTOLAKIS:
Let me try to bring down the 21-argumer.t to its bare essentials.
Is it true that we should 22 not have de facto regulation, things that sort of evolve 23 from the way we are doing business, and acquire a policy 24 status without ever really going through the process that 25 establishes proper policy?
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C.
20005 (202) 842-0034
266
~1 And one of those is a 10 to the minus.4 per 2.
reactor year, core damage frequency, which is sort of
~
3 accepted now as a goal by everyone, and it does establish 4
policy in the sense that'it is not consistent.
It is more 5
conservative than the QHOs.
If you work backwards, taking, 6
you know, rough estimates of the contributions of level 2 7
and so on.
So at an intellectual level, this is the primary 8
motivation.
You realize that you are doing something that 9
has not really been scrutinized properly to acquire the 10 status of a policy.
11 I can see no other reason as to why we really want i
12 to elevate it.
It hasn't been scrutinized, yet it is being
'13-used.
14 MR. KING:
That is clearly one advantage of trying 15 to elevate it.
I think, yon. low, you are right, we are 16 using the 10 to the minus 4th in a number of areas.
If we 17 get that into the safety goal policy, either as a-18 fundamental goal or a subsidiary objective, I think it would 19 provide some more firm foundation for --
20 DR. APOSTOLAKIS:
Yeah.
21 MR. KING:
-- for the basis that we are using for 22 a lot of things we are doing right now.
23 DR. APOSTOLAKIS:
Exactly.
In other words, it 24 will be scrutinized.
25 MR. KING:
It vill be scrutinized.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
267 1
DR. APOSTOLAKIS:
And we will understand its
()
2 relation to other regulations.
I think you guys have also 3
listed those.
4 MR. KING:
Yeah.
5 DR. APOSTOLAKIS:
Because when you do that, then 6
you really pay attention.
And the major drawback is a 7
practical one, that there is only so much that you can do 8
with limited resources, and you told us last time and I 9
think you will tell us today again, well, it will require 10 staff resources to make and implement revision, that the 11 risk-informed, performance-based initiative that we have 12 been working on for the last year and a half, may suffer, or 13 may be affected, impacted on adversely.
That's the real 14 practical concern that makes us hesitate to do this.
Is 15 that correct?
Are these the two basic reasons and 16 everything else is really gravy?
17 MR. KING:
Those are two basic reasons.
I think 18 there is another basic reason, and that is, if we do elevate i
19 this and put it into safety goal policy, are there other I
20 impacts that we haven't thought of, in terms of legal 21 implications, opportunity for misunderstanding what that 22 means in terms of looking at a plant or a group of plants i
23 and their safety?
Is there some additional guidance we need j
24 to put out in terms of how do you calculate?
You know, what 25 are the ground rules with the methodology?
You know, what ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
268 1
are we getting ourselves into if we do elevate that to that
(
2 level?
3 DR. APOSTOLAKIS:
Okay.
I think one of the I
l 4
arguments you just gave is valid, namely, the problem of 5
potential misperceptions and saying, you know, gee, this 6
plant now is unsafe, when in fact it meets all the adequate 7
protection criteria.
8 But there are ways around it.
They may be time 9
consuming, but there are ways around it.
10 The other one, about the legal implications and 11' all that, it seems to me that is really the benefit of 12 scrutinizing something before you start using it as a goal, 13 that you will understand the implications and its i
14 relationship to the rest of the regulatory system.
So that
)
15 is -- well, it is not so clear cut that it is a benefit, but 16 there may be some things that will come up that we will not 17 like.
18 So these are really the basic arguments.
Right?
19 MR. KING:
Yes.
20 DR. APOSTOLAKIS:
We agree that these are the 21 basic arguments.
And any decision as to whether we should 22 elevate the CDF to the same level as the QHOs has to take 23 these three or four arguments into account.
-24 MR. KING:
And weight the benefits versus the j
25 drawbacks.
1 O.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
269 1
DR. APOSTOLAKIS:
Versus'the drawbacks.
()
2 MR. KING:
Yeah.
3 DR. APOSTOLAKIS:
Okay.
4 MR. KING:
And,.you know, and recognizing the fact 5
that today we are proceeding to use CDF, LERF, in the Reg.
6 Guides, in regulatory analysis, in advance reactor reviews.
7 We are not sitting here waiting for this to happen.
8 DR. APOSTOLAKIS:
So perhaps you can go to slide 9 9
then, we covered the others, if you agree.
Since I wasted 10 your time, I am trying to help.
11 MR. KING:
No, I'll be glad to go to slide 9.
12 DR. APOSTOLAKIS:
I think the essence of the other 13 arguments has already been discussed, unless you fellows 14 feel that I did not do it justice.
)
15 MR. KING:
No, slide 7 is just an attempt to say 16 some of'these things are connected and we need to look at 17 them as a package.
We did -- slide 8, ultimately, we are ic going to have to make a decision on --
19 DR. APOSTOLAKIS:
Yeah.
20 MR. KING:
On looking at these things, and we 21 tried to write down, how are we going to make decision?
22 DR. APOSTOLAKIS:
Exactly.
23 MR. KING:
One of the things we want to look at, 24 so that's a set of criteria we would propose to use.
And 25 then when we go to slide 9, what are going to do a this O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
270 1
point?
Well, what we are going to do at this point is-we
()
2 are going to provide a paper to the Commission at the end of 3
this month that sort of lays out these issues, and lays out 4
the considerations that we don't see any urgency at this 5
time for revision.
6 DR APOSTOLAKIS:
Now, fom, in terms of this 7
agency, when we decide that something is not urgent, does 8
that mean we never do it?
9 MR. KING:
No.
10 DR. APOSTOLAKIS:
So would you give the Commission 11 then a timetable and say it is not urgent, but we will start 12 in the year 2001?
We may have a different Commission then.
13 But, still, are you under any obligation to tell them when?
i 14 If there is no urgency, fine.
How is that different from we 15 don't need it?
16 MR. KING:
I think we are obligated to ultimately 17 come back and recommend an action plan to the Commission.
18 DR. APOSTOLAKIS:
Oh, you have to.
Okay.
j i
19 MR. KING:
And I think at this point, today, I am 20 not prepared to tell them on what schedule we will do that.
21 It is a question of resources -- on all the other things 22 that are on our plate versus what do we need to adequately 4
23 do this job.
I would hope by the time the paper goes up we 24 would be able to say something about that, that we need 25 another year to study it, or we need six months and we'll
[~
ANN RILEY & ASSOCIATES, LTD.
A Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
271 1
'get back'to you.
Ii D) 2 I think, clearly, if we get into revising the 3
-policy, it is a two year process, because you have to go 4
through, prepare a package for public comments.
It's like 5
doing a rulemaking.
Get the public comments, we come to I
6 ACRS at the draft stage, the final stage, CRGR, and so 0
7 forth.
So it is a two year process.
8 DR. APOSTOLAKIS:
So there is a difference in 9
saying what is urgent, and this is important, but it's not 10 urgent, and this is unimportant --
11 MR. KING:
Yes.
12 DR. APOSTOLAKIS:
-- because you have to come up 13 with a plan for the important thing which is not urgent.
14 That's what I want you to understand.
(
15 MR. KING:
Yes.
We're not saying --
16 DR. APOSTOLAKIS:
We're not really tabling this 17 indefinitely.
18 MR. KING:
No.
L 19 DR. APOSTOLAKIS:
Okay.
l 20 MR. KING:
The intent is not to table it 1
21 indefinitely.
l 22 DR. APOSTOLAKIS:
Okay.
l 23 MR. KING:
The intent is to give it the right i
24 priority and figure out what the schedule is and come back 25 with a recommendation.
t O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034 i
~
'1 1
DR. BARTON:
George, if it takes six months to do
()
2
-- six years to go do that, why do it at all?
I don't know 3
what Tom's schedule is, but suppose it is based on 4
priorities other, the action plan goes out to the year 2002, 5
Why do it at all?
6 DR. APOSTOLAKIS:
Presumably the benefits of doing j
i 7
it will still be there.
1 8
DR. BARTON:
To me, it's if it ain't broke, don't 9
fix it.
10 CHAIRMAN SEALE:
I think that it is important, but i
11 it's important that you do it right, and they are not ready
)
i 12 to do it right yet.
13 DR. BARTON:
Well, that's what I'm saying, too.
14 CRAIRMAN SEALE:
So we need more time to be ready.
)
15 to do it right.
16 DR. APOSTOLAKIS:
But my point is when we say we 17 need more time, do we put the period there or do we say 18 period, therefore, we will do this at such and such a date, 19 because there is a big difference in my mind between those 20 two, and Tom said yes, we will give some --
21 DR. BARTON:
I think he clarified it with the fact 22 that his action plan is going to have some milestones in it, 23 as opposed to you look at a generic safety issue that may be 24 ranked high and it's 14 years old because it doesn't have an 25 act' ion plan.
1 ANN RILEY & ASSOCIATES, LTD.
\\
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1
)
273 l'
DR. MILLER:
I'm certain the Commission will not 2
let them get away without an action plan.
3 MR. KING:
I'm sure the Commission is looking for 4
a plan and a schedule today.
I'm not prepared to give you 5
that but, you know, we are going to have to come up with 6
that.
But I am prepared to say we're not proposing this be 7
put on the shelf indefinitely.
8 DR. APOSTOLAKIS:
Let me ask another question.
At 9
this point, my understanding is that the regulatory guides 10 that we are just completing and related activities consume a
{
11 significant amount of time of the staff.
12 MR. KING:
Yes.
13 DR. APOSTOLAKIS:
Correct?
14 MR. KING:
Yes.
15 DR. APOSTOLAKIS:
How long is that going to go on, 16 do you know?
Can you guess?
At the same level.
17 MR. KING:
It won't be at the same level.
I think 18 we've got some ideas for some follow-on activity in 19 risk-informed.
Some of it could result in additional reg 20 guides, some of it could just result in some additional 21 studies looking at, for example, the NRC inspection program 22 in a risk-informed fashion.
23 DR. APOSTOLAKIS:
Would you say --
24 MR. KING:
I would say it's still going to tie up 25 in Researcn probably three or four people full time.
ANN RILEY & ASSOCIATES, LTD.
\\
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
274
'l DR. APOSTOLAKIS:
For how long?
(
2' Mk. KING:
Probably at least the next couple of 3
years.
4 DR. APOSTOLAKIS:
And then things will go back to 5
sort of normal?
6 MR. KING:
Let me --
7 DR. BARTON:
Does the staff then get involved in 8
reviewing applications if this was such a good deal?
I 9
mean, you go from developing the --
10 MR. KING:
Another thing we have talked about is 11 reviewing applications, particularly ones that are pushing 12 the guidelines, and having some teams do that would be 13 Research, NRR teams.
How many of those will come in, I 14 don't know, but we are committed to work together to deal 15 with those first of a kind or those tougher ones.
16 MR. NEWBERRY:
Yes.
I think that's the resource 17 question, and right now, it is a question mark, what 18 response will we get to the reg guides?
How many license
]
19 amendments will we get from the utilities?
20 CHAIRMAN SEALE:
I think that was an issue we 21 raised at our last meeting, as a matter of fact.
22 DR. MILLER:
Right.
If you've done all this and 23 you get no applications, it's all for naught.
You hope you 24 get some applications.
Vny even go through this exercise if 25 you don't get use of it?
If you get use of it, you're going 1
I O
ANN RILEY & ASSOCIATES, LTD.
I Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
275 1
to have more staff resources.
(
2 MR. KING:
And we have the activity that you heard 3
Bob Christy talk about at your subcommittee meeting.
That 4
could-lead to a number of areas that are candidates for a 5
risk-informed approach.
We're committed to work on the 6-pilots with them.
That will take some staff resources, but 7
what comes out of that may take more.
8 MR. NEWBERRY:
And there's going to be a broad
-j 9
spectrum, based on what I have seen, on resources necessary 10 to look at some of these pilots.
Some are rather simple and 11 some where you could end up proposing to change a regulation i
12 to look at a design' basis event like large break LOCA 13 differently, that's a very resource --
14 MR. KING:
Yes.
()
15
.DR. MILLER:
I can't see being back to whatever 16 normal is d.or quite a long time.
17 MR. KING:
We've got the PRA standards activity 18 we're working with ASME on; we've got the certification
-19 programs the owners groups are working on that we're going 20 to get involved in.
21 DR. APOSTOLAKIS:
Is there an alternative to 22 revising the safety goal policy statement and do something, 23 send a message that CDF is treated as a fundamental goal but i
24 not go through --
25 DR. BARTON:
But why?
Why?
I don't see a ANN RILEY & ASSOCIATES, LTD.
i
\\
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 276 1
compelling reason.
Why the urgency, I guess?
What's the I
()
2 compelling reason to go do that?
3 DR. APOSTOLAKIS:
Well, I don't -- there are two 4
things that are happening now that have not really been 5
. scrutinized, the ten to the minus four and the ten to the 6
minus three.
I think it's widely used, the ten to the minus 7
three or thereabouts is widely used as a real criteria in 8
the sense that the industry takes action or.the NRC staff 9
takes action if a contributor without frequency is 10 identified.
11 DR. MILLER:
That's also flexible, a matter of 12 judgment.
13 DR. APOSTOLAKIS:
Of course.
14 DR. MILLER:
Other factors go into that judgment.
)
15 DR. APOSTOLAKIS:
No, I think if you hit ten to 16 the minus three, it's hard to find other factors.
You are 17 really now talking about serious stuff.
18 DR. MILLER:
If you hit five times ten to the 1
19 minus three, there may be other factors you take into --
20 DR. APOSTOLAKIS:
Well, you see, that's exactly 21 why I want this debate, because these things -- you know, 22 you have many people interpreting them in different ways, 23 and these things are happening right now without the proper 24 scrutiny,-without the proper debate.
So that is what 25 bothers me.
l ANN RILEY & ASSOCIATES, LTD.
Court Reporters s-1250 I Street, N.W., Suite 300 Washington, D.C.
20005 (202) 842-0034
i 1
277 1
1 DR. MILLER:
Has it caused a problem?
(~h i
i 2
DR. APOSTOLAKIS:
I don't know.
\\_/
3 DR. MILLER:
I would say we wait a couple of years 4
and have the implementation --
5 DR. APOSTOLAKIS:
Oh, a couple of years is fine.
6 I'm not saying we should do it tomorrow.
7 DR. MILLER:
No, have the implementation flowing 8
and see what happene and see -- we may nct want to do it at 9
all.
10 DR. APOSTOLAKIS:
Tnat's not the question you 11 asked me.
You didn't ask me whether I wanted to wait for i
12 two years.
You asked me why do I have to worry about these 13 things, and I answered that question.
14 DR. MILLER:
Okay.
(_,/
15 DR. APOSTOLAKIS:
If your question was, are you 16 willing to wait for two years, my answer is yes.
17 DR. MILLER:
I'm saying in two or three years, you 18 may have --
19 DR. APOSTOLAKIS:
I don't think there is urgency 20 to it, btu there is a reason why you should be concerned.
21 That doesn't mean that, you know, I have to do it tomorrow.
22 That's what I'm saying.
23 DR. MILLER:
You may have less or more reason to 24 be concerned in three years.
25 DR. APOSTOLAKIS:
Well, I don't want to speculate.
[#l ANN RILEY & ASSOCIATES, LTD.
\\'
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
278 1
But the idea of doing things because everybody does it and
()
2 it sounds okay_ bothers me, because we are scrutinizing other 3
things to death, so I think the fundamental policy statement 1
4 of the Commission should go through that scrutiny, in my 5
opinion.
But I fully agree that it is not an' urgent 6
problem.
That's why I'm asking the question, when we say 7
it's.not urgent, what does that mean really?
Are we tabling 8
it or there will be a plan?
And Tom said that there will be 9
a plan.
Okay.
Fine.
So let's go on.
Do you want to 10 continue and talk about this or --
11 MR. KING:
Well, let me mention, in terms of a 12 letter, I said up front we would like a letter on this 13 subject.
14 DR. APOSTOLAKIS:
We already have a draft.
15 MR. KING:
Okay.
I think the points that are 16
-important to us are the urgency issue, do you agree that 17 it's not an urgent matter; that we need to carefully look at 18 the benefits versus the drawbacks and that there are a 19 number of things to be considered in doing that.
Of the 20 eight issues, are there any you feel are not really an issue 21 or did we miss something.
Maybe there ought to be nine or 22 ten issues.
I think those are the things that are of 23 immediate importance to us in terms of you writing a letter.
24-DR. APOSTOLAKIS:
Yes.
25' MR. KING:
Now, maybe you want to say more than
]
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
279 1-that.
f~)
2 DR.'APOSTOLAKIS:
I don't think you missed V
3 anything, but in my mind, the real driver here is the issue 4
of consistency, and you shouldn't be more conservative than 5
the overall safety goal statement just because you feel like 6
it or it's not reasonable to be -- I mean, either you are 7
consistent with the goals or you're not, and if you're not, 8
then'you're establishing your policy.
If you're 9
establishing your policy, then you should pay the price.
In 10 other words, you have to go through the process.
That's the 11 main argument in my mind for doing it.
12 Again, when you're going to do it is a very 13 difficult question, and I agree that it doesn't have to be
-14 done immediately.
()
15 By the way, the ACRS warned the Commission back in 16 1987 not to do that, to make sure that when we establish 17 goals, they will not be so conservative as to be de facto 18 new policy.
I can give you une quote 19 MR. KING:
Yes.
And that's why we couldn't find a i
20
-way to make the suggested plant performance guideline ten to 21 the minus six work, because it in effect was a de facto new 22 goal.
23 DR. APOSTOLAKIS:
I think we're in agreement here.
24 The thing that really bothers me, though, if you want to 25 really have input, rather than going through the -- is this
[s~ )
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
i 280 1
issue of land contamination.
You really have an opportunity
()
2 to convince me one way or another.
What is it about it?
3 Why do I have to worry about it all of a sudden?
Because 4
it's important?
5 DR. KRESS:
If you looked at land contamination as i
6 a dose per unit area to the land, that is a societal. risk.
7 You can convert that with the population to the total number 8
of deaths.
I think you can make those two things be 9
' consistent with each other if you -- and I don't think-you 1'O
'need both I think one or the other will --
)
i 11 DR. APOSTOLAKIS:
One or the other?
12 DR. KRESS:
Total number of deaths or total land 13 contamination.
I think you could make those consistent.
14 DR. POWERS:
I think I view land contamination in h
15 a different way.
I hear a lot of people tell me, oh, gee, j
16 you guys are spending a lot of time worrying about nuclear 17 power plant accidents, but most industrial facilities, and i
18 they could list several, are much more hazardous and you're 19 over-regulating this industry.
When you think about that, 20 you see, well, it's not quite fair.
If I have an oil 21 refinery blow up on me, it can, indeed, hurt members of the 22 public, but after it's done, they clean it up and the public 23 moves right up next to the site boundary.
~
24 When, instead, I look at the Chernobyl accident,
^5.
and I see a town of 50,000 population that cannot be ANN RILEY & ASSOCIATES, LTD, Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
281 1
occupied for another 98 years, I see this accident and a L
I
()
2 refinery' accident are qualitatively different, and the 1
3 difference lies in the land contamination, and that is a 1
4 risk measure that one ought to bear in mind.
I 5
When I look at the world-wide safety goals that 1
6 countries have set up, I find some countries who tacitly 7
assume no member of the public will be injured by a reactor j
8 accident and focus exclusively on the land contamination, 9'
and I say gee, it's missing frum our regulations, and yet 10 here it's the focus of those countries' entire safety goal 11 structure, and why is that?
And so I raise the question of, 12 is not land contamination a peculiarity of the nuclear 13 enterprise and should' figure in our safety considerations?
14 MR. KING:
The way I look at it is we do look at
)
15 land contamination when we do environmental impact 16 statements today.
We do look at land contamination when we 17 do a regulatory analysis.
It's a cost that's factored into 18 the equation.
19 DR. POWERS:
I'll have to interrupt you and say 20 that I find these arguments very persuasive, saying yes, the 21 land contamination is a peculiarity and a concern here, but 22 because of the peculiarity of the society and whatnot, it 23-need not be in this first tier, that it can be viewed as a 24 partitioned or second-tier type requirement, and we put it
-25 into cost-benefit analysis and environmental impact
(
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
282
-1 statements that there's a whole other block of regulations
()
'2 where tha' can~be taken into account.
3 I like that argument that you advance.
4 MR. KING:
It's not ignored, 5
DR. KRESS:
And that's where I think it belongs, 6
because in a safety goal statement, if you were, say, to 7
have one on early deaths, late deaths, societal total deaths 8
and land contamination, you've way over-specified.
One of 9
those will rule and you don't need the others because the
-10 others -- one of them will rule.
11 DR. POWERS:
I will agree with you that one of 12 them will rule.
I disagree with you strongly on which one 13 rules, but I will agree with you that one of them rules.
14 But I think the --
15' DR. KRESS:
I didn't say which one I thought 16 ruled.
17 DR. POWERS:
But I bet you have in the past.
18 The difficulty I see and the reason I'm happy to 19
-- that I find persuasive the arguments that have been made 20 on this is that you want a hierarchy of structure.
I want 21-to be able to follow from the topmost level of policy down 22 through its implementation in a cascading fashion, and I'm
-23 running into a problem when I bring in land contamination if 24 I go from the Atomic Energy Act downwards.
25 I have a much easier time going from the law that
]
-1
.[\\ I ANN RILEY & ASSOCIATES, LTD.
d Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 283. escapes me right now that gives rise to environmental impact.
(
2 statements to get to land contamination as an issue -- NEPA y
3 rules.
I have a much easier way to get there, and so I'm 4
uncomfortable with the Commission saying our compelling 5
authority is the Atomic Energy Act and vur top level of 6
requirements will come from that, but there is another j
7 implicit top level requirement of all federal agencies, and 8
that is they obey all federal laws.
9 So we factor in when we come in our second tier, 10 we have come from that implicit requirement to get to this 11 second tier requirement, and it seems like it hangs together 12 very nicely.
I mean, I thought your arguments there were 13 compelling.
14 DR. FONTANA:
Tom, could you clarify for me, in
()
15 making decisions, total societal cost is not only health 16 costs, but it's also the cost, dollar cost of land put out 17 of circulation and that sort of thing.
18 MR. KING:
Yes.
19 DR. FONTANA:
Whose responsibility is it to look 20 at that?
21 MR. KING:
NRC looks at it when they do regulatory 22 analysis and they're looking at --
23 DR. POWERS:
Well, they also look at it when they 24 do the environmental impact statement.
25 MR. KING:
Yes.
In the environmental impact O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
284 1
statement, they look at, you know, potential for
()
2 contaminating land,'for contaminating it so bad that you 3
can't clean it up and-you permanently have to, you know, 4
quarantine the land, and there's.some probability associated 5
with that and some judgments are made that it's, you know, 6
not a major impact because the probability is so low.
7 DR. FONTANA:
And a cost-benefit assessment is 8
made at that point?
9 MR. KING:
In regulatory analysis, there's dollars 10 associated with permanently condemned land versus the cost 11 of cleaning up other segments of land, and that's factored 12 into the cost-benefit equation.
In the environmental impact 13 statement, I don't think there's dollars associated with it; 14 as I recall, it's just a probability of how much land
()
15 contamination you're going to get.
16 DR. POWERS:
Well, it's a weighting.
17 DR. FONTANA:
Because it's potentially a real 18 cost.
19 MR. KING:
It's a weighting against are there less L
l
'20 costly alternatives to the proposed action.
Under the SAMDA 21 piece of the environmental look, yes, then there is a 22 weighting.
23 DR. KRESS:
Excuse me.
I was maintaining that 24 your LERF in essence represents a good surrogate for land 25
' contamination, although it may -- you know, I don't know i
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
285 1
what the value of land contamination that represents.
It's
{
'2 a risk'of land contamination, and it's -- so you've already 3
gotten a measure in there.
4 DR. POWERS:
I guess I don't understand that, Tom.
5 If I -- and it may be a misunderstanding on my part on LERF 6
because I find that a very confusing concept right now.
If 7
I go through an accident management effort and say I will 8
try to manage the source term and I find a strategy by which 9
I divide that source term that comes from an early 10 containment failure by a factor of two, I find no impact 11 when I go through like the 1.174 analysis.
12 DR. KRESS:
Oh, yes.
That will show up in your 13 LERF.
It should, if you view LERF as a frequency 14 probability of an early release of a substantial quality of
()
15 fission products.
I 16 DR. POWERS:
But it's like infinities.
A 17 substantial divided by two is still substantial, and I 18 haven't changed the probability at all.
So I get no credit 19 for -- I find no credit for myself to do accident management 20 to mitigate source term.
If I can affect probabilities,
)
21 that's good, but affecting the magnitude of the source term 22 would have no impact.
23 DR. KRESS:
Well, it certainly is linear on land 24 contamination.
A factor of two will give you a factor of 25 two on land contamination.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Streec, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
I 286 a
1 DR. POWERS:
But I don't get anything through LERF j
l l
2 that way.
(
3 DR. KRESS:
Yes.
LERF gives you that precisely, I 4
think.
5 DR. POWERS:
No.
LERF is a probability that I'm 6
going to have a large release.
7 DR. KRESS:
I know.
So if you change the large 8
release by a factor of two by mitigating the fission j
9 products, then you've changed the probability of land 10 contamination at a certain level.
11 DR. POWERS:
I did in reality, but I don't get it 12 through LERF.
I change the probability not one whit.
l 13 DR. APOSTOLAKIS:
We have to go on now 14 DR. KRESS:
But it ought to be implicit in the way 15 you calculate LERF.
16 DR. POWERS:
But it's not.
17 DR. KRESS:
Well, I think it is.
18 THE COURT:
Okay.
I think it's very clear that 19 these issues need to be resolved.
20 DR. APOSTOLAKIS:
I just remembered another factor 21 that -- element that will make the societal risk different i
22 from just being the product of the number of people times 23 the individual risk, and this is the -- and that was also, 24 by the way, in the report the ACRS produced in 1980.
It's 25 the aversion towards large accidents, that an accident that l
f(
)
ANN RILEY & ASSOCIATES, LTD.
j Court Reporters 1250 I Street, N.W.,
Suite 300 Wasnington, D.C.
20005 (202) 842-0034 i
s
~
287 1
kills 500 people and an accident that kills a thousand 1
()
2 people, the one that kills a thousand people is ncb just 3'
twice less desirable.
4 In other words, if you have a major catastrophe, 5
you have to pay a penalty for it, and they usually put the 6
risk aversion factor, so that kind of thing would clearly 7
make societal risk different from the individual risk.
In 8
fact, the ACRS on its tables had the number of people to the 9
1.2 power, okay, to calculate the expected societal impact, 10 which is clearly a non-linear function.
So that was an 11 essential element that was missing from our earlier 12 discussion, and I knew something was bothering me and it 13 finally came.
So for the record, I want to say that.
14 So it seems to me that we have general agreement 15 here that land contamination is handled appropriately right 16 now.
We shouldn't really recommend anything like putting it 17 in the safety goal policy statement?
18 MR. KING:
In the discussions the staff has had in 19 trying to put together a paper, we did not have any 20 dissenting views on land contamination.
We felt that the 21 way it was treated now was sufficient, that it is covered.
22 DR. APOSTOLAKIS:
My only question is, the results 23 of these regulatory analysis, are they in general more 24 conservative than one would be getting from core damage 25 frequency and LERF considerations?
()
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
288 1
MR. KING:
I'm not sure what you mean by
(
2 conservative.
3 DR. APOSTOLAKIS:
Well, I don't know the details 4
of what you do --
5 MR. KING:
They use PRA information, whatever the 6
best available is.
If it's --
1 7
DR. APOSTOLAKIS:
Well, let me put it this way at h
8 the higher level.
Somebody does a PRA, they meet the CDF i
1 9
code, they meet the LERF code, they meet the QHOs, then you 10 do a regulatory analysis involving land contamination and 11 come back and say, no, no, no, you have to do this to your 12 plant.
That, in my mind, is more conservative.
13 MR. KING:
In doing regulatory analysis, there's a 14 cost-benefit piece of the equation which does not come in O
.t f
15 when you're doing a regular PRA.
v 16 DR. APOSTOLAKIS:
Right.
17 DR. KRESS:
You have to pass two tests before you 18 go implement a new rule or a back-fit, and that is, is there 19 substantial improvement in safety which strictly looks at 20 things like delta CDF, delta LERF.
If it passes that test, 21 then it also has to be cost beneficial.
That's where the i
22 land contamination, you know, and other economic factors 23 come into play, 24 DR. APOSTOLAKIS:
But you could come up with some 25 requirements that would not have.come up by looking only at ANN RILEY & ASSOCIATES, LTD.
A Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
289 l'
core damage frequency and LERF.
()
2 MR. KING:
You mean is there something that 3
DR. APOSTOLAKIS:
Is that true?
4 MR. KING:
-- economically makes sense-that's cost 5
beneficial.but doesn't improve safety very much?
6 DR. APOSTOLAKIS:
But you are bringing -- yes --
7 land contamination into the picture.
8 MR. KING:
In the world of the environmental 9
impact statement, that test is applied to things called 10 SAMDAs, severe accident mitigation design alternatives.
A 11 new plant is built; they have to look cost benefit only, 12 would it make sense to add another diesel containment vent, 13 whatever it may be.
Forget the fact that it may not buy you 14 much in terms of absolute safety; if it's cost beneficial,
()
15 figuring in land contamination and these other economic 16 factors, the environmental rules right now suggest you go-17 ahead and make that change to the plant.
18 DR. FONTANA:
Does it also address the bigger 19 question, do you build a nuclear plant as compared to a 20 gas-powered plant?
21 MR. KING:
The environmental impact statements 22 have to address that as well.
23 DR. FONTANA:
Okay.
24 DR. KRESS:
Tom, your cost-benefit regulatory 25 analysis uses $2,000 per man / rem.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters j
1250 I Street, N.W.,
Suite 300 I
Washington, D.C. 20005 i
(202) 842-0034
1 290 1
MR. KING:
Yes.
'v) 2 DR KRESS:
It's my impression that that value is t
3 used as is and all you calculated is man / rems, and that that 4
implicitly includes the cost of land contamination.
5 MR. KING:
No, i
6 DR. KRESS:
Is that not how that's handled.
1 7
MR. KING:
No.
When the thousand dollars per 8
man / rem was established years ago, that was the thought.
It 9
covered everything.
But now it has been separated out.
The 10 new $2,000 number is health effects only.
11 DR. KRESS:
It's just health effects only.
12 MR. KING:
Yes.
13 DR. KRESS:
Okay.
I was remembering that wrong.
14 DR. APOSTOLAKIS:
I think we're running out of
()
15 time, so I really want to cover the major points that I 16 think are relevant here.
17 On your slide 20, you talk about defense in depth 18 and adequate protection.
Do you think -- I mean, you 19 mentioned earlier that one of the potential problems of 20 elevating CDF to a fundamental level is that we may create 21 perceptions that a number of plants are unsafe, and, in 22 fact, I was looking at the IPE report that your office 23 issued recently, and I think there are about 15 units, nine 24 licensees, 15 PWR units that are above the ten to the minus 25 four number.
You don't remember that, but --
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 5
291 1-MR. KING:
There are some that are --
()
~ 2 DR. APOSTOLAKIS:
Well, I can read you the 3
sentence, if you want.
4-MR. KING:
Okay.
5 DR. APOSTOLAKIS:
Okay?
That's page 7-3.
6 MR. KING:
Right.
That's internal events only.
7 DR. APOSTOLAKIS:
Internal events only.
So the 8
actual numbers may, in fact, be higher given the limitations 9
of IPEs and so on.
So there may be a perception that these 10 are unsafe.
11' You are raising the question of adequate 12 protection here, which is a valid point.
The question is 13 really this:
If we decide to elevate the CDF to the 14 fundamental level, should we also say something about
()
15 adequate protection?
That is not a trivial task.
16 In other words, in a simple case, the goal is ten 17 to the minus four, adequate protection is ten to the minus 18 three.
No, I know you can't say yes, but something to that 19 effect, which incidentally, again, was in the 1980 report 20 from the ACRS to the Commission.
Obviously the Commission 21 didn't buy it.
But these are not new issues.
22 So I don't know how you can get around this 23 potential perception problem without saying something 24 quantitatively or give a range for adegrate protection.
In 25 fact, in one of our reports maybe a year or so ago, we i
h ANN RILEY & ASSOCIATES, LTD.
%~
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
292
{
11 stated very-clearly that the adequate protection -- the CDF
()
2 in response to the. adequate protection criteria is higher 3
.than the goal.
I mean, that's something that everybody 4
accepts, right?
So I don't know how you would handle that.
5 It seems to me you have to say something, and that 6
something:cannot be adequate protection is higher.
So are 7
we going to be looking for another number or a region now or 8
a rangei Because the fact is, I think ten to the minus 9
three is the adequate protection.
{
l 10 MR. KING:
I can give you my personal view.
We 11
-have not concluded --
12 DR. APOSTOLAKIS:
Sure, I realize that.
13 MR. KING:
I think we need -- it would make sense 14 to me to have the safety goal policy discussed, the "how
()
15 safe is safe enough" concept, and the adequate protection 16 concept.
17 DR. APOSTOLAKIS:
Yes.
I agree.
18 MR. KING:
They have already defined numerically 19 the "how safe is safe enough" with the QHOs.
I personally 20 don't think you need to quantify adequate pIntectic 21 think you need to get across the point that it is a cm cr. t 22 that says this is the minimum you need in terms of 23 protection, you're out of cost-benefit space when you're 24 down in that level.
It's achieved by complying with the 25 regulations in-terms of interpretation today, but I think to 1
ANN RILEY & ASSOCI.ATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
1 293 1-get in to quantify it, one isn't necessary and two could
, ()
2 lead to a lot of problems.
But I think it ought to be 3
discussed, the relationship of all those things ought to be 4
laid out.
5 DR. APOSTOLAKIS:
That is quantification or 6
quantification.
I think what you are saying is it is not 7
necessary to say it's 10 to the minus 3, and I agree, 8
because, you know, if you are at 9, 10 to the minus 4, you
)
l 9
know, that doesn't mean you are safe.
I mean you have this 10 imprecision.
But some numerical guidance, I think,.would be 11 needed, and very carefully phrased.
Because the factor --
12 and, you see, that's what bothers me, that we have all these i
13 factor things, and I have, you know, in a different life I 14 was on the other side, doing PRAs and so on.
I have seen
()
15 it.
If the results are close to 10 to the minus 3, they are 16 not published, and the utilities are looking for ways to 17 change the plan, change the procedures, whatever, do a 18 better analysis, so that they will not'have to publish 10 to 19 the minus 3.
What does that tell me?
It tells me that the 20 industry has a number there in that range that they consider 21 unacceptable, and I think the Staff thinks that way, too.
22 As we approach the 10 to the minus 3, we are really now
)
23 pushing now the envelope of adequate protection.
Okay?
Now 24 are we happy leaving it at that, so there is a common 25 understanding?
Or do we want to say something more about
/I ANN RILEY & ASSOCIATES, LTD.
l
\\w #
Court Reporters 1250 I Street, N.N.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
294 1
'it?
O 2
I' don't have the answer, either.
\\j 3
MR. KING:
I'm not sure the benefits of 4
quantifying it that way, the drawbacks.
5 DR. APOSTOLAKIS:
That's right.
6 MR. KING:
That's my personal opinion.
7 DR. APOSTOLAKIS:
Okay, that's a legitimate point 8
of view.
You have a minute.
Do you have anything that you 9
think is of extreme importance?
10
[ Laughter.]
11 DR. APOSTOLAKIS:
Because, you know, everything we 12 discussed with the subcommittee we discuss in here.
I am 13 trying to go down to the heart of the argument here, and so 14 far I have not sensed any disagreement between the Staff and 15 members of the committee.
That doesn't mean that there may.
j 16 not he in the future, but is there anything that we have not 17
_ covered and you-feel should be brought to our attention?
I 18 think you have covered everything.
19 MR. KING:
I think we have covered all the major 20 points.
21 DR. APOSTOLAKIS:
That's why we have subcommittee 22 meetings; right?
~
23 DR. WALLIS:
Well, I was sitting here for some 24 time, not knowing what was going on, and it would help if 25 you could give me a two-minute summary of what you thought ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
295 1-was going on, in very plain English.
()
2 MR. KING:
We have had a safety goal policy since 3
1986 that has been used in various -- to form the basis for 4
various' Staff actions, and elements of it or things derived 5'
from it show up in various Staff documents, reg guides, reg 6
analysis guidelines.
7 Things have also happened that make that out of 8
date.
We are doing things above and beyond what the safety.
9 goal policy originally intended when it was written.
The 10 question is, should we come back and update. that safety goal 11 policy to reflect what we are currently doing, and to also 12 address some areas that were recognized as maybe not 13 addressed when it was originally written, and the question 14 I
is, one, if we are going to do that, what are those areas
(,e 15 that we need to update; and two, what is the priority that 16 we ought to give to this activity.
17 DR. APOSTOLAKIS:
And we know much more now, the 18 PRA, the state of the art has advanced.
I mean, remember,
-19 the original statement was published in '86, but the debate 20 started in 1980, so it really reflects the technology of the 21.
early '80s.
22 MR. KING:
And the Commission guidance on how to 23 use it didn't come out till 1990.
So it boils down to a 24
. question of what are the benefits and what are the drawbacks
.25 of doing this, and we are trying to make a decision where do ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
296 1
we go from here.
What do we tell -- what recommendation do
()
2 we give to the Commission.
And we are at the point where we 3
are saying we really need some more time to look at'this, 4
it's not an urgent matter at this point in time.
5 DR. APOSTOLAKIS:
Okay.
Anything else, Tom?
6 Anything else from members of the committee?
7 Back to you, Mr. Chairman.
One minute before 8
10:00, 9
CHAIRMAN SEALE:
George,.I really appreciate how 10 much we missed you yesterday.
11
[ Laughter.]
12 DR. SHACK:
I'm glad your watch is wrong, George.
13 You actually left us with three minutes.
14' CHAIRMAN SEALE:
Tom, let me say I'm very -- I h
15 have great confidence in the effort that's going forward 16 because.of the very succinct way that you demonstrated that 17-
.you have not lost sight t-
.ne objectives, and your summary 18 that you gave for Graham's benefit, and I should say for the 19 benefit of all of us.
20 Scott, do you have anything you want to add to 21 what's been said?
l 22 MR. NEWBERRY:
No, I don't.
I think the only 23 thing I said was on the resources side.
NRR will be -- I 4
24 mean there is a concern there.
We will have to respond to 25 the industry proposals to implement the reg guides, and I
'l J
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 1
297 i
i 1
think that's part of the equation.
()
2 CHAIRMAN SEALE:
I think that we share with you 3
the concern.that any proposals and the implementation of any 4
of the results of the pilot studies be given high priority 5
to give proper recognition of the very clear expenditure of 6
resources and the talent and so forth that those proposals 7
represent on the part of the industry, and we look forward 8
to hearing from them.
9 Do the' industry people have anything they want to 10 say?
11 Okay.
With that, we are going to take -- thank 12 you again very much.
We are going to recess for 15 minutes 13 and be back at a quarter after.
14
[ Recess.]
)
15 CHAIRMAN SEALE:
We'll come back into order.
16 At this time we have a presentation from the Staff 17 of the proposed final SRP sections and reg guides for j
}
18 risk-informed performance-based regulation and related 19 matters.
20 Again, George, you are chairman of this j
21 subcommittee, so I will turn it over to you to introduce our 22 speakers, and any other introductory remarks you wish to 23 make.
24 DR. APOSTOLAKIS:
Thank you.
Again, we met,cith 25 the Staff, the subcommittee met on February 19th, and we ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
298 i
1
- discussed regulatory guides.
There were several comments I
()
2 that were made at the time, and the Staff promised to 3
address them and tell us today what your position is, as I 4
recall, and I suggest again, in the interest of time, that 5
we limit the presentation of the general material that was 6
presented before and'maybe focus on the issues that were i
7 raised and how you propose to resolve them.
Is that a 8
reasonable way to proceed?
Okay.
So the floor is yours.
l 9
MR. FISCHER:
Thank you.
My name is Dave Fischer, 10 and I am in NRR's mechanical engineering branch.
We had a 11 team of people develop the guidance documents and 12 risk-informed in-service testing.
Most of the team members 13-are here.
I think Gareth Parry, who is also on the team, is I
14-in the audience and we have some other people that helped O(,,/
15 out considerably.
They are in the audience to answer 16 questions, if need be.
17 What I would like to do -- this slide is an 18 outline.
Basically what I want to do is cover five key 19 points.
That is how we changed the document based on 20 changes that were made to the general guidance documents, 21 based on changes -- or excuse me, based on public comments 22 that we received on the reg guides and SRP, based on i
23 interactions of the pilot plant, based on comments with the
)
24 ACRS subcommittee, and based on OGC comments., and I will 25 primarily focus on the changes made based on the ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
l 1
i 299 j
i subcommittee.
2' The background -- there are just some key dates 3
here, if the committee wants to look at some key dates 4
related to our development of the reg guides and SRP on IST, 5--
they are here.
6 The first group of major changes was things that 7
we did to the risk-informed IST program in response to 8
changes that were made to the general guidance, that because 9
they were_being developed in parallel, we had changes that 10 we had.to kind of mirror in the IST guidance when they were 11 made to the general guidance.
12 Most of them -- what we did was restructured the 13 in-service testing guidance documents, and we changed the 14 content of the guidance documents, primarily in the areas of
~'N (Q
15 PRA and integrated decision-making and performance 16 monitoring to comply with the general guidance.
So I don't 17 want to go into any more detail on that.
18 The public comments that we got on the 19 risk-informed IST guidance -- there are several what I will 20 call major changes that we made based on public comments and 21 I will quickly run through several of those.
22 Performance monitoring for high and low safety 1
23 significant components was clarified.
There'was some 24 confusion.
What we meant by performance monitoring for all 25 of the components in the risk-informed IST program is the ANN RILEY & ASSOCIATES, LTD.
\\
Court Reporters.
1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
300 1
inservice testing program itself.
There wasn't additional
()
.2 monitoring we were suggesting be done like counting number 3
of demands and counting failures on demand and what-not.
4 We were not' suggesting that there be a whole new f
i 5
monitoring program for' inservice testing.
It is in fact i
1 6
just the inservice testing itself.
7 We also clarified the relationship to the 8
maintenance rule.
We in response to some changes that were 9
made to the general Reg Guide said that the monitoring 10 programs should be coordinated, so what we ended up saying 11 in the IST guidance is that it may be advantageous to adjust 12' the maintenance rule performance criteria to meet the 13 guidance in the risk-informed IST guidance document 14 The documentation that is in the risk-informed IST
)
15 guidance was significantly reduced and clarified.
We 16 elaborated on what should be included in the licensee's l
17 risk-informed IST program submittal and we elaborated on 18 what should be available on site and made it much more clear 19 what we wanted in terms of documentation.
20 We talked about the cumulative impact of all-21 risk-informed IST program changes and this is a section that 22 may be a little bit confusing, but what we were trying to do 23 was provide some guidance on how licensees could change 24 their risk informed IST program after the staff had done its 25 initial review of the program, so that they could make
/'
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
}
l 301 I
subsequent changes when they developed more information.
2 As they work their way through this risk-informed 3
, program, they'd make changes to their program.
What we 4-clarified is that a cumulative impact of all changes,.that 5
is initial approval of the program plus later changes, I
6 should comply with the acceptance guidelines in Reg Guide 7
1.175, which refers you back to Reg Guide 1.174, and what 8
this does is it provides the licensees with some flexibility 9
to move around in those regions of Reg Guide 1.174 for core 10 damage frequency and large early release without coming back 11 to the Staff for approval.
12 We also clarified that we didn' t-expect licensees 13 to go back and adjust the failure rates in their PRAs every 14 time they had a failure of an IST component.
We suggested 15 that this be done on a periodic basis and we deleted the 16 specific periodicity with which licensees needed to go back 17 and re-evaluate their overall program.
18' In terms of what we did in response to the pilot 19 plant reviews, we clarified that guidelines on the scope 20 level of detail and quality of PRA for IST application.
21 We did this mostly because of the limited role PRA 22 plays in the decision-making process.
That is, in the IST 23 context what we did was we identified components that were 24 candidates for reduced testing and also candidates for 25 enhanced testing.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
302 1
We did not try to identify components that you 1(
2 could just stop testing on or you could run to failure, so 3
the PkA was used in a very limited role in this application 4
and we revised the PRA guidance and the guidance documents 5
appropriately.
6 We took the guidance that was in the general Reg 7
Guide on defense-in-depth and margin of safety and tried to 8
make it more specific to the IST application.
This was 9
largely in response to the way the pilot plant licensee had 10 described how they were going to deal with defense-in-depth 11 and margin of safety.
12 There was a section in the draft guidance 13 documents that talked about making changes just to the test 14 interval and we deleted that section and now it just talks 15 about making changes to the program, both test intervals and 16 test methods.
17 The reason we did that was because we really 18 didn't want licensees to make changes just to the frequency.
19 We didn't want them to take the current IST test method and 20 continue to do that and just extend the intervals for the 21 lows.
We wanted them to take a proactive look at the 22 testing that was done on both the highs and lows and make 23 sure that they were doing an effective test.
24 The other two are kind of trivial changes.
25 With regards to what we did in response to the O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
i l
303 1
ACRS subcommittee comments, we clarified in the guidance f$)
2 document that components may be omitted from the PRA model v
3
_for any one of a number of reasons.
We made it clear that 4
just because a component is not modeled, that doesn't mean 5
it's an unimportant component, but we specified some reasons 6
why some'-- some rational reasons why they might be deleted 7
from the models and then we also provided some guidance for 8'
the integrated decision-making process for how they should 9
go through and deal with unmodeled components.
'10 We deleted the equation where the failure is equal l
11 to half a Lambda T and we also made reference NUREG 6141 and 12 Dr. Apostolakis's paper on state of knowledge.
13 The last item on this viewgraph is that we --
14 DR. APOSTOLAKIS:
By the way, my middle initial is
)
15 "E"
16 MR. FISCHER:
Okay.
17 DR. APOSTOLAKIS:
Not "A"
18 MR. FISCHER:
Thank you.
I mean I got the 19 uncertainty right this time.
20
[ Laughter.]
21 MR. FISCHER:
The last bullet is kind of a 22 significant change.
23 We clarified in the guidance document that the 24 test intervals for components have to meet two criteria, and 25 the first one is that the test interval should be i
O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034 i
304
\\
5L significantly less than the expected time to failure of the
()
2~
component as assumed in the PRA and we provided an example 3
.at_the suggestion of the subcommittee that that be an order 4
of magnitude less.
5 We had this in the draft document.
We were going 6
to'take it out and then we ended up putting it back in.
7 We feel it is important that the expected time --
8 excuse me ---that the interval be significantly less than 9
the expected time of failure.
10 Then the other piece of this thing is that the 11 next sentence says in addition there has to be sufficient 12 margin in the component's capability to get you to the next 13.
test and for an MOV that is like required -- the torque 14 that's the output of the motor has to have sufficient margin
)
15 above the required torque to get you to the next test 16 interval and they have to be able to project that upon 17 components.
18 So really in the risk-informed IST program there 19 are two aspects.
There is kind of a PRA aspect where you i
20 have to have an interval less than the expected time to 1
21 failure, but you also have to have this mechanistic margin 22 to get to the next test.
1 23 There are also a few changes that -- I'll just 24 highlight the first three on this viewgraph that we made in 25 response to Office of General Counsel.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
305 1
We added a section up-front on both the Reg Guide
()
2 and the SRP to clarify that licensees can still come in with 3
purely deterministic IST program changes and they do not 4
need'to come in with risk-informed programs and meet all of 5
this guidance.
6 We changed a couple of the titles of the sections 7
in the Reg Guide and SRP to be more descriptive of the 8
content of those sections and we clarified then the 9
risk-informed IST Reg Guide that licensees are expected to 10 have processes and procedures in place related to making 11 program changes.
12 We had said that in the SRP, that the reviewer was 13 supposed to go and look at the procedures and processes the 14 licensee had, but we kind of omitted it in the Reg Guides so
()
.15 we made the Reg Guide and the SRP consistent and OGC picked 16
- up on that and we made those changes.
17 Those are the major changes that we made to the 18 document since it wenc out for public comment and I would 19 ask if any of the other team members have anything they 20 would like to add or -- Dr. Apostolakis?
21 DR. APOSTOLAKIS:
So essentially you did not 22 contest any of the comments that we made last time?
23 MR. FISCHER:
No, we thought they were very useful 24 comments.
25 DR. APOSTOLAKIS:
Okay.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
306 1
MR. FISCHER:
There weren't a lot of comments, a b' )
2 lotLof changes that we made, but we'took all the 3
subcommittee's comments to heart and were able to make some 4
changes that we thought were reasonable.
5 Yes, sir?
6 DR. WALLIS:
Is this just a catalogue of 7
essentially housekeeping changes or do they respond to some l
8 change in experience or philosophy or something more 9
fundamental?
10 MR. FISCHER:
I would not characterize them as 1
11 housekeeping changes.
12 I think that I would characterize them as the 13 Staff has been learning over the last two years, working to 14 develop general guidance on risk-informed applications, and
()
15 we have thought long and hard about things like
)
16 defense-in-depth and margins and we have made some changes 17 to be consistent with what we have learned in the last two ~
j 18 years.
19 We have also been in parallel with developing the 20 general guidance.
We have been working with the pilot 21 licensees and the pilot licensees have been reviewing the 22 guidance that we have prepared on risk-informed IST and 23 giving us some feedback about what is reasonable, what is 24 not reasonable, and we have some of those changes as I have i
25 been described reflected in changes to the guidance O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
307 1
documents.
2 I don't think they are' housekeeping at all.
I 3
think some of.them are just making it more practicable, 4
making it a little bit more -- I want to say cost-effective 5
for licensees to implement a risk-informed program.
6 We had in the draft that went out for public 7
comment an awful lot of onerous things in there and partly 8
because we were thinking that the guidance documents --
9 partly because the Staff was thinking that maybe what we 10 were doing was taking low safety significant components and 11 almost not testing them.
12 You know, we evolved into a position where we 13 decided that we want both high and low safety significant 14 components to continue to be tested and to continue to be
\\
15 operable over the interval, and we used PRA a lot less than 16 maybe we originally intended.
17 DR. APOSTOLAKIS:
Thank you very much.
So-who is 18 next?
l 19 MS. GILLES:
Technical specifications.
20 DR. APOSTOLAKIS:
All right.
Technical 21 specifications.
22 MS. GILLES:
Good morning.
My name is Nanette 23 Gilles.
I am at NRR in the Technical Specifications Branch.
24 The rest of the Technical Specifications team is 25 here with me -- Roy Woods from the Office of Research and b
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
308 1
Millard Wohl from NRR's PRA Branch.
2 We are going to discuss the summary of the major 3
changes to the risk-informed technical specifications 4
Regulatory Guide and standard review plan.
5 They are broken down into two categories -- first, 6
those changes made based on comments by the subcommittee 7
from our meecing with them on February 19th, and second, the 8
rest of the changes due to public or Staff comments.
9 The first change we made based on an ACRS PRA 10 subcommittee comment had to do with the section entitled 11 Scope of PRA for Tech Spec Changes -- that's Section 4.3.2 12 in the Reg Guide and there was a statement in there 13 concerning shutdown risk analysis for certain types of 14 technical specification changes and that in some' cases where 15 a full shutdown risk analysis was not available, 16 semi-quantitative analysis made be appropriate and the 17 comment by the subcommittee was what did we mean by 18 semi-quantitative analyses?
19 We added some examples which include separate 20 fault tree analysis or failure modes and effects analysis.
21 The second comment by the subcommittee was in 22 Section 4.3.3.1 on modeling detail needed for technical 23 specification changes..
There was a lengthy section that 24 discussed what you might do in a case where the technical 25 specification system you were concerned with was not O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
309 1
completely modeled in your PRA.
(
)
2 In that case, the subcommittee thought that 3
perhaps there was too much detail for such a subject in the i
4 absence of an example, 5
At the subcommittee meeting we had thought that it i
6 would be possible to reference another document where 7
perhaps the text came from.
Upon research after that I
8 meeting, we discovered that this text did not come from 9
NUREG 6141, the handbook, which we had originally thought it 10 was original text in the Reg Guide, and so we could not 11 provide a reference.
12 Upon further discussion with some of the other 13 authors we concluded that there was a real life example out 14 there in regards to a Westinghouse topical report for
/( j\\
15 reactor protection system technical specification changes, 16 and so we have included that as an example where you might 17 need to create separate fault trees and combine them with 18 the existing PRA.
19 CHAIRMAN SEALE:
Let me make sure I understand.
20 You replicated the example from the Westinghouse 21 report or you referred to the example in the Westinghouse 22 report?
23 MS. GILLES:
We referred to the example, not to 24 the Westinghouse report specifically, but to the example of 25 the case used in the Westinghouse report.
(m i
4 ANN RILEY & ASSOCIATES, LTD.
' x
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
310 1-DR. APOSTOLAKIS:
Which section is this?
j
- (
)
2 MS. GILLES:
It's page 15, 4 3.3.1.
3 It is in the copy you received yesterday -- should 4
be highlighted, _the changes that were made.
5 DR. APOSTOLAKIS:
So for example detailed separate 6-fault tree modeling of the reactor protection system 7
combined with the existing PRA model -- is that the one?
-8 MS. GILLES:
Yes.
9 DR. APOSTOLAKIS:
Okay.
10 MS. GILLES:
The'next comment appeared in Section 11 4.4, which is the section on acceptance guidelines for 12 technical _ specification changes.
This' appears on page 24.
13 In the item that described the incremental conditional core 14 damage probability guideline there was a statement that said
()
15 that the ICCDP contribution should be distributed time such 16 that any increase in the associated instantaneous risk is 17 small and the subcommittee comment was that instantaneous 18 was not the appropriate term and that conditional was more 19 appropriate, so that change was made.
20 In Section 5.1, which 10 entitled the Three-Tier 21 Implementation Approach, the comment made was that the 22' discussion there did not reflect an earlier discussion of 23 the same subject in Section 4.3 and that the three-tiered 24 implementation approach really was there to help ensure 25 defense-in-depth was adequately maintained.
g ANN RILEY & ASSOCIATES, LTD.
\\~/
Court Reporters 1250 I Streec, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
311 1
The Section 5.1 was changed to discuss
()
2 defense-in-depth similar to the previous discussion.
The 3
' discussion on the principles of performance based 4
implementation and monitoring strategies be employed to 5
account for uncertainties has been deleted based on the 6
subcommittee's comments.
That is Section 5.1 on page 25.
7 The one comment from the subcommittee that we did 8-not incorporate was the comment to consider using the 9
incremental conditional core damage probability guideline as 10 an acceptance guidelines for determining the Tier 2 11 configurations.
12 After some additional thought and discussion, it 13 was concluded that it would not be an easy measure to use 14 for Tier 2 because the second system that you bring into the
()
15 analysis at that-point most likely will have a different 16 allowed outage time than the one you are considering for the j
17 base system and could likely have a shorter allowed outage 18 time'so that you would not be looking at the same time 19 period, allowed outage time period, for both systems.
20 Based on that we decided not to include this as an 21 acceptance guideline for Tier 2 but to leave what was I
22 originally in the document.
1 4
23 DR. APOSTOLAKIS:
Well, what you really are 24 concerned about is that you may have what you call 25 potentially high risk configurations in which equipment in j
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 i
Washington, D.C.
20005 (202) 842-0034 l
312 1
addition to that associated with the change were to be taken
()
2.
out of service simultaneously, so you are really concerned 3
about the overlapping right?
4 MS. GILLES:
Right.
5 DR. APOSTOLAKIS:
Not necessarily that throughout 6
the AOT --
7 MS. GILLES:
Right.
8 DR. APOSTOLAKIS:
Now when you identify a 9
potentially high risk configuration in this sense, it seems 10 to me you have done most of the work because now you know I 11 am looking at this component and I am concerned that this 12 other component may be out too.
13 Calculating the ICCDP for the combination should 14 not really be that difficult.
()
15 MS. GILLES:
But it would require the licensee to 16 do more.than we had asked them previously, which means it 17 would require them to along with finding importance measures 18 for all those other systems, which is what we had envisioned 19 earlier, they would have to also look up the allowed outage 20 times for all those other systems and tech specs and then go 21 back and do another ICCDB calculation based on whichever of 22 the two times was shortest, because that would be the 23 longest time you would be allowed to be in the combination 24 configuration.
25 DR. APOSTOLAKIS:
But wouldn't that have to be
[/l ANN RILEY & ASSOCIATES, LTD.
A-Court Reporters 1250 I Street, N.W.,
Suite 300 J
Washington, D.C.
20005 (202) 842-0034 l
313 1
done by the evaluator here anyway?
I mean maybe not 2
numerically but the moment someone says this is a 3
potentially'high risk configuration then the.t person 4
postulates that two or three' components are out at the same 5
time, right?
6 MS. GILLES:
Right.
7 DR. APL "OLAKIS:
So --
But we ' id not require them to bring 8
MS. GILLES:
d 9
in the allowed outage times associated with the other 10 systems previously.
I 11 DR. APOSTOLAKIS:
I understand, so after this 12 identification is done, then what are you supposed to do?
I
'l 13 mean you are not doing the quantification.
14 All you are saying is, look, make sure that this 15 other --
16 MS. GILLES:
When they identify high risk 17 configurations we.have asked them for those that they 18 perceive to be outliers, to' write some kind of restriction 19 either in procedures or in the technical specifications 20 themselves for those particular configurations.
21 DR. APOSTOLAKIS:
And how is this identification 22 going to be done?
23 MS. GILLES:
Well, what we had talked about and 24 suggested previously was to take the system you are 25 considering for the change, assume that is out of service, ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
314 1
and then-run importance measures on the rest of the systems.
()
2 DR. APOSTOLAKIS:
And take the highest?
3 MS. GILLES:
Right -- or any that are outliers 4
above a certain threshold which we did not define for the 5
licensees.
6 DR. APOSTOLAKIS:
Is that here in the Guide?
7 MS. GILLES:
It is in there in very general terms, 8
not specific.
9 DR. APOSTOLAKIS:
And then you still need the 11 0 Third Tier then?
11 MS. GILLES:
Correct.
12 DR. APOSTOLAKIS:
The overall configuration risk 13 management.
Well, I guess it could be -- so you are saying 14 that the way you have it now will be less demanding in terms
()
15 of resources and --
16 MS. GILLES:
We believe so.
17 DR. APOSTOLAKIS:
Yes.
18 MS. GILLES:
In addition, we felt a little uneasy 19 about inserting an acceptance guideline that the industry 20 and the public had not had a chance to comment on at this 21 point in the process.
22 DR. APOSTOLAKIS:
Yes.
Okay -- and we still don't 23 have criteria on that include the frequency of outages, 24 right?
This is per outage?
25 MS. GILLES:
Per outage and then the licensee ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
315 1
would have to estimate the frequency they expect to use the
[v\\
2 outage when they are doing the calculation for the Delta 3
core damage frequency.
4 DR. APOSTOLAKIS:
Okay -- so that is the only 5
place that if you have very high frequency you will catch it 6
and say that is not allowed?
7 MS. GILLES:
Right, but again that is an estimate 8
based on how they expect to use that allowed outage time 9
once it is in place.
10 DR. APOSTOLAKIS:
On the average how many are we 11 talking about per year?
12 MS. GILLES:
How many usages?
13 DR. APOSTOLAKIS:
Yes.
Outages.
14 MS. GILLES:
I wouldn't say more than one or two
)
15 in most cases probably.
16 DR. APOSTOLAKIS:
That is what I suspected.
17 MR. WOHL:
Most of these extended AOTs are 'or 18 preventive maintenance.
We have to keep remembering that 19 because also the ICCDP is lower than for corrective 20 maintenance.
21 DR. APOSTOLAKIS:
Now what if someone came back 22 and said, okay, you think that your Second Tier, that you 23 have identified a potentially high risk configuration, but I 24 took the time and I calculated the ICCDP assuming that both 25 of them are out and I'm still below your goal.
[
ANN RILEY & ASSOCIATES, LTD.
\\-
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 316 1-Would you buy that argument?
\\
2 MS. GILLES:
I believe so.
Yes.
Good argument.
[U 3
DR. APOSTOLAKIS:
And do you say that anywhere?
4 MS. GILLES:
No, we don't.
1 J
5 DR. APOSTOLAKIS:
Do you think that is something 6
that should be said or is it understood?
7 I mean it is a matter of adding a sentence 8
'perhaps?
9' MS. GILLES:
It could be added.
10 DR. APOSTOLAKIS:
So you promise that you will do 11 it or it could be done?
12
[ Laughter.)
13 DR. MILLER:
Maybe could be or should be.
14 DR. APOSTOLAKIS:
Always give some ideas to people 15 because --
16 MS. GILLES:
Right.
17 DR. APOSTOLAKIS:
-- instead of saying no, I will 18 make sure this will never be down at the same time with 19 that, I did the calculations ~and it really does not, 20 according to your rules.
21 MS. GILLES:
That would fit in because we just 22 suggest the other way to do it, so we could suggest this way 23 also.
24 DR. APOSTOLAKIS:
1 Other words, what you can say 25 is that you define on page it was the old page 11 --
1 TiNN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suit? 300 Washington, D.C.
20005 (202) 842-0034 i
317 1
14S. GILLES:
Page 12?
j( )
2 DR. APOSTOLAKIS:
It's still page 11.
In the 3
footnotes you define ICCDP and ICLERP -- what is ICLERP?
4 MS. GILLES:
I guess so.
IC LERP is how I see it.
5 DR. APOSTOLAKIS:
And the other one?
6 MS. GILLES:
ICCDF -- I don't think you can 7
make --
8 DR. APOSTOLAKIS:
You have to find a vowel and put 9
it in.
10 Somewhere here on this page you might want to say 11 that these two concepts could also be used in a more general 12 way to involved more than one item's being out if analysts 13 deem that this is something that would help their 14 argument -- something to that effect.
15 In other words, you are not requiring it but this 16 is not a definition that is rigid.
You know, it gives me a l'7 measure of the conditional core damage frequency given that 18 things are out of service, but what you really want to see 19 is, you know, you want to see it applied to individual 20 components.
21 MS. GILLES:
Right.
22 DR. APOSTOLAKIS:
If they want to do more, that is 23 their business, so a sentence or two to that effect probably 24 would help and that then could be used in the Second Tier 25 argument.
O ANN RILEY & ASSOCIATES,.LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005
'(202) 842-0034
318 l'
That would make it more flexible, in other words.
(
-2 MS. GILLES:
Okay.
3 DR. APOSTOLAKIS:
I mean, it doesn't restrict 4
anything.
5 MS. GILLES:
Right.
6' MR. WOHL:
So far most of the-Tier 2s that we have 7
are some of the more. obvious combinations, diesels in 8
switchyard work, diesels in turbine-driven aux feed pumps, 9
' things of that sort, which are easily spottable up front by 10 people with no PRA background whatsoever. 'But this would be 11 an additional enhancement.
12 DR. APOSTOLAKIS:
Okay.
Good.
So next slide.
13 MS. GILLES:
I believe that covers the changes 14 that -- based on comments from the subcommittee with one 15 exception.
There was a request to provide some information 16 on the uncertainty discussion.
17 DR. APOSTOLAKIS:
I thought there were a couple of 18 more comments, though.
19 On page 18, the former page 18, which seems to be 20 also the current page 18, paragraph 1.
I remember you gave 21' an answer to that, but I don't remember what it was.
After 22 the test the component's unavailability resets to zero or 23 falls in the Boolean expression.
So it's still valid?
24 And then on the next page you're talking about 25 errors of restoration after the test.
Are these consistent O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250.I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
i 319 l
l' or --
()
2 MS. GILLES:
I guess I though we had discussed 3
those during the subcommittee meeting --
4 DR. APOSTOLAKIS:
Yes, we d.id.
5 MS. GILLES:
And concluded that what was there was 6
adequate.
7 DR. APOSTOLAKIS:
Well, we're saying here on page 8
18 that the test limit at risk is normally estimated by 9
assuming that the surveillance test of a component detects 10 the failures and that after-the test the componert's 11 unavailability resets to zero or falls in the Boolean 12 expression.
13 But then on the next page we say downtime to 14 conduct the test, errors of restcration after the test.
So
)
15 the component unavailability is not set equal to zero then, 16 because you may have that error.
I mean, are paragraphs 1 17 and 4 consistent, or are they talking about different 18 things?
19 I remember we had a discussion, but I don't 20 remember a resolution.
21 MR. WOHL:
I don't see a contradiction here.
We
.22 can have a perfectly good test and say that the equipment's 23
'available, but there can be errors of restoration after the i
24 test, valves mispositioned, things of that sort.
So --
25 DR. APOSTOLAKIS:
They are two different things, l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300
{
Washington, D.C.
20005 (202) 842-0034 l
320 l
.you're saying.
)
~2 -
MR. WOHL:
Yes, I think so.
%/
i 3:
'DR. APOSTOLAKIS:
Two different things.
4 MR. ~WOHL:
Yes.
5 DR. APOSTOLAKIS:
Also in paragraph 3 it says in 6
most PRAs the quantification scheme is equivalent to 7
assuming nonstaggered testing.
8 Is~that really true?
9~
No, I.think we agreed though the PRAs do it 10 wrongly.
They take lambda tau over 2.and multiply it, 11 right?
Yes, that's wrong.
But they don't assume that it's 12 nonstaggered.
They're just doing it incorrectly.
13 MR. PARRY:
Can I add a comment here?
This is 14-Gareth Parry from the staff.
It's effectively making an O
(,j 15 assumption that it's a nonstaggered test.
Because you take 16 the unavailabilities, this lambda T by 2, and you just i
i 17 multiply them.
18 DR. APOSTOLAKIS:
No, but in the nonstaggered 19 itest, in.the sequential test, it's lambda tau squared over 20 3.
'In the staggered it's lambda tau squared over 24 or 21 something.
22 MR. PARRY:
Right.
23 DR. APOSTOLAKIS:
But what they are doing is J24 la1mbda tau squared over 4, which is neither.
It's plain 25 wrong.
l O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
321 1
MR. PARRY:
No, it's --
()
2 DR..APOSTOLAKIS:
The problem is that if you want 3
to. modify your computer codes it's such a major undertaking 4-that whether you want to live with an error of dividing by 4 5
instead of 3, these numbers, I mean, you probably -- but the 6
thing is that they are really wrong.
There is a coupling 7
that is introduced by the tests --
8 MR. PARRY:
Yes.
9 DR. APOSTOLAKIS:
The way you do the test.
But 10 this statement is not really correct.
I would drop it.
In 11 most PRAs, a quantification, I would drop it.
I don't want 12 to say that in most PRAs it'c done incorrectly, but 13 staggered or sequential test strategies are commonly used, 14 and t' appropriate expression for the unavailability should 15 be employed.
And in fact I think NUREG-4161 that was cited 16 earlier --
17 MS. GILLES:
Yes.
18 DR. APOSTOLAKIS:
Does give all these.
Are you
)
19 citing that report?
20 MS. GILLES:
It's in the reference book.
21 DR. APOSTOLAKIS:
Yes, let's see.
22 MS. GILLES:
It's also cited in the appendix.
23 DR. APOSTOLAKIS:
It's before the appendix?
24 MS.-GILLES:
It's in the reference list and then 25 cited in the appendix also separately.
ANN RILEY & ASSOCIATES, LTD.
1 Court Reporters 1250 I Street, N.W, Suite 300
' Washington, D.C.
20005 (202) 842-0034
322 1-DR. APOSTOLAKIS:
So the reference list is --
2 MS. GILLES:
Is at the end.
3 DR. APOSTOLAKIS:
At the very end?
I 4
MS. GILLES:
Page 27, and I believe it's the last 5
reference on the list, which is'actually on page 28.
6 DR. APOSTOLAKIS:
Oh, it was not 4161?
7 MR. WOHL:
6141.
8 MS. GILLES:
6141.
9 DR. APOSTOLAKIS: '6141.
10 MS. GILLES:
It's on page 28, number 15.
11 DR. APOSTOLAKIS:
6141 is says here; right?
12 MS. GILLES:
Yes.
j 13 DR. APOSTOLAKIS:
Yes.
Okay.
So why don't you 14 modify here in paragraph 3 --
)
15 MS. GILLES:
You want to strike that sentence?
Is 16 that --
17 DR. APOSTOLAKIS:
Strike that sentence, because
~
18 it's incorrect.
And then say -- where you say staggered or 19 sequential test strategies are commonly used, and then 20 strike in most PRAs --
21 MS. GILLES:
Right.
22 DR. APOSTOLAKIS:
And -- okay.
Then you have the 1
23 right sentence.
The risk impacts of adopting different test 24 strategies, sequential versus staggered, should be evaluated 25 to determine where there is an impact -- okay?
Parenthesis ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
323 1
NUREG whatever it is, 6141.
That sends a clear message.
m
(
)
2 And then leave it at that.
3 MS. GILLES:
Okay.
4 DR. APOSTOLAKIS:
You don't have to get into the
.5 details.
6 MS. GILLES:
Anything else?
7 DR. APOSTOLAKIS:
Are you done?
8 MS. GILLES:
The next slide covered the changes 9
based on public comments, which the subcommittee has already 10 seen.
11 DR. APOSTOLAKIS:
Now -- oh, you gave us the basis 12 for the criteria, right?
13 MS. GILLES:
Yes.
14 DR. APOSTOLAKIS:
I remember when we asked you, i
/%(,)
15 you gave them, but you don't put them in the report.
16 MS. GILLES:
No.
17 DR. APOSTOLAKIS:
And with the reason being?
18 MS. GILLES:
I don't know.
I guess we --
19 DR. APOSTOLAKIS:
Is there any reason why the user 20 of the regulatory guide should not know where the criteria 21 came from?
I mean, I'm just asking.
I remember what you 22 gave us made sense.
23 MR. KING:
There is no reason you can't put it in 24 there.
Generally to debrief and just give the bottom line j
25 we don't put a whole bunch of bases in regulatory guides.
l l
(f~ )
ANN RILEY & ASSOCIATES, LTD.
N/
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 l
l
324 1
DR. APOSTOLAKIS:
Well, would the footnote giving I.. )
(~/
maybe semiquantitative heuristic argument as to why these 2
3 numbers were selected be enlightening to the licensees and 4
to everybody involved?
{
5 I'm not asking for a major treatise.
6 MS. GILLES:
Sure.
7 DR. APOSTOLAKIS:
I mean, what you gave us last 8
time I thought, you know, was convincing.
I mean, you have l
9 to do something, right?
10 MS. GILLES:
Okay.
We can --
11 DR. APOSTOLAKIS:
And that something has some sort 12 of a basis.
Is there any reason why we don't want to 13 explain that?
14 MS. GILLES:
I don't think there's any reason we 15 can't put a footnote.
16 DR. APOSTOLAKIS:
In a footnote.
In a footnote.
17 I think that would be helpful.
18 Now you say that the other comments are on your i
19 responses to the public comments?
20 MS. GILLES:
Yes, the changes made.
Would you 21 like to hear those?
22 DR. APOSTOLAKIS:
Okay.
Do you have anything 23 spectacular there?
24 MS. GILLES:
The subcommittee has heard them, 25 but --
A ANN RILEY & ASSOCIATES, LTD.
A s#
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
b l
j 325 1
DR. APOSTOLAKIS:
And we did not contest any of 2_
these?
-3 MS. GILLES:
No.
4 DR. APOSTOLAKIS:
Okay.
Does anybody want to do 5
that now?
6 MS. GILLES:
And the final item you asked for was 7
an explanation of the statement regarding uncertainties for 8
technical specification changes.
9 DR. APOSTOLAKIS:
Where is that?
10 MS. GILLES:
Do you recall that?
11 DR. APOSTOLAKIS:
On page 5?
12 MS. GILLES:
It's in section -- page 5 of the 13 slides.
I'm sorry.
14 DR. APOSTOLAKIS:
Oh.
Oh, oh.
I'd
(,/
15 MS. GILLES:
It's in -- the statements in
)
16 section -- page 19, section 4.3.5.
17 DR. APOSTOLAKIS:
Now where did we say that?
18 MS. GILLES:
The first sentence, which says that 19 sensitivity analysis have shown that the risk resulting from i
20 Tech Spec AOT changes is relatively insensitive to 21 uncertainties.
And you asked us to bring some information 22 to back that up to the full Committee meeting.
23 DR. APOSTOLAKIS:
Oh, okay, so you're still 24' defining that it's a statement?
Okay.
25 MS. GILLES:
Yes.
And this is a copy of a page
~
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 12b0 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 i
326 1
out of one of the requests for information to the licensee 2
.and their response regarding uncertainties, _and I'm going to 3-let Mr. Wohl explain --
4 MR. WOHL:
This is for one of the candidate CE 5
owners group plants.that are being considered as a group of 6
plants.
The first column of numbers represents the 7
statistical parameters with the current AOT, and the second 8
column represents the parameters with the extended AOT.
9 They are three days to seven days.
And you can see that we 10 were trying to discuss last time that the shape of the 11 distribution that is a probability density function of the l
12 CDF remains just about the same, and since all of these 13 higher moments are just about the same the point estimate 14 goes up a little bit.
15 We feel that there's -- that that's ample 16 justification for making the subtraction that we do when we 17 calculate ICCDPs.
The shape of the distribution stays 18 essentially the same.
You can see 1.99E minus 9 for 19 variance for the current AOT, 2E minus 3 for the extended 20 AOT, et cetera, for the skewness and kurtosis.
21 So there isn't any real change in the shape of the
-22 distribution, and subtracting the values from each other 23 makes sense under these conditions.
This was done using the 24 LHS procedure and the TEMAC code, which are codes developed 25 for the NRC by Sandia a number of years ago.
O ANN._RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
327 1
DR. APOSTOLAKIS:
And what's the purpose of saying 2
all this?
3 MR. WOHL:
Well, we wanted to show that in fact 4
the shape of the probability density function of the core 5
damage frequency is not changed by extending the current AOT 6
from three to an extended value of seven -- just to give you 7
a feel.
8 DR. APOSTOLAKIS:
It's a minimal impact.
You're 9
right.
i 10 MR. WOHL:
Yes.
It's just not going to be -- if 11 that shape of the distribution changed in a very distorted 12 fashion, then we'd have to reexamine the way in which we're 13 calculating things.
14 DR. APOSTOLAKIS:
Right.
Right.
That's okay.
15 Now, there is a typo in 19.
Are we supposed to 16 tell you --
17 MS. GILLES; Sure.
18 DR. APOSTOLAKIS:
Two lines -- the penultimate 19 line.
Subject the plant to a variation in its exposure --
20 no apostrophe there.
21 MS. GILLES:
Which line are you at?
22 DR. APOSTOLAKIS:
Line before last.
23 MS. GILLES:
Oh, okay.
Thank you.
24 DR. APOSTOLAKIS:
There are two things you are 25 going to do, right?
One is add a few lines on the basis for
(
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C.
20005 (202) 842-0034
f.
328
~1 choosing the particular criteria you chose.
()
2 MS. GILES:
Yes.
3 DR. APOSTOLAKIS:
And the other one is you will i
4 cite -- I mean this staggered testing business, --
5 MS. GILES:
Right.
6 DR. APOSTOLAKIS:
-- you will cite the NUREG 7
report?
l 8
MS. GILES:
Correct.
9 DR. APOSTOLAKIS:
And drop that sentence that --
10 MS. GILES:
Drop that sentence.
11 DR. APOSTOLAKIS:
Yeah.
12 MS. GILES:
And we are also going to add statement 13 suggesting using the ICCDP in combination of systems.
For 14 Tier 2.
15 DR. APOSTOLAKIS:
Any other
- any comments from 16 members?
Yes.
17 DR. WALLIS:
I have my naive comment here.
Do you 18 find that the licensees understand all this stuff and how to 19 implement in it a professional way without fudging?
20 MS. GILES:
Yes.
21 DR. WALLIS:
They do.
22 MS. GILES:
The licensees that we have worked with 13 to date --
24 DR. WALI f J :
So they really are up to snuff?
25 MS. GIIMS :
Yes.
O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
329 1
DR. WALLIS:
Okay.
Thank you.
[
)
2 DR. APOSTOLAKIS:
In fact, yeah, you cite the 3
combustion engineering piece of work, that was really very 4
good.
That was really very good.
5 Do you have anything else to say,. lady and 6
gentlemen?
Anything to add?
7 MS. GILES:
No.
8 DR. APOSTOLAKIS:
Okay.
Thank you very much.
9 Is there more?
Quality assurance.
10 MR. GRAMM:
Good morning.
We are here to talk 11 about graded QA.
My name is Bob Gramm from NRR.
With me is 12 Steve Dinsmore, also from NRR, and Roy Woods.
We have other 13 members of the team that participated in the Reg. Guide 14 development in the audience.
()
15 Today we will briefly cover the resolution of some 16 major public comments we received.
We will go over some of 17 the Subcommittee ACRS comments and propose a resolution of i
18 those.
And, in parallel, the Office of General Counsel has 19 provided us comments and we will address those.
20 The current draft of the Reg. Guide is dated j
21 3/2/98.
The copy we have provided you has some bar line 22 markings so you can distinguish where we have made changes, 23 either additions or deletions.
We apologize, some of it is 24 redline and strikeout text, and some you have to do a text j
25 comparison manually.
ANN RILEY & ASSOCIATES, LTD.
-/
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 j
330 1
I
.1 An overview of some of the major public comments
()
2 we received.
Several people thought we should'have a 3
greater reflection of the volunteer plan activity in the 4
guidance document-We have made specific changes in the QA 5
control section for low safety significant items to amplify.
6 upon what quality controls would be appropriate in those 7
circumstances.
And we added some text to describe 50.59 8
procedure, implementing procedure change control which'will 9
also be necessary for this particular effort, since the QA 10 program only contains a high level description and the bulk 11 of the details on how to actually conduct the program will 12-be in the implementing procedures.
13 A number of commentators indicated that the 14 proposed Reg. Guide was excessively burdensome and would not t(,jr 15 yield any benefits.
We have addressed that in a couple of 16 areas.
We have reduced the burden requirement in terms of 17-the information to be submitted, and we have also amplified 18 on the fact -- a system-by-system categorization approach 19 could be entertained.
20 The original Reg. Guide alluded -- could be read 21 that you would need to risk rank the entire plant.
This is 22 very time consuming and very expensive to do at'the front 23 end of the project, and so we have laid out an approach 24 where it could be done on a system-by-system basis, with 25 appropriate bounding studies in that particular situation.
b-ANN RILEY & ASSOCIATES, LTD.
\\w /
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 I
i t
331 1
Many comments were received for low safety 2
significant items.
We were expecting an excessive amount of 3
root cause analysis and performance monitoring.
The 4
document has been revised to focus in on failure monitoring,
'5 and the fact that we have a graduated analysis.
There would 6
be an apparent cause to at least gain some understanding of 7
the failure and that would trigger a root cause analysis as 8.
appropriate.
So it would be a graduated failure -- an 9
analysis going into more rigor as necessary.
10 Also, the commentators indicated that we should-11 give greater credit for the maintenance rule.
And, in line 12 with the overall Reg. Guide, we have inserted some text that 13 could be a starting point, both for the safety significance 14 determination, as well as monitoring.
As discussed by the 15 other application guides, there would be -- have to be some 16 augmentation required to utilize that type of approach fully 17 for the graded QA purposes.
I 18 Very briefly, at our last Subcommittee meeting, 19 South Texas was here and gave an overview presentation, so 20 this is just a synopsis.
They have an-approved QA program i
21' to implement graded QA.
At this point in time they are 22 revising procedures, training staff, developing the 23 infrastructure necessary to fully implement that.
They are 24 anticipating an April or May time frame when it will become 25 operational.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
332 1
The savings projected from this were substantial (3) 2 on the part of South Texas.
And they provided also at that 1
3 time an indication of the -- they have only ranked a limited i
4 number of systems, but they do have a scheduled effort to 5
proceed with risk ranking e'cher systems in the near future.
)
6 CHAIRMAN SEALE:
And I understand that you are 7
making, or preparing information to help the inspection 8
people to be able to monitor their program based on the 9
revised -- this revised pilot?
10 MR. GRAMM:
That is correct.
In parallel, we are 11 developing an inspection procedure.
12 CHAIRMAN SEALE:
Okay.
13 MR. GRAMM:
Some of the other areas where the Reg.
14 Guide was changed, we conformed it with the NUREG Reg.
15 Guide, particularly on the safety principles.
We clarified 16 what QA is appropriate for high safety significant but i
17 non-safety-related items.
There was some confusions that we 18 were advocating a full Appendix B approach, and that is not 19 the case.
We are saying it is voluntary up to and including 20 that level of controls.
And we discussed the treatment if 21 an item is originally low and re-categorized high at some 22 future point, how that is treated.
23 I'll spend a little bit more time here on the 24 Subcommittee comments.
We received a comment that the 25 categorization section, the tone of it should be a little b
ANN RILEY & ASSOCIATES, LTD.
\\-
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 333 1
bit less aggressive.
We have made some minor edits, l h 2
particularly on page 12 of the revised document.
V 3
DR. APOSTOLAKIS:
So let's look at them.
Page 12, 4
is that the same page 12 in the draft?
5 MR. GRAMM:
The page numbering is probably 6
different.
7 DR. APOSTOLAKIS:
Let's see.
8 MR. GRAMM:
It's Section 4.1.2.1 would be the 9
comments section number.
10 DR. APCSTOLAKIS:
Well, system functions.
Basic 11 events.
So, one, two, oh, you have changed.the thing.
So I 12 take it that these little bars here are the changes, right?
13 So this is -- when each SSC is categorized, the safety l
14 significance of all the functions that the SSC supports must 15 be known.
16 Again, must be -- what does that mean?
17 MR. DINSMORE:
This is Steven Dinsmore.
Since I j
18 changed this sentence, I'll respond.
19 DR. APOSTOLAKIS:
Yeah.
1 20 MR. DINSMORE:
Well, we weren't too sure how to 21 change the tone.
We wanted it to be clear that we were 22 interested in looking at the functions, and one of the 23 things you said was that the reason we are looking at the 24 functions is because we want to categorize the components, 25-which is correct.
So that's essentially what this says.
j
(
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
334 1
That we want to categorize the components.
But to
(
2 categorize the components, we have to know the safety 3
significance of the functions they support.
4 DR. APOSTOLAKIS:
I thought the main argument that 5
I advanced was that the various importance measures have 6
built into them the significance of these safety functions.
7-So why make such a big deal out of it?
I mean --
8 MR. DINSMORE:
Well, partly because, again, 9
there's only about 2,000 components in the PRA and there's i
10 20,000 safety related SSCs and there's a lot of functions 11 which are -- which they are going to look at which don't 12 have much to do with the PRA.
So when we work our way dcun 13 to those functions, they need to know the safety 14 significance of the functions which those components are
(~s
(,/
15 supporting.
But they are not going to have a direct j
16-importance measure on which to hang their hat, I guess.
i 17 DR. APOSTOLAKIS:
But you are not saying anything 18 here about the components that are not in the PRA.
If that 19 is your primary reason for doing this.
20 MR. DINSMORE:
Well, it is supposed to be for all 21 components, PRA or not.
22 DR. APOSTOLAKIS:
Yeah, but the PRA components, 23 again, you know, that's there.
I mean that's why 24 Fussell-Vesely came out to be, you know, X,
because that 25 component supports important safety functions.
So to come ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
335 1
'back and say now'let's look at the safety functions again, I
~
()
2 l don't know what'the purpose is.
3 MR..DINSMORE:
Well, there might be a lot of
)
4 component that are not modeled again because.you have a 5
diesel.
You model a diesel, there's two or three oil i
6 systems, there's all kinds of stuff involved.
7 So when you are doing each one of those i
8 components, you have to know -- I mean you'could say, well, i
9 okay, he has to know that this supports the diesel.
I mean j
10 there is just so much that is not modeled.
11 DR. APOSTOLAKIS:
Sub-components, you mean?
I 12 MR. DINSMORE:
Yeah.
I 13 DR. APOSTOLAKIS:
Components of the diesel itself?
14 MR. DINSMORE:
Right..Because the plant naming
)
15 scheme is on a much lower level than the basic events and 16 this provides a very clear -- see, we have to be able to go l
l 17
.back in a couple of years and figure out why they'put each I
18 one of these specific components in each place.
And'if they
]
19 do it system function by system function, we can find every 20
-component, and exactly why -- what this component does and i
21 why it is where it is.
22 Because the expert panel, of course, has this 23
' authority to put things where they -- where they feel they 1
24 are justified to be put.
But what we require is that we
)
25 would be able to find everything, and find why they put it l
O ANN RILEY & ASSOCIATES, LTD.
j Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 i
(202) 842-0034
336 1
there.
That is all kind of tied in to why we want to at
[ )\\
2 least know what system functions are being supported.
w.
3 DR. APOSTOLAKIS:
So the expert panel will process 4
this information?
5 MR. DINSMORE:
Yeah, the way it worked at South 6
Texas is the working group assembles it and gives it to the 7
expert panel and they kind of would -- they made the 8
decision on it.
9 DR. APOSTOLAKIS:
Well, now what -- I mean it 10 seems to me that an additional piece of information that is 11 very important here is, when you say, when each SSC is 12 categorized, the safety significance of all the functions 13 that the SSC supports must be known, shouldn't I also know 14 how many redundant SSCs I have doing this?
r(,)y 15 See, you are going back and breaking up the PRA 16 model into its piece again.
17 MR. DINSMORE:
Well, if you have --
18 DR. APOSTOLAKIS:
Okay.
I am telling you that 19 component X supports high pressure ir.jection, it is an 20 important safety function.
And that's all I am telling you.
21 Well, I am not giving you the full picture, because at the 22 same time there may be two other components that are 23 redundant to this one that can do the same thing.
So the 24 panel should know that, right?
25 MR. DINSMORE:
Well, I think the panel probably es
(
)
ANN RILEY & ASSOCIATES, LTD.
\\/
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
337 1
does.
If they -- if they could come up with an argument V[ h 2
that there is absolutely no CCFs or stuff between these 3
things, they could put it as low.
4 But I am not sure, if you have four diesels and 5
each individual -- you can't say that all four diesels are 6
low just because they have low Fussell-Vesely.
I mean --
7 DR. APOSTOLAKIS:
That is a different argument 8
now.
Each diesel --
9 MR. DINSMORE:
Well, it's the same thing as 10 redundancy.
You got four --
11 DR. APOSTOLAKIS:
I am trying to understand the 12 argument.
If I have four diesels, as a group, they are very 13 important.
14 MR. DINSMORE:
Right.
15 DR. APOSTOLAKIS:
That's why we are talking about 16 functions.
But each one of them, that's why I have four, is i
17 not as important as the function of the group.
So when I 18 talk about diesel A, I should be fully aware of the fact 19 that I~also have three others.
And that's what I meant by 20 the tone of.this section.
That you want them to say that A 21 supports a very important function, because that is kind of 22 conservative, but you don't want to emphasize also that you 23 have three other diesels waiting for you.
24' And it seems to me that all this stuff comes into 25 the'importance measures.
So we are backing off.
()
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
338 1
MR. DINSMORE: _Well, I guess -- well, are you kind
-()
2-
'of indicating that each individual diesel should be low?
3
~Because --
4 DR. APOSTOLAKIS:
Well, it should be -- take the 5
appropriate value, because you have four of them.
I mean 6
that stands to reasons.
That's why_they spent money buying 7-four of them.
Otherwise, they would buy only one.
I mean 8
you have to give them credit for that.
9 MR. DINSMORE:
Well, again, they can -- they can, 10 of course, place the things in low.
We are not saying that 11
-- we very clear' state that if there is a high safety 12 significant function, they may place the stuff in low, but 13 they have to have a good engineering reason for it.
14 If they -- and what we want to make sure is that 15 it is clearly laid out, and that when they put this stuff in
'16 low, they are very aware of exactly what is going on.
With 17 the diesels, everybody, of course, knows, but there's maybe 18 components in the plant which are supporting functions which 19 aren't so clear.
l 20 DR. APOSTOLAKIS:
My impression is that the tone 21 is still conservative here.
I don't know whether anyone 22 else on the Committee feels the same way.
But it seems that 23 there is great reluctance to go down to the component level 24 and use importance measures there.
And I will grant you 25 that if the component is not there, then, of course, you ANN RILEY & ASSOCIATES, LTD.
)
5 Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005
)
(202) 842-0034
339 1
have to do something else.
()
2' But I am talking about components that are l
3-modeled.
Like if basic event importance measures, rather 4
than system function importance measures, are used to 5
.directly category SSCs at the component level, the 6
. categorization process becomes more dependent on PRA l
7 characteristics such as blah, blah, blah, blah.
l
'8 In other words, I am dragging my feet.
You want 9
to go to a lower level, but, boy, look at all the problems 10 PRAs have.
Wow! You really want me to do that?
And that's 11 what makes me uncomfortable with this.
12 DR. POWERS:
On the other hand, I mean you have 13-got to recognize that the PRA is not sufficiently complete 14 to hand?
all 20,000, 50,000.
()
15 Ms DINSMORE:
Twenty to thirty.
16 POWERS:
A bunch of the devices that an 17 enginee at the plant has'to confront, procurement officers 18 have to procure, things like that.
I mean it becomes an 19 element of practicality here.
20 DR. APOSTOLAKIS:
But my problem with this is that 21 it does not give me a clear -- clear guidance as to what to 22 do with the components that are not modeled, the 20,000 23 components that are not modeled.
It makes sweeping 24 statements, ycu know, look at the function.
Because even if i
i 25
-- first of all, I have to understand why that component is l
l
.O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034 i
340 1
not modeled.
We discussed that before.
Is it screened out?
[\\ s) 2 MR. DINSMORE:
We put it in there.
But the first 3
thing you have to determine is why the component wasn't 4
modeled.
5 DR. APOSTOLAKIS:
Okay.
So whether it was 6
screened out or was just not part of the PRA model, they are 7
two different things, right?
But both of them are telling 8
me something.
The fact that it is not in the PRA, it seems 9
to me, tells you something about the importance of that 10 component.
Because if it was important, it would have been 11 in the PRA.
Because if an important component has been left 12 out of all these PRAs, we have been doing in 20 years, 13 somebody would be screaming bloody murder.
14 MR. DINSMORE:
Well, that's kind of --
()
15 DR. APOSTOLAKIS:
So these are not important 16 components if they are out.
17 MR. DINSMORE:
Well, that comes back to the 18 sentence that you read that says, if you are going to do it 19 with the basic events, they we are going to pay a lot more 20 attention to what is in the PRA.
That's exactly the reason 21 that sentence is in there.
But if you are --
}
22 DR. APOSTOLAKIS:
Which is the whole idea of 23 risk-informed regulation, right?
I mean if you do it at the 24 system function level, then you are not doing risk-informed l
25 regulation, are you?
i O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 j
i
341 1
MR. DINSMORE:
Well, we are trying to grade the (Gj quality assurance controls on a whole bunch of components 2
3
'and the easy -- or the most straightforward way to do it, so 4
that it is well documented and that we can go look at why, 5
is to do it by the system functions.
6 DR. POWERS:
It seems to me, George, there's a 7
tremendous latitude granted the PRA analyst on how he draws 8
his boxes, and what he calls items within a plant.
And that 9
one has to anticipate that there will be diversity in doing 10 that.
That people will say, oh, fair enough, I-have data on 11 failure of diesel generators, the rates.
I don't have so 12 much on oil service pumps to those diesel generators, though 13 I know that contributes.
So rather than mining a tiny 14 little data base that I might have on service pumps type,
()
15 draw a box around the thing that is big, then do the PRA 16 analysis.
17 But the QA guy, he has to worry about each one of 18 the things inside that box, and you have got to accommodate 19 that somehow in establishing your QA program.
20 DR. APOSTOLAKIS:
And my point is that the amount 21 of attention you will pay to the little pieces inside has to 22 rely heavily on how important that piece is, or how 23 important the component is.
24 DR. POWERS:
I don't think he will argue with you 25 on that.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034 2
342 1
DR. APOSTOLAKIS:
Well, but the document doesn't
/
2 say that.
The document goes all the way to the safety 3
function.
See, if I have an importance measure for the 4
diesel, then I think South Texas told us there are, what, a
5 couple of thousand components for each diesel.
6 MR. DINSMORE:
Something like that.
7 DR. APOSTOLAKIS:
3500 or so, I remember.
If I 8
already have three or four diesels in the plant, 9
Fussell-Vesely and risk achievement worth will come up with 10 numbers that will tell me each diesel now has this 11 importance level.
12 MR. DINSMORE:
Right.
13 DR. APOSTOLAKIS:
That should be my starting point 14 for deciding what to do with the components inside this i
15 component, not the fact that the diesel supports the supply 16 of electric power, which is a major function, an extremely 17 important function, and that's what I get from this.
That 18 the diesels are supposed to give you emergency power, that's 19 an extremely important function, and that really is my 20 starting point for saying something about the little 21 component inside each diesel.
Because then the fact that I 22 have one or three diesels becomes almost irrelevant.
23 Or you are placing it back on the licensee.
The 24 burden is on the licensee to convince you that three diesels 25 is better than one.
ANN RILEY & ASSOCIATES, LTD.
i Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 343 1
MR. DINSMORE:
That sounds reasonable.
Are you
(
2 saying'that the diesels should be low safety significant, 3
each diesel in the power plant?
4 DR. APOSTOLAKIS:
What I am saying is these are 5
components.that are in the PRA.
I should get the two 6
measures of importance that the PRA gives me.
Okay.
This 7
should be my starting point.
8.
Now, then I will look at the individual pieces of i
9 the components.
The PRA hasn't done it.
Okay.
Hasn't done l
10 it.
So I don't have importance measures for the 2,000 11 components of each diesel.
12 MR. DINSMORE:
Right.
13 DR. APOSTOLAKIS:
But I already have a boundary 14 value.
That, as a whole, a super-component, this is their-()
15 significance, which is important because it is high.
But 16 then within it, again, I may have little pieces that are not 17 that important.
So what do I do.
Either I use my judgment, 18 or if it is really important, I may even do a 19 mini-probabalistic analysis for the component.
20 MR. DINSMORE:
But your assumption there was that i
21 the diesel would be high based on the component level --
1 22 DR. APOSTOLAKIS:
Well, I guess.
=
23 MR. DINSMORE:
Well, you have four -- six of them.
24 DR. APOSTOLAKIS:
But I would not start at the 25 function level.
I would say that has been taken care of by O
ANN RILEY & ASSOCIATES, LTD.
I Court Reporters 1250 I Street, N.W.,
Suite 300 j
Washington, D.C.
20005 (202) 842-0034 i
i
'344 1
the PRA already.
You'are talking about big things here.
We
()
2 are not going to go back to the accident sequences again and
~
3-dissect them.
That's my point.
4 The use of'the word safety function bothers me 5
because it essentially says, well, the PRA does all these 6
things, but we really want to go back to the safety 7
functions.
Then why did they spent $2 million doing it?
8 See, at that level, you are so high that the PRA i
9 does model these things.
If there are safety functions -- I 10 mean all of them are there.
11 MR. DINSMORE:
Well, I am sorry, sir, but you mean 12 that at the component level, importance measures for diesel 13 is low, then we should put diesels as low significant 14
. components?
I mean that is the logical conclusion.
()
15 DR. APOSTOLAKIS:
But I am not saying -- yeah.
16 MR. DINSMORE:
And I am not sure that we were 17 willing to go that far at this time.
But so what you are 18 talking -- you don't mean the tone, you mean that the idea 19 is incorrect.
20 DR. APOSTOLAKIS:
Well, I said the tone because it 1
21 puts so much emphasis on the system function.
What do you 22 mean by system?
You mean low pressure injection, for 23 example, the whole system?
24 MR. DINSMORE:
You know, we used systems because 25 systems are pretty well defined chunks of the plant.
1 h
ANN RILEY & ASSOCIATES, LTD.
\\~/
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
345 1
DR. APOSTOLAKIS:
In essence, what you are saying
()
2 is I don't believe any of the importance measures you are 3.
producing for me.
4 MR. DINSMORE:
We are saying that the individual 5
components level on -- the individual component measures on 6
individual components is not sufficient to actually reduce-7 QA controls on that component without further evaluation.
8 DR. APOSTOLAKIS:
That is not really what you are 9
saying.
10 MR. DINSMORE:
Well, we define further. evaluation 11 by determining what the system function is and how important 12 that function is and then, based on that, you can look at 13 your individual components and make your decision.
14 DR. APOSTOLAKIS:
The essential element there is
)
15 also how many of these do I have?
16 MR. DINSMORE:
Well, they can do that when they 17 are making their final decision.
18 DR. APOSTOLAKIS:
But you didn't think it was i
19 appropriate to say it because that is against quality 20 assurance?
It is for relaxing it, right?
21 It seems to me that has a very conservative tone 22 for something whose impact on the failure rates is really 23 not understood.
24 CHAIRMAN SEALE:
Timidity rears its head.
25 DR. APOSTOLAKIS:
All right. Let's go on.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
346 1
MR. GRAMM:
The other change that was_made, there A
2
'was a table that had caused some confusion where we had
([
3 tried to articulate how they function with the matrix.
We 4
removed that particular table.
'5 An option was inserted to deal with direct ~
6 manipulation of the uncertainty distribution during the 7
bounding analysis.
8 We elaborated on what types of justification could 9
be used for categorizing items as low safety significant 10 when looking at probability considerations.
11 DR. APOSTOLAKIS:
Where is this now?
12 MR. GRAMM:
Where is it?
13 DR. APOSTOLAKIS:
Yes, which page.
14 MR. GRAMM:
It would -- it is now located on page 15 15.
16 DR. APOSTOLAKIS:
15.
17 MR. GRAMM:
And basically we are trying to 18 distinguish what was the factor that led to the reliability 19 and if it is directly associated with QA, then you would end 20 up with a different result.
21 There was a suggestion made that we should target 22 the QA controls based on CCF factors.
We have looked at the 23 Reg Guide and feel that in general it allows for 24 categorization and assignment of QA controls is appropriate.
25 In several cases we discuss common cause considerations.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
l 347 I
i 1
We didn't address any particular amplification on f
2 that.particular comment.
3 As Steve said earlier, we did expand on the 4
discussion about looking at the reasons why SSCs are not 5
modelled during the traditional review and take that into 6
account during the expert panel considerations --
7 DR. MILLER:
Where?
8 MR. GRAMM:
Page 8?
9 DR. MILLER:
Page 8?
10 MR. GRAMM:
Yes.
11 DR. MILLER:
Where on page'8 is it?
12 MR. GRAMM:
It is at the bottom of the page in 13 Section'4.
There was some redlined text in the new 14 document.
15 MR. DINSMORE:
We essentially added the effluent 16 monitoring.
This was just to give an idea'of what type of 17 stuff was not modeled.
18 DR. APOSTOIAKIS:
Are you saying anywhere that 19 some stuff is not modeled because it was screened out?
20-MR, DINSMORE:
That's on page 11.
I think it just 21 says to determine first why it is not modeled and then 22 guided by this determination you should investigate -- that j
23 determination should guide your traditional engineering 24 investigation.
25 MR. GRAMM:
On the final slide are some comments I
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W, Suite 300 Washington, D.C. 20005 (202) 842-0034 i
348 1
made by our. legal department.
They wanted to clarify
-y--
'2
. language where we use "must and "shall" within the context.
-L 3
of the Reg Guide.
4 We have added clarification w:.mt that means when a 5
licensee voluntarily adopts _the Reg Guide.
6 We have added clarification on 50.54 (a), change 7
controls, what that_ covers, and that is about the essence of 8
the OGC comments.
9.
That concludes our presentation.
10 DR. APOSTOLAKIS:
Any comments?
11 DR. MILLER:
Can we go back to page 8?
The two 12 sentences before Section 4 says for those seeking SSCs that 13 are low safety significant, relaxation in QA control may be 14 proposed.
()
15 Is there a way we can make that stronger, like 16 "should be considered" or --
17 MR. GRAMM:
Certainly, we could add that.
18 DR. MILLER:
That was kind of the -- it was a tone 19 issue -- like, gee, let's propose it, is what I would like 20 to see in there.
21 MR. GRAMM:
That's obviou91y the items that we are 22 focusing in on in terms of relaxing those controls and so we 23 can add that language in.
Yes, sir.
24 DR. MILLER:
Then I had -- maybe this is going the 25 opposite.way.
Let's go above there.
Let's go under Item 4 O
' ANN RILEY & ASSOCIATES, L TD.
Court Reporters 1250 I Street, N.W.,
Suite 300
""""i"S'"e4?: 63
349 1
where you say the licensee should evaluate its-risk studies
()
2 to determine the extent to which quantitative-and 3
qualitative risk insights -- you mean may or will be?
Is 4
that what you mean?
5 MR. GRAMM:
I don't think "may" is the operative 6
word because they may look at it and consider it is not 7
appropriate in the context of graded QA.
8 DR. WALLIS:
Could yoa clarify something for me?
9 When you have public comments, they look to me 10 like comments from the nuclear industry.
11 Is that the case?
12 MR. GRAMM:
In this particular case the comments 13 ran the gamut from interested individuals to owners-group to 14 NEI.
There-was about 40 sets of comments that filled a-()
15
. couple ring binders of material.
16 DR. WALLIS:
So you do get comments from members 17 of the public?
18 MR. GRAMM:
We did have members of the public as 19 well as individual licensees and NEI provided a 20 consolidation of comments from the nuclear industry.
21 DR. WALLIS:
And are those useful comments?
22 MR. GRAMM:
We found constructive comments within 23 that and we have discussed some of the changes we have made 24 in regards to this document.
25 We did not go over some comments made that we O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
i 350 1
1.
chose not to react to.
Some comments were out of scope.
()
2 DR. APOSTOLAKIS:
Anything'else?
Thank you.
3 MR. DINSMORE:
I would just like to!say one thing.
4 These guys are putting 50-70 percent of their I
i 5
components in low --
6 DR. APOSTOLAKIS:
These guys?
7 MR. DINSMORE:
Well, South Texas -- even with the 8
setup that there is now, so it is usable.
You might not 9
think it is the best way to do it, but it is possible to do 10 it and you get a fairly large chunk of stuff that you are 11 putting in low, which everybody is going to agree is low.
12 DR. POWERS:
Let's pursue that comment just a 13 little bit.
14 Are you fairly comfortable that this is -- we have
()
15 talked a lot about going slow around here and learning from 16 these things.
17 Is this going to give you a good education in how 18 well this '";.xs?
19 MR. DINSMORE:
Tremendous education.
20 DR. POWERS:
So you think that it is entirely 21 conceivable at the end of some number of years of using this 22 that you might well want to go back and revise the tone and 23 thrust on this?
24 MR. DINSMORE:
Yes.
25 DR. PC WERS :
But you are going to derive a h
ANN RILEY & ASSOCIATES, LTD.
/
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C.
20005 (202) 842-0034
l
.351 1
database from this pilot experience?
2 MR.-DINSMORE:
Yes.
3.
DR. POWERS:
Now the other point is maybe for the 4
edification of the whole committee.
5.
When you get your public comments, you in fact 6
expose every comment, even though you'might discuss here
{
I
\\
7 only some fraction of those comments?.
8 MR. GRAMM:
That is. correct.
As I understand it, 9
what we will have in the Commission paper forwarding these 10 is a characterization of the comments.
11 In many cases we compared and summarized similar 12 comments and we will have a disposition for each one of 13 those.
14 DR. APOSTOLAKIS:
Why would it not be wise at this 15 time to have in this Guide two sections with clear 16 statements of what you expect the licensee to do?
17
'In one section you are saying dealing with 18 components that are in the PRA.
The licensee comes :ba here 19 with Fussell, Vesely and Rau.
I want you to take these two 20 into account and these additional elements and bring up 21 functions and so on and so on, but your starting point is i
22' the importance measures, and then as I said it is legitimate 23 to say, gee, I really don't want to go blindly the 24 importance measures.
I want to understand more about it --
25 can maybe move some things around.
t i
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
(
352 1
The other section says components that are not in I
(
2 the PRA.
Next question, why aren't they?
Have they been I
3 screened out?
That tells me something about their 4
importance.
Then think about this and this and that when 5
you categorize it.
There are other things that are 6
completely outside the scope of the PRA but we in Quality 7
Assurance have other objectives that go beyond severe 8
accidents, so now we are approaching this approach.
9 That in my mind would have been a very clear 10 statement of where you are coming from.
Now I am confused 11 as to what exactly you want to do.
12 You are starting with safety functions.
You are 13 saying something about importance measures.
Then you are 14 going back to the safety functions.
It is not clear to me O
\\,,/
15 if I were a license what would be expected of me.
16 That is my problem, and then the more fundamental 17 problem is that we are arguing about something whose impact 18 is not understood.
19 MR. GRAMM:
If I could try to answer that --
I 20 DR. APOSTOLAKIS:
Sure.
l 21 MR. GRAMM:
The document I think is trying to say 22 we would use the PSA information where available for 23 screening purposes and then the expert panel comes in with a 24 human element, brings in the engineering and operations 25 perspectives, looks at additional considerations and makes a O
ANN RILEY & ASSOCIATES, LTD.
t\\-
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
353 1
judgment -- is that PSA information sufficient?
Do we have 1
()
2 more?
And then the safety functions need to be understood j
3 to fully develop that rationale for that process.
4 DR. APOSTOLAKIS:
And where is that?
Can you j
5 point me to the page where that is said, because that is 6
exactly what I can't find.
7 That is what I mean by saying that the tone is not 8
right.
If you said look, this is what the PRA says, I know 9
it has problems, I'll go to the expert panel, I'll give them 10 more advice, and they will make a decision, that makes sense 11 to me.
But I have a hard time following all these arguments 12 about safety functions, when they're categorized the system 13 functions. require proper operation and so on.
It's a very 14 different approach.
It doesn't say what you just told me.
()
15 And then I -- a separate section on components that are not 16 there.
Yes, I think that's a legitimate concern.
17 MR. DINSMORE:
We wanted a process which would 18 cover everything.
19 DR. APOSTOLAKIS:
I understand that, and that's 20 why I'm confused.
21 MR. DINSMORE:
And if we learn later on how to 22 split ir op int.elligently during 1.he different applications 23 as the pilots come in, we'll probably improve the process, 24 DR. APOSTOLAKIS:
Well, we do have IPEs for all 25 the plants, so, you know, it wouldn't be so hard to know 4
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 l
i 354 1:
what's left out.
j
[ \\
2 MR. GRAMM:
I think in terms of the overall I
g 3
characterization in section 4.3 for the integrated 4
assessment --
5 DR. APOSTOLAKIS:
Page?
6 MR. GRAMM: -Page number 18 at the top. -Basically 1
7 the panel should evaluate traditional engineering, I
8 probabilistic, qualitative information regarding systems and 9'
system functions within'the defined scope.
And where it's 10 basically saying all of those are input to the panel when 11 they make their final decision on how to rank the particular 12 components.
i 13 DR. APOSTOLAKIS:
And in my opinion, I mean, if 1 14 wanted to be a'little harsh, I would say yes, the panel will
()
15 do a good job here if they have not read the first 17 pages.
16 I mean, I have to give you my frank opinion on 17 this, because the first 17 pages are going to confuse the 18 hell out of them.
.19 I MR. GRAMM:
And I think we're left with a 20 situation where very important equipment such as reactor 21 pressure vessels is not modeled, and we needed to lay out a 22 process where that would be considered.
We have 23 instrumentation, radiation monitors, there is a wide gamut 24 of items.
In fact, the majority are not modeled.
So we put 25 together a process that would use risk informacion but also
/]
ANN RILEY & ASSOCIATES, LTD.
\\-s<
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
355 1
be able to treat other commodities.
2
-DR. APOSTOLAKIS:
I think the problem is what 3
Steve said, that you are trying to cover all of these items 4
at the same time, in, you know, the same discussion.
And 5
that's where the discomfort, my discomfort, comes from, that 6
if there had been separate sections discussing different 7
components, depending on whether they are part of the PRA or i
8 not and so on, maybe life would have been easier for both of 9
us.
10 DR. MILLER:
Could that be changed simply by
)
11 putting something in that context in the introduction?
)
12 DR. APOSTOLAKIS:
I don't think it can be changed 13 by adding a sentence.
I think it can be changed by doing a 14 serious editing job to the document.
()
15 DR. MILLER:
Kind of like a total reorganization 16 of the document.
17~
DR. APOSTOLAKIS:
Yes.
18 MR. DINSMORE:
You would also have to define all 19 of this.
I mean, the way it's set up now, the process kind 20 of deals with everything.
21 DR. APOSTOLAKIS:
At the same time.
And that's 22 where the problem is.
23 MR. DINSMORE:
Same time.
24 DR. APOSTOLAKIS:
In my opinion --
25 DR. MILLER:
But it clearly says it's dealing with
[
ANN RILEY & ASSOCIATES, LTD.
\\w-Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
356 1
the same'-- everything --
2 DR. APOSTOLAKIS:
Hmmm?
3 DR. MILLER:
If it clearly says in the 4
introduction it's dealing with-everything, I think that 5
would clarify it, wouldn't it?.
6 DR. APOSTOLAKIS:
No, because standard components 7
are telling the PRAs that we now have IPEs for all the 8
plants should be treated differently.
I mean, _you should 9
really treat the Fussell, Vesely, and the risk achievement 10 worth measures with a little_more respect, because'a lot of 11 that stuff about safety functions and so on -- that's built 12 into it.
13 DR. MILLER:
Right.
14 DR. APOSTOLAKIS:
Yes.
Now for a component that's 15
-not modeled very well where we have problems with the models 16 or it's left out completely, then you take a different-17 approach.
Then you revert back to being more conservative i
i 18 and taking small steps.
19 DR. MILLER:
You're almost recommending is almost 20 a different reg guide, one for those that have been modeled 21 and one for those that have not been modeled.
22 DR. APOSTOLAKIS:
No, the same reg guide but 23 different chapters.
24 DR. MILLER:
Whichever.
25 DR. APOSTOLAKIS:
When is this due?
(V)
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
357 1
DR. MILLER:
Next week.
()
2 MR. GRAMM:
The target is the end of the month.
3 DR. APOSTOLAKIS:
To the EDO's office?
4 MR. WOODS:
If it's going to get to the EDO's 5
office at the end of the month, it has to go into 6
concurrence process about the lith of March.
So we're 7
talking about a week and a half.
8 DR. APOSTOLAKIS:
Well, I guess -- any other 9
discussion?
10 DR. WALLIS:
Well, again, I'm the new member on 11 this committee, I guess it is.
12 DR. APOSTOLAKIS:
Yes.
13 DR. WALLIS:
Maybe this is just whatever this is 14 here, regulatory guidees.
But it would seem to me that
(~ss t
(_)
15 reading this -- there's an awful lot of words here which are a
16 all subject to interpretation.
So one could quibble about 17 hundreds of sentences in here if one wished to.
And yet the 1
18 other extreme would be to have some kind of a checklist o;.
19 multiple-choice thing that the license just goes through 20 where it's quite clear what action has to be performed.
21 Is it really very complicated for some licensee to 22 figure out what all these words mean and what then has to be 23 done in response to all these words?
Can't some kind of a 24 routine procedure which is simpler be worked out where you
~
25 do this and then you do that and you do something else and l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
358 1
it's. clear what you're doing?
Or is this just the culture
[)
2 of the NRC that is being reflected in this sort of document?
N--
3 MR. DINSMORE:
I would say after several 4
submittals once we really understand all the implications of i
5 everything we could develop that, but we've only had one 6
submittal, and we need several more.
7 MR. WOODS:
If you do that, you have in mind a j
8 particular application, and what process you write down with 9
your checklist would fit that application very well, and the 10 trouble is someone comes in with quite a different 11 application and you've found that that specific list that i
12 you've made isn't appropriate.
So you have to write more 13 general guidance.
And that's how it got like it is.
This 14 wouldn't fit if you did that.
I'll guarantee you it
]
(3 15 wouldn't fit all the different cases you'd want to apply it s_j 16 to.
l 17 DR. WALLIS:
I just wonder if there isn't 18 something that's sort of between the two extremes which is i
19 easier to administer.
If I write this sort of stuff for how 20 I'm going to grade students on a report or something, it 21 would be absolutely useless, because neither they nor I 22 would understand what's actually required.
23 DR. POWERS:
I think, Graham, that the difficulty 24 that's confronted, if they were to write a reg guide for 25 grading students on an exam, I think it would be a very
(
ANN RILEY & ASSOCIATES, LTD.
A-Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
i 359 l
1 terse document, and it could approach a checklist.
I think 1
()
2 if on the other hand they were to write a reg guide for your 3
3 student's career, that might be a more involuted and 1
4 complicated document, and have a lot more sections and 5
subsections.
And I think that's the analogy they're drawing l
6 here, that they're writing for a rather elaborate plant, and 7
if they had to write one for diesel generators, it might be 8
a good deal more terse.
9 Is that a fair characterization?
10 MR. DINSMORE:
I think it's excellent.
11 DR. WALLIS:
Well, isn't it really that if the 12
-measures were clearer, then the document could be more i
13 terse?
14 MR. WOODS:
Yes.
That's true.
O l( )
15 DR. POWERS:
Clearly one of the handicaps that in 16 all fairness plagues the application of risk-informed j
17 techniques here is that we do not have a quantitative 18 understanding of how quality assurance and quality control 19 measures affect risk.
If we had that, I think these 20 gentlemen's job would be much, much easier.
21 DR. WALLIS:
Is there hope of moving in that 22 direction?
23 MR. GRAMM:
I think as we see the pilot of the 24 volunteer plants evolve and gather some information at the 25 site we will have a feedback loop.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
360 1
DR. APOSTOLAKIS:
Well, I thought the Office of 2
Research was going to start a project sometime to tell us
-3
-what reliability -- for example, are the quality assurance 4
regulations -- well, they don't have regulations, but, I 5
mean, requirements for oth r industries as stringent as 6-ours?
7 MR. GRAMM:
I think you'll see a wide variation, 8
but industries involved with complex activities, be it 9
manufacturing or design, have a QA program.
It can range 10 from an ISO-9000-type program to other commercial types of 11 quality programs, but they have to have the basic 12 ingredients that are described in the criterion of the 13 appendix.
14 DR. APOSTOLAKIS:
But it would be interesting, for 15 example, to compare failure rate of major -- failure rates 16 of major equipment that are common, 17 MR. GRAMM: -It would be interesting.
Throughout 18 this process we have asked if anyone has that data.
It has 19 not been forthcoming.
There have been searches made by the 20 licensees, and unfortunately the data bases don't 21 discriminate.
It's all lumped together.
I 22 DR. APOSTOLAKIS:
I think you're right, but I'm 1
23 talking now about the project in the future that might try 24 to do that in a more serious way and see whether the j
25 differences justify the effort.
Because you have to also b
ANN RILEY & ASSOCIATES, LTD.
O Court Reporters 1250 I Street, N.W.,
Suite 300 I
2 84 b3 i
361 1
specify the difference in effort.
But we're hopelessly
()
2 behind schedule here, so --
3 DR. MILLER:
One comment Dr. Wallis raised is 4
grading of student reports.
I always require in a student 5
report a table of contents and a list of definitions up 6
front.
I always wonder why we don't do that in reg guides.
7 I think this would be much simpler for me to understand if 8
the thing had a table of contents and a list of definitions 9
up front, which I suggested I thought some time ago, but 10 obviously none of the reg guides, whether it be these or any 11 of the other ones, do that.
Basic reporting requirements 12 you do that.
I don't understand why we don't.
That's a 13 broader issue.
I still think it would be much simpler to 14 use if you had those simple things.
I agree with Dr. Wallis
()
15 coming in cold this looks like an overwhelming complicated q
16 document.
I can only say it's quantum leaps from what we i
17 saw a year ago or so.
18 DR. POWERS:
For someone who spent altogether too i
19 much of his time reviewing QA plans for at least DOE nuclear 20 facilities, I will compliment the staff for a very clear 21 document.
22 (Laughter.]
23 And certainly concise.
24 DR. APOSTOLAKIS:
Which confirms once again that 25 clarity is in the eyes of the beholder.
O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
362 1
(Laughter.]
2 DR. APOSTOLAKIS:
Thank you very much.
3 I guess the next session is mine, too, which comes 4
as a surprise.
5 CHAIRMAN SEALE:
Yes, that's correct, and we have 6
I think some guests here that want to hear us, so we better 7
get started on it.
8 DR. APOSTOLAKIS:
Okay.
So we received a white 9
paper, I suppose it is -- first of all, are there any staff 10 members that will sit up here, or is it us only?
Is this a 11 discussion with --
)
12 CHAIRMAN SEALE:
This is a discussion.
13 DR. APOSTOLAKIS:
Just us?
I 14 Oh, you want to make an introduction?
15 A draft of a white paper from the Commission with
{
16 the title Risk-Informed Performance-Based Regulation which 17 defines or discusses some of the terms that people are using i
18 these days freely and perhaps do not mean the same thing to 19 everyone, things such as risk assessment, risk-based, 20 risk-informed, performance-based and so on.
I think -- I've 21 read it and I think it's a pretty good document.
It does 22 give a clear idea as to what these terms mean, especially 23 towards the end.
I mean, Item 4, risk-based, the Commission 24 has never endorsed risk-based.
Risk-informed makes more 25 sense.
Performance-based and all that.
I think this is ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
363 1
pretty good.
()
2 My only comment is that I would change this 3_
traditional-approach tone again
---I seem to have a problem 4
with tones -- and in the preamble say something that --
5 something like what we said in our letter to the Commission 6
dated December 16, that uncertainty is a major issue in 7
regulating nuclear power, and in the early days the 8
technology for quantifying that uncertainty was not 9
available, so that so-called traditional approach evolved 10 with safety margins, defense in depth, and so on, as it is 11 correctly. stated here it is to some degree or at some level
-12 risk informed. 'But basically they were trying to deal with 1
13 the uncertainty in our state of knowledge, okay, that we 14 really don't know very well what may go wrong, okay?
())
(~
15 DR. APOSTOLAKIS:
Then PRA comes along and U
i 16 starts answering that question.
Again,-it is not just the i
17 probability part, by putting together things in the form of 18 sequences of events -- the integration in other words -- so 19 now the way we are handling uncertainty becomes different.
i 20 It has an element, a dimension that is handled by 21 what is known as PRA but there are still other elements that 22 are outside the PRA for some reason, okay?
23 We feel that they both have problems in some areas 24 and we are trying to create a new system that combines the 25 best elements of each. okay, and this is the new era of
()
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 364 1
1
' risk-informed regulation.
2 So it is not really that I am saying something
(
3 fundamentally different from what is in the document.
It's 4
just the way it is presented, I think.
5 DR. KRESS:
I agree with you, George, with one 6
'possible exception.
What you are saying'there is a why --
7 it's not a what.
What is the traditional approach?
And I 8
think we are trying to get a definition in here and I would I.
9 say that this falls just a little bit short of a true 10 definition of what a traditional approach is, and if you 11 want to put why the traditional approach in there I wouldn't l
12 care there either, because that gives one some information 13 that is useful, so I would say it's there because of unknown 14 uncertainties and we need to be_ conservative, but that tells 15 the why.
16 Then I would say, well, what it is -- how did we 17 account for these uncertainties?
18 DR. APOSTOLAKIS:
Right.
19 DR. KRESS:
We did it this way.
20 DR. APOSTOLAKIS:
Exactly.
21 DR. KRESS:
And that is what I would say.
22 DR. APOSTOLAKIS:
Exactly -- and that is what --
23 maybe I didn't phrase it very well, but that is exactly why
)
i l
24 I want that there, plus I mean if you ask someone a straight 25 question, is PRA really that different from the existing ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 1
365 1
approach? -- not really.
I mean we're doing certain things
()
2 different and so on.
3 Yet many, many times people come in here and talk l
4 as if PRA would replace the way they are doing business, 5
which is utterly false.
You never replace.
You ndd.
You i
1 6
modify.
You take a few thingt, out.
7 The reason why I would like this central theme of 8
uncertainty is to say look, you are still dealing, you have 9
been dealing with the san i thing from day one --
10 uncertainty.
11 As our state of knowledge evolves, we are 12 beginning to do certain things differently but fundamentally 13 we are addressing the same problem and a clear statement 14 somewhere here that neither system is perfect I think would
)
15 go a long way towards helping people understand what is 16 going on.
17 See, very often they new kid, PRA, is scrutinized 18 to death as if the existing system were perfect.
19 DR. POWERS:
On the other hand, one has to concede 20 that the existing system has had a reasonable amount of 21 success.
We haven't been destroying reactors all over, left 22 and right, so when one comes in and proposes to make a 23 change for a system that appears on the face of it to have a 24 reasonable amount of success I think it is probably right, 25 probably prudent to give it close scrutiny especially in the L
1 l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters
)
1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 l
(202) 842-0034 l
l
366 1
areas.where it seems. adventurous: to apply.
.m(
i-
-2 DR.~APOSTOLAKIS:
And I fully agree with that, V
3 except for your basic premise.
i
'4 (Laughter.]
5 DR.'APOSTOLAKIS:
I do agree with the. statement 6
'that you should scrutinize the new thing, but you should 7
scrutinize'it in the proper context and the context is not j
8-right.
That is my argument.
9 The context is that the current system has served 10 the industry and the Commission well.
Well, I take 11 exception.to that.
12 Maybe from the safety perspective, although even 13 that'is also doubtful, but it's bringing the industry to its 14 knees, so success again -- you know, f)
_ what criteria do you
(,/
15 use?
11 6 Now it is.none of our business --
17 DR. POWERS:
Let me ask about this --
18 DR. APOSTOLAKIS:
It is none of our business to 19 worry about economic well-being of the industry, I guess, 20 but I am not sure that that is really -- I saw somewhere in 21 fact that the Atomic Energy -- where did I see that?
I 22 think --
-23 DR. POWERS:
You read the Atomic Energy Act, which 24 says that, yes, it is not an unbridled examination of 25-safety --
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
367 l'
-DR. APOSTOLAKIS:
Right.
(
2-DR. POWERS:
-- that a certain amount of 3
encouragement to the development of nuclear energy is built 4
into the Atomic Energy Act.
That is a fair statement.
5 Now an the empirical evidence, saying that it is 6
bringing the irdustry to its knees strikes me as a tad 7
strong --
8 DR. APOSTOLAKIS:
It is --
9 DR. POWERS:
-- when I see that nuclear power has 10 set records for its contribution to the nation's energy mix.
11 DR. APOSTOLAKIS:
Well, I took an extreme 12 position --
13 DR. POWERS:
Yeah, you took a real extreme 14 position there and if you are going to critique that way, 15 then we will go througn a point by point on this.
16 DR. APOSTOLAKIS:
Well, let me then come b1:k to 17 your argument and say that when we say that the existing 18 system has served us well, we should qualify that, because I 19 don't know what "well" means, because if that were the case, 20 I don't think this agency would be spending all these i
21 resources it's expending now to move to this new system, so 22 something is wrong someplace.
23 DR. MILLER:
No --
24 DR. POWERS:
What I think -- the contextual 25 argument that you make is one that you have made previously
()
ANN RILE'l & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
368 1
and that's been written up into an ACRS letter to the
()
2 Comn.ission in which we suggested that the proper context is 3
to see does this new methodology add value'and focus and it 4
strikes me that that is the point that you want to make is 5
that we knew more than we knew when the nuclear enterprise 6
started.
7 When the nuclear enterprise started there were 8
real uncertainties.
There was fear that the core damage 9
frequency might be as high as 10 to the minus 6.
I am 10 literally quoting from one of the original siting studies on 11 a DOE site, but there were big uncertainties about the 12 magnitude of that.
13 There were no industrial standards that would 14 guide how one would set up a safe nuclear industry and a
()
15 shotgun approach was taken -- that is, I don't know what is 16 most important so I will assume everything is important.
17 We now have 3,000 reactor years of operational 18 experience'with big systems.
We have reasonable databases.
19 Suddenly we have the capacity to focus and now what you want 20 to have.is a systematic process for focus.
21 You want to have a systematic process for
-22 gathering additional data and that is what your j
23 probabilistic risk methods are giving you is a systematic 24 way to focus and if the licensees derive economic benefit 25 out of that, that's good.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
1 i
369 1
If society derives safety benefit of it, that,
()
2
'too, is good, and it seems that both are in the offing.
3 DR. KRESS:
George, I think --
f.
4 DR. APOSTOLAKIS:
I don't disagree with anything l
l 5.
you have said.
l 6
DR. KRESS:
I think the Commission has already 7
expressed its desire to move into E more risk-informed type 8
of regulatory approach and I don't think we should be i
9 debating the reasons for that or whether it is good or bad 10 or why.'
11 I think we are talking about given that as a 12 direction we are going, let's be sure we understand what all 13 these things mean.
14 DR. APOSTOLAKIS:
Yes.
()
15
.DR.
KRESS:
So I don't think this particular white 16 paper is the place to go into the reasons we are changing 17 and the reasons we are going this way.
This particular 18 paper is to clarify what it is that we are talking about and 19 I think that she -- in Item 2 in her enumerated things, she 20 gives a pretty good description of what is wrong with the 21 present approach and why we shouldn't 22 DR. BARTON:
The present system treats everything 23 at the same level.
24 DR. APOSTOLAKIS:
Don't misunderstand what I am 25 saying.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
370 1
I am not asking for a major change here.
What I
/-(j 2
am saying is --
3 DR. KRESS:
For a few words.
4 DR. APOSTOLAKIS:
Yes.
A few words, but let me 5
come back to your point, Tom.
6 We read on page 1, second paragraph, "To 7
understand and apply the commitment expressed in this policy 8
statement it is important that the NRC, the regulated 9
community, and the public at large have a common 10 understanding of the terms and concepts involved, an 11 awareness of how these concepts are to be applied - " and I 12 would say, what I am saying is that there should also be an 13 awareness of how these concepts evolved.
14 I mean if our objective with the statement is to
/
15 bring the regulators, the regulated and the public at large 16 to a common understanding, a few words as to what were the 17 concerns and how -- exactly what Dana said:
In the early 18 days this was our state of knowledge, this is what we were 19 doing.
20 I think that would go a long way towards bringing 21 everybody to the same understanding.
22 DR. KRESS:
Well, what du you think about Item 2 23 and maybe moving its location, because I think it is 24 misplaced where it is and working that into the introduction 25 a little more?
O ANN RILEY & ASSOCIATES, LTD.
~
Court Reporters 1250 I Street, N. W,,, Suite 300 Washington, D.C.
20005 (202) 842-0034
371 1
DR.-APOSTOLAKIS:
I haven't thought about it but
()
2 that sounds --
3 DR. KRESS:
The reason it is there of course is i
4 because it follows on with you tell what the traditional 5
approach is and you say, well, why not stick with it?
- Well, 6
these are the problems.
That is why it is where it is.
)
7 Maybe'it is misplaced.
I didn't think it should 8
be enumerated because everything in here is a definition.
9 DR. APOSTOLAKIS:
It is a definition.
You're 10 right.
11 DR. KRESS:
Except for that and I think it is 12 misplaced a little bit and maybe you can use that paragraph 13 to do what you are talking about.
14 DR. APOSTOLAKIS:
I think that makes perfect
()
15 sense.
16 In other words, you have an introductory paragraph 17 that gives an overall view and I think have -- Dana said --
18 we discussed a lot of that stuff when we were preparing that i
19 letter in December, and then part of the current Item 2 can 20 be the introduction and part of it can be at the end of Item 21 1.
22 In other words, you say what the traditional 23 approach is, and then you~ finish by giving some of --
24-DR. KRESS:
-- its problems.
25 DR. APOSTOLAKIS:
the problems with it, and i
O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D C. 20005 (202) 842-0034
372 1
then you go to risk assessment, and perhaps -- do we get 2
many problems?
3 DR. POWERS:
Let me ask you this, George.
4 DR. APOSTOLAKIS:
Yes?
5 DR. POWERS:
The traditional approach in the sense 6-of it being comprehensive and deterministic is not without 7
its benefits.
8 The risk-informed approach is not without its 9
costs.
10 Isn't there room for balance in your description 11 of the traditional approach and if you are going to point 12 out its problems to point out its advantages?
13 DR. APOSTOuAKIS:
Yes.
14 DR. POWERS:
There is one that looms large in my 15 mind is that if you have to design a system where it's i
16 pretty easy to set your design requirements, whereas in a j
17 more probabilistic approach, especially one combined with 18 something very nebulous called integrated evaluation, it is 19 very difficult to set design requirements.
20 DR. APOSTOLAKIS:
No, I think you are right, and 21 one way of doing it is to have the traditional approach, say 22 how it evolved, then end that item with some of the 23 problems.
24 Then you go to the risk assessment, how it evolved 25 and how it addresses some of the problems of the traditional ANN RILEY & ASSOCIATES, LTD.
Court Reporters l'50 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
373
,1 approach, and then end with the problems of the risk
()
2 approach, and then say ah, but now'the one that we'have been 3
accusing all along can come to the rescue here.
4 So you are actually doing what Dana wants in one 5
particular way, or you --
6 DR. KRESS:
I think you guys are making something 7
entirely different out of this paper than what it is 8'
intended to be, frankly, by doing that, but I think that may L
9 be good stuff and worthwhile thinking about.
10 I think this is probably not the right pan'.
.o 11 put that into.
12 DR.'APOSTOLAKIS:
The only thing that makes me 13 want to do it is this sentence that the Commission would 14 like the NRC, the regulated community and the public at
()
15 large to have a common understanding of the terms and 16 concepts -- second paragraph of the first page.
l 17 DR. KRESS:
Well, that is the thing that makes me
.le not want to put that in there, because you are confusing the 19 issue when you put all this evolution and reasons for having 20 it, and then what this is is a -- to me is a pure, out and 21 out definition of terms for clarify, and you know, those l
'22 things ought to be terse, short, to the point and get right 23_
to it, and let people understand what we are dealing with.
24 All these other things you can talk.about in 25 another paper.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C. 20005 (202) 842-0034
1 374 1
DR. MILLER:
So you want to minimize any
()
2 description of process?
3 DR. KRESS:
I want to make this clear and I want 4
to get extraneous information out of it and --
5 DR. MILLER:
For example, one thing I would like 6
to see somewhere, maybe here or someplace else is the 7
process of doing a PRA.
8 The process of doing a PRA is very important to 9
our better understanding of the system and the fact is the 10 result of the PRA may be not as important as the process.
11 DR. POWERS:
I don't think that belongs in this 12 document.
13 DR. MILLER:
It just hit me when I read all this, 14 I would like to have that in there.
(~ \\
\\
')
15 I think George said that early-on when he talked j
s 16 about the PRA.
The most important thing is the process.
17 DR. APOSTOLAKIS:
Just don't say that the numbers 18 don't count.
19 DR. MILLER:
I didn't say that.
I said the 20 process may be as important as the --
21 DR. APOSTOLAKIS:
The process is extremely 22 important but the numbers do count as well.
23 DR. POWERS:
It we accede to Dr. Kress's argument, 24 which I think has merit --
25 DR. MILLER:
I think it does.
['~)'t ANN RILEY & ASSOCIATES, LTD.
\\_
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
375 1
DR. POWERS:
-- and return to some of the
.r I
2 definitions, I find definitions in here that -- one in
(
3 particularly just is a fingernail on a blackboard to me.
4 DR. KRESS:
Right.
I agree with you.
1 5
DR. POWERS:
Other expressions of risk include 6
core damage frequency and large early release frequency.
7 DR. KRESS:
Yeah, that's one that bugs me.
And I 8
was suggesting in my -- George, did you get my comments I 9
wrote down?
10 DR. APOSTOLAKIS:
On this?
-11 DR. KRESS:
Yeah.
j i
12 DR. APOSTOLAKIS:
Yes.
13 DR. KRESS:
I'm not -- with written comments that 14 you may not have seen.
()
15 DR. POWERS:
I would tend to write that sentence 16 as other commonly misused terms to express risk in an 17 acceptable way or --
18 DR. KRESS:
No, well, I suggested -- I suggested 19 this be handled in a footnote.
20 DR. APOSTOLAKIS:
What are we being asked to do?
21 CHAIRMAN SEALE:
We have been asked to comment on i
22 this document.
23 DR. APOSTOLAKIS:
What does that mean?
We like 12 4 it, we don't like it, or here is a way to improve it?
25 CHAIRMAN SEALE:
No, I think we have been asked to O
ANN RILEY & ASSOCIATES, LTD Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
376 1
make --
()
2 DR. KRESS:
She would like to know if it is --
3 DR. MILLER:
Critique it.
4 CHAIRMAN SEALE:
Yes.
Substantial comments on 5
content.
6 DR. POWERS:
Certainly, we have as'ed the k
7 Commissioner to comment on documents that we have written 8
and she has graciously done so, and --
9 DR. APOSTOLAKIS:
Yeah, but are we offering 10 constructive suggestions?
Or are we just saying we don't 11 like this?
12 DR. KRESS:
No, I think constructive suggestions.
13 DR. POWERS:
No, I think she is looking for 14 constructive suggestions.
15 DR. KRESS:
If we don't like it, she would accept 16 that comment.
17 DR. MILLER:
If we substantiate it.
18 DR. APOSTOLAKIS:
I am still bothered by -- I mean 19 the intent is to bring everybody to a common understanding.
)
20-I think we can --
21 CHAIRMAN SEALE:
In terms of the definitions.
22 DR. APOSTOLAKIS:
We can help tremendously by 23 adding just a few sentences here and there, placing the 24 whole thing in context.
I am not talking about a major 25 treatise.
(
ANN RILEY & ASSOCIATES, LTD.
I Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
)
377 1
DR. POWERS:
Well, if I go on and look at just the
()
2 definitions, there is another I think is unfair and that is 3
the risk-based where it says relies exclusively, with 4
exclusivelv underlined.
5 DR. MILLER:
That's wrong.
6 DR. KRESS:
Well, that's -- yeah.
7 DR. POWERS:
I think that is a strict definition 8
there, but it lacks the generosity that you accord other 9
systems where there is some mixing and additions and 10 deletions.
11 What I think, the failing that makes one cautious 12 about risk-based types of regulatory decision making is the 13 cost of calculating the risk becomes so excessive, so 14 difficult that you are reluctant to make decisions.
15 CHAIRMAN SEALE:
And the risk of incompleteness.
16 DR. POWERS:
Well, there are all those failings of 17 the risk methodology.
That, really, what you would rather 18
-- what the Commission, I think, in opting for a much more 19 risk-informed kind of approach was saying, gee, the PRAs and 20 those kinds of analyses are so useful to me, not just 21 because of the numbern ad whatnot, but because they give me 22 an idea of what areas I ought to work most diligently in and 23 what areas that I can be -- I can let s1.:w. because, 24 clearly, there are so many things that I san't treat 25 everything.
(
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
378 1
And the risk information is, even when the numbers
()
2 are imprecise and subject to uncertainty, it is very 3
valuable.
And I couldn't do that if I became too enamored 4
with this risk methodology that I said my rules were 5
risk-based.
6 CHAIRMAN SEALE:
I think that definition is 7
perhaps defensive and it overly -- it deletes from the 8
completeness of the -- in its present form.
It just deletes 9
from the completeness.
10 DR. KRESS:
I had a comment.
My comment on 14, 11 it's similar to Dana's, but what I -- the problem I had with 12 that is it is possible attribute she lists there, it is not 13 a definition.
14 DR. APOSTOLAKIS:
Now, let me come to the Item 5, 15 risk-informed.
And why you need this background, I 16 mentioned' earlier.
This is not really accurate.
A 17 risk-informed approach to regulatory decision making 18 represents a philosophy to be used in all regulatory matters j
19 whereby risk insights are considered together with other 20 factors.
I 21 Well, I submit that this has been done since 1955.
1 22 DR. POWERS:
Yes, and there are many, many people 1
23 that say absolutely, we have always been risk-informed.
24 DR. MILLER:
We just have better tools now.
25 DR. POWERS:
And, in fact, the difference now, and ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
379 1
_maybe one to make clear, worth making clear, is when we say
()
2 risk-informed now or risk-based, or use the work risk at 3
all, we are really saying quantitative risk.
4 DR. APOSTOLAKIS:
Exactly.
Exactly.
5 DR. KRESS:
Well, you have to keep in mind she 6
defined risk insights.
7 DR. APOSTOLAKIS:
First of all, is this from the 8
Chairman or from the Commission?
9 DR. KRESS:
It is from the Chairman, it has got 10 her name.
11 DR. APOSTOLAKIS:
Oh, from the Chairman.
12 DR. KRESS:
But she defined risk insights in a 13 previous paragraph.
And when you --
14 DR. APOSTOLAKIS:
Where?
Where?
)
15 DR. KRESS:
Paragraph 3.
16 DR. APOSTOLAKIS:
Oh, the risk insights yield by 17 these risk assessment techniques.
18 DR. KRESS:
She defines what she means by risk 19 insights.
And if you take that definition, I think that 20 makes.--
21 DR. APOSTOLAKIS:
Well, I am not sure.
22 DR. KRESS:
Well, I think --
23 DR. APOSTOLAKIS:
That the word number or quantify 24 is nowhere.
A risk assessment is a systematic method for 25 assessing system performance.
4 O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
380 1
DR. KRESS:
She talks about changes in CDF, LERF,
()
2 the containment failure probability.
Those are quantified 3
numbers.
4 DR. POWERS:
Well, let me ask you this question, 5
George.
6 DR APOSTOLAKIS:
Well, that is the only place, 7
yeah.
You're right.
Yeah.
8 DR. POWERS:
If one has this measure of risk that, 9
presumably, you and I come in agreement with on what risk 10 means, is PRA the only way to get there?
11 DR. APOSTOLAKIS:
To get those numbers?
12 DR. POWERS:
Yeah.
13 DR. APOSTOLAKIS:
Yeah.
14 DR. POWERS:
It is.
)
15 DR. APOSTOLAKIS:
Yeah.
Again, let me come to the 16 philosophy issue here.
What I am trying to say.
There are 17 two points which I think can be said in two sentences.
18 Perhaps Ron will have a semicolon, so it will be a long 19 sentence.
l 20 DR. MILLER:
A whole, an entire page.
21 DR. POWERS:
And the esteemed Chairman --
l 22 DR. APOSTOLAKIS:
I know you don't like 23 semicolons, but 24 DR. POWERS:
A former Chairman of this Commission, 25 or this Committee established an executive order that has i
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
381 1
yet to~be rescinded, and it said, to the effect that there 2
will be no semicolons.
3
[ Laughter.)
4 DR. APOSTOLAKIS:
We can try it with a new 5
administration.
6 I am not talking -- again, I am not talking about i
7 a major treatise.
I want a sentence to support this common 8
understanding that says, look, the issue here has been 9
uncertainty from day one.
Okay.
We will find fancy words, 10 but, esst tially, that is the message.
11 The question is you -- well, the question.
It is 12 a statement of fact that you are handling uncertainty based 13 on your current state of knowledge.
For a long time, up 14 until the mid-70s, we had not quantified parts of that 15 uncertainty.
And what evolved was what is termed here 16 traditional approach based on defense in-depth and so on.
17 What risk insights -- unquantified risk insights 18 were indeed used.
Now, this new technology comes along.
19 And now we can quantify good chunks of this risk.
20 DR. MILLER:
And reduce our uncertainty.
21 DR. APOSTOLAKIS:
Yeah, and eGtablish programs, 22 whatever, you know.
j 23 DR. MILLER:
Can you put that -- you have already 24 used up one page.
j 25 DR. APOSTOLAKIS:
So this -- I am not done, you
/"'T -
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
I 382 1-are interrupting.
w (g) 2 DR.' MILLER:
You already used up'one page.
Can 3-you put that in one --
4 DR. BARTON:
It's only a paragraph, that's all.
5 DR. APOSTOLAKIS:
The art of it is to reduce it to 6
a sentence, so that's why I have -- so now this new 7
technology comes along.
You are quantifying part of it, and 8
being risk-informed.now means being quantified I
9 risk-informed.
Okay.
And what is required is a blending of 10:
the two, because we don't quantify all the uncertainties.
11 And I think that is a clear message, and you still 12 need these definitions, by the way.
13 DR. KRESS:
Actually, you can get that meaning out 14 of the definitions.
15 DR. APOSTOLAKIS:
But I would like to point them z
16 out.
I don't want this to be fully informed -- informative 17 to Tom Kress.
Because Tom Kress doesn't need to be 18 informed.
19 DR. KRESS:
Right.
But if you are talking about 20 just, maybe an extension of the introductory paragraph by a 21 couple of sentences.
22 DR. APOSTOLAKIS:
Yes, that's waat I mean.
23 DR. KRESS:
I think that's all right, because that 24 doesn't obfuscate things too badly.
So that might be worth 25 ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
~-
383 1
[ Laughter.]
()
2 DR. APOSTOLAKIS:
I was never arguing for a major 3
treatise here.
A few sentences.
4 DR. KRESS:
That might not be too bad.
A sentence 5
of two.
6 DR. APOSTOLAKIS:
Yeah.
7 DR. KRESS:
Why don't you --
8 DR. APOSTOLAKIS:
And then adopt the rest of the 9
document to what went before.
Right.
I mean you want --
10 CHAIRMAN SEALE:
There's some modifications there, 11 too.
12 DR. APOSTOLAKIS:
Yeah, modifications.
13 DR. BARTON:
I think we ought to give George a 14 shot at a couple of sentences and see --
(-
15 DR. KRESS:
Do you want to craft up a sentence 16 that says --
17 DR. APOSTOLAKIS:
I have two letters that I have 18 to write.
Well, when is this due?
Do we need to say 19 something this time?
20 CHAIRMAN SEALE:
John.
21 DR. BARTON:
Hopefully, this week.
22 DR. KRESS:
Yeah, we ought to do something this 23 week.
24 DR. POWERS:
I will point out that the author went 25 out of her way --
[/
)
ANN RILEY & ASSOCIATES, LTD.
Court Reporters I
A-1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 i
~
384
-1 CHAIRMAN SEALE:
Yes.
()
2 DR~ POWERS:
-- to get prompt comments to us on 3
our letter to Congress.
I-think we owe her a similar-4 courtesy.
~5 DR. KRESS:
I do, too.
6 DR. APOSTOLAKIS:
Besides, we are working for her.
7 DR. LARKINS:
You have got till noon, Saturday, 8
George.
9 CHAIRMAN SEALE:
Gee, George, you go right to the 10 nexus, don't you?
11 HDR. APOSTOLAKIS:
Well, one sentence.
Well, is 12~
the Committee then charging me with the task of giving a 13 proposal?
14 CHAIRMAN SEALE:
Well, I would make -- I would
()
15' like to divide the task if I may.
Tom has made some 1C comments in what I would say the body of some of the 17 definitions.that follow, which I think are appropriate and 18 helpful.
19 So what I'would ask is that George would take a 20 look at the introductory part of it to incorporate the 21 comments that he has suggested,-and that we ask Tom to give 22 us a cleaned up or a modified version of the definitions and 23 so on that you have, integrating your comments here.
We 24 merge those together then, Mike, and have a document that we 25-can take a look at.
O' ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 1
(202) 842-0034 i
385 1
DR. KRESS:
I would certainly like to have
( ).
2 George's look at my changes to the definitions --
3 CHAIRMAN SEALE:
Sure.
4 DR. APOSTOLAKIS:
Why don't we say the following, 5
Tom and I will do it.
6 CHAIRMAN SEALE:
Well, fine.
Fine.
'7 DR. APOSTOLAKIS:
And we are adults, we can split 8
among ourselves.
9
.DR. KRESS:
We can work together.
10 CHAIRMAN SEALE:
And I have suggested a division 11 of labor initially.
All right.
'12 DR. APOSTOLAKIS:
Okay.
13 CHAIRMAN SEALE:
Okay.
John, are there any other 14 insights you can give us that might be helpfu' as we --
()
_15 DR. LARKINS:
No, I assume at some point later we 16-will probably just go through it paragraph by paragraph --
17 CHAIRMAN SEALE:
Okay.
Fine.
18 Jim,' do you believe that this would be responsive, 19 the approach that we have outlined here would be responsive
'20 to the Chairman's request?
21 MR. JOHNSON:
Yes, I think it would.
22 DR. APOSTOLAKIS:
Jim, can you come to the 23 microphone?
24' CHAIRMAN SEALE:
We would like to have that on the l
25 record-.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 J
386 1
MR. JOHNSON:
I don't know if I want this on the
().
2 record.
But this is Jim Johnson from the Chairman's office 3
and I-think it would be responsive to the Chairman's 4
request.
5 DR. APOSTOLAKIS:
Do you know what else would 6
really help, Jim.
Can we get an electronic version of this?
7 MR. JOHNSON:
Yes, you can get an electronic 8
version.
9 DR. APOSTOLAKIS:
Okay.
Send it soon?
10 MR. JOHNSON:
We will send it to Dr. Larkins 11 shortly after lunch.
12 DR. APOSTOLAKIS:
Okay.
Great.
13 CHAIRMAN SEALE:
Fine.
14 DR. APOSTOLAKIS:
Appreciate that.
Thank you.
)
15 CHAIRMAN SEALE:
All right.
Well, I guess at this 16 point, there are no other comments.
Unless other members of 17 the Committee would like to make a comment?
18 DR. POWERS:
Sounds good.
19 CHAIRMAN SEALE:
Sounds good.
20 DR. POWERS:
Good plan.
21 CHAIRMAN SEALE:
Okay.
Well, we will recess for 22 lunch for one hour.
I would remind Tom and Dana that we 23 have a meeting with John.
24
[Whereupon, at 12:17 p.m.,
the meeting was 25 recessed, to reconvene at 1:17 p.m.,
this same day.]
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
387 1
AFTERNOON SESSION
(
):
2
[1:20 p.m.]
3 CHAIRMAN SEALE:
I apologize for the lateness of 4
the P&P Committee in getting back, but about two-thirds of 5
us are here now.
Our session this afternoon starts with a l
6 report of-the Subcommittee on Reliability and Probabilistic 7
Risk Assessment.
8' George,'you're still swimming in the barrel.
9 You have 30 minutes.
i 10 DR. APOSTOLAKIS:
Next time I should be here from 11 day 1.
12 CHAIRMAN SEALE:
Yes.
13 DR. APOSTOLAKIS:
Well, the subcommittee met on 14 the 19th and 20th of February.
We had presentations by the 15 staff on the regulatory guides, and we made comments, and 16 this morning they told us how they decided to respond to 17 these comments.
18 We also had a discussion on elevating-the core 19 damage frequency goal to a fundamental level, and again we 20 covered that this morning.
21 With respect to -- we also had a presentation from 22 Mr. Bradley of NEI.
He didn't quite -- he was not prepared 23 to talk about the core damage frequency as a fundamental 24 goal.
He just told us that the letter that NEI had sent to 25 the chairman was still the position of NEI, namely that
()
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
388 1
doing so can only cause harm in the sense that plants that
()
2 do meet all the current regulations all of a sudden will 3
find themselves as being unsafe in some sense because the-4 core damage frequency will be greater.
5 DR. BARTON:
So how do the plants get over that?
6 Is that going to require them to make mods or do changes or 7
something?
8 DR. APOSTOLAKIS:
No, no, no, no, no.
I think --
9 well, we'll come -- we'll discuss it when we discuss the I
10 letter that I drafted.
But basically you cannot just talk 11 about core damage frequency.
You have to at the same time 12 raise the issue of adequate protection and what that means 13 in terms of frequencies.
Otherwise you are becoming 14 vulnerable.
But we can discuss that in my letter.
I was
)
15 very careful to address that question in the letter.
16 We also heard from Joe Murphy on performance -- on 17 what he is doing on performance-based regulation.
I don't 18 think it was anything that we didn't really know already.
19 The only thing was that he -- I think what came out of the 20 discussion was that a performance measure that is the result
'21 of a model that has many assumptions does not really qualify 22 as a performance criterion.
In other words, if it's only a i
23 matter of calculations, he was willing to go along, you 24 know.
In other words, a performance criterion does not have 25 to be observable itself.
It can be the result of i
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
l 389 1
calculations that involve observable quantities.
. (J) 2 DR. BARTON:
Okay.
3 DR. APOSTOLAKIS:
A trivial example is the 4
unavailability of a system.
You never really measure 5
unavailability, you calculate it by dividing the downtime by 6
the total time, for example.
But that's a trivial 7
calculation.
Nobody disputes it.
So it's okay.
l 8
But if the metric is the result of calculations 9
and assumptions, then he was reluctant to say that that 10 would qualify as a performance criterion, which is a very 11 interesting thought, and certainly deserves proper respect.
12 So that was basically it from Joe.
13 We also heard from several industry people.
Tom 14 Cook of Southern California Edison talked about their x,)
15 technical specification pilot application.
Mr. Rosen from 16 South Texas told us about the recent developments in GQA, 17 and I think they were repeated this morning.
Then we had a 18 presentation by Mr. Christy of Performance Technology on the 19 new risk-informed / performance-based regulation pilot program 20 and what he called Task 0 where according to him this is 21 really a major effort on the part of the industry.
It's 22 supported by NEI, and a number of utilities have agreed to 23 participate.
24 DR. BARTON:
Also, it's a major impact on the 25 staff, t r,o.
1 l
ID ANN RILEY & ASSOCIATES, LTD.
k-I Court Reporters 1250 I Street, N.W.,
Suite 300 i
Washington, D.C.
20005 (202) 842-0034
390 1
DR. APOSTOLAKIS:
It will be; yes.
()
2 So one of his -- and the objective is really very 3
ambitious, to revisit the whole body of regulations and see 4
what changes we can make using this new technology, and in 5
Task 0 they've identified I believe three pilots, and they 6
expect, depending on how successful those pilots are, then 7-they.will go ahead with the other tasks.
I think there's a 8
total of six tasks.
9 DR. BARTON:
Right.
10 DR. APOSTOLAKIS:
The final task, of course, will 11 change the whole world.
i 1
12 Now task --
)
i 13 DR. BARTON:
He also put a qualifier on that, not 14 just how successful, either NRC bought off on what comes out
()
15-Task 0 or it's all over, as I remember.
j 16 DR. APOSTOLAKIS:
I confronted him with that.
17 DR. BARTON:
Okay.
18 DR. APOSTOLAKIS:
Because he was saying that Task 19 0 will be successful if the NRC approves the pilots.
i 20 DR. BARTON:
Right.
21 DR. APOSTOLAKIS:
And I said well, gee, you really i
22 don't mean that.
23 DR. BARTON:
Right.
24 DR. APOSTOLAKIS:
If the NRC reviews them and j
25 makes a decision in a timely manner, then you should say O,
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
391 1
that Task 0 is successful.
()
2 DR..BARTON:
Do you agree with that, George?
3 DR. APOSTOLAKIS:
Not quite, but -- well, he 4
didn't say just --
5 DR. BARTON:
He's leaning that way now?
6 DR..APOSTOLAKIS:
I don't know.
7 DR. BARTON:
Okay.
8 DR. APOSTOLAKIS:
I don't know.
But he cannot 9
really come here and say --
10 DR. BARTON:
Right.
11 DR. APOSTOLAKIS:
Unless you approve this we will 12 not go on.
That cannot be --
13 DR. SEACK:
He can say it.
14 DR. BARTON:
He did say it.
15 DR. APOSTOLAKIS:
And he did.
And he did, yes.
16 So one of the pilots is the hydrogen -- deals with 17 the hydrogen control requirements, and they want to delay 18 containment hydrogen sampling time from 30 minutes to 90 19 minutes.
And they will supply risk-informed arguments why 20 that is justifiable, and they claim that no other procedure 21 or hardware changes are proposed, so defense in depth is 22 maintained, safety margins are maintained, and so on.
23 So that's one.
24 The other pilot has to do with diesel generators, 25 and in fact that presentation was made by Mr. Rosen again, O
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
392 1
who happened to be here, from South Texas, and that's a
()
2' design-basis event.
It requires that the diesels be able to 3.
start within ten seconds given the ccncurrent loss of 4
offsite power and a large LOCA.
And they want to eliminate 5
this concurrent this coexistence of these two events from 6
the FSAR as a design-basis event.
And they have contacted 7
the diesel engine manufacturers to see what interval would 8
make sense.
They're not sure yet.
It might be 60 seconds 9
or he said maybe up to 90 seconds, as I remember.
10 DR. SHACK:
I thought it was 30 to 60 seconds.
11 CHAIRMAN SEALE:
Yes, 30 to 60 was what I 12 remember, George.
13 DR. APOSTOLAKIS:
I remember vaguely 90, too, but 14 anyway --
)
15 DR. SHACK:
It was the diesel guy that told him 16 that in fact the 30 was easy.
17 DR. BARTON:
That's right, they were going for 60, 18 and 30 was easy, yes, but 60 to 90.
19 DR. APOSTOLAKIS:
So that's something that will 20 come down the pipeline soon, and basically that's really 21 what we discussed last time.
The main subjects were the 22 elevation of CDF to fundamental level.
That was the first 23 time we heard the staff give us the presentation that they 24 gave today.
And they told us they were leaning towards 25 recommending delaying action, and evidently today they
+
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034
393 1
seemed to also think the same way.
The regulatory guides,
( )
2 we heard, you know, how they responded to our comments and 3
this new initiative.
4 So that basically covers what was done.
5 Now, since I have a'few minutes Actually the 6
next item is writing letters.
Right.
7 CHAIRMAN SEALE:
Yes.
I was first going to ask 8
everybody if they had any advice for you on how to handle 9
this reg. guide problem.
10 DR. APOSTOLAKIS:
Yes, I wanted to ask that 11 question myself.
12 Well, first of all, do we need the reporter 13 anymore?
I mean, if this is --
14 CHAIRMAN SEALE:
I don't know.
Do we?
(_j 15-MR. MARKLEY:
No.
16 CHAIRMAN SEALE:
No?
17 DR. APOSTOLAKIS:
If we are moving into letter
'18 writing.
19 CHAIRMAN SEALE:
Okay.
Fine.
20 I think you get an early day off then.
21
[Whereupon, at 1:30 p.m.,
the public meeting was 22 concluded.]
23 24 25 I
ANN RILEY & ASSOCIATES, LTD.
\\s Court Reporters 1250 I Street, N.W.,
Suite 300 Washington, D.C.
20005 (202) 842-0034 1
i REPORTER'S CERTIFICATE i
11s is to certify that the attached proceedings before the_ United-States Nuclear Regulatory Commission in the matter of:
NAME OF PROCEEDING:
449TH ADVISORY' COMMITTEE ON REACTOR SAFEGUARDS i
DOCKET NUMBER:
j PLACE OF PROCEEDING:
Rockville, MD 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 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.
JM Hundl y Official Reporter Ann Riley & Associates, Ltd.-
i s.,
a
g j
t O
e st e
'n t
y t
Ce r
i m
r Rm e
Ne b
m t
w o
oa e
C t
t N
l S
8 l
t
,s u
9 s
t ny o
F 9
n oc c
S 1
o S
R 3
i i
i s
l s
s io O
,g C
i h
m vP n
A c
mo el K
o a
i r
- C Ra M
t y
o L
n r
l o
o aG s
t i
a i
t t
a l
a e g "hty m
u s
tn t e aR ee o
e t
t f h
S r s
da oa T
e e
r el PS ti c P
nu UN
}
O
., [
2 O
y
=
c ri ol g
f o n
P i
h
,^
d cl n
a a o e
orG n
m p
o m
h y
i pt s
ae ino ca f
vec n
d a eie o
n nS r sr o r
i o
a h
ve s
p i
p C
i t
er v
a it n
ir o
a aR w
e f
t r
r d
N e
l n
p d obh n
e e
e t
a f
s d h t sdi e
et a aew n
r s e c er a
i OP o s r ed l
f pi o ad e p
o ov s
e n
i r e sesc o
e pr at no a
so s o sdor d
eicp e
t p
f r
f d ud eo s
t u
ae s n bt e
P s e s a t
t u
coo i
n t
q e
ee n
e h
e hh s
r r
t h t t d o
r t
s se e
e r
sgt sr e
t n
e t
un uan h e
e i
ct t t l
c sm s
saae S
s e
ihhh i
R d
et d
sa www ost o
CA TaS T
O
3
)r O
e s
tt n
e o
l
)r c
eS t R d: 8 n
t
)
9 r eC a/
el A a
3 t S s
t eR 7 o
oo 9
t rt l
C/
F pn S
7 R A D
do 1
C7 6 C
nit
/
9(
aa A/h f
d o
s 1 t n
6 e
1 p n
u e 9/
/4e o
sm 5(d i
sim t
"n n-1 a
/8 oi v
e o
(
ie e
t c ae t
l lacs e
dr oen i
t e n
d d 8
Gof o
n e 9
r e r
/
n t pd n
ar d
y 1
e o
ce f
d u
f en 8 e e
i t
o at o a
f d t
a r
dSut d )o o;
e Ok g
el qn n
t l
t r p
da e e em s o m
c ntndm i
i r e l f a
e eae mm ae 7
o f
B m m "e t o7 de t
9 c
a
/
c9 et 9
P m
ma odns e/
da n
R i
cnf y r2 d
S i
i
/
eu e c a7 vp
/
d ou r
r e
G rfdi
(
l Sao o o a py u
R l
t p f
n s
Ro o
l
)
f g 7 a s
a Ct sa f
i r
l AFaf a
9 n
n e
t y
/
,Doo st 2i a
n sCgn e
1 n e
l r
dfe a 9 o P
g
/
ef yo i
i ot t k s
( t c
d t
t e a sl 8 a i
e a
a 0 e en m
enf r r
g o
e at l
2 sis d n n d
t o
r ain t
s o
s a e 7i u a gi s
r sol f
ovn 9
t iein mm n
Sissia n
r so r a Yo aleuc ad Cc Cmio i
vg k
i v
h n Er R
s i
u o
R n
Cf Sf N
I
=
O
l n
4 o
O y
e it t
v a
e it r
fa c
e s
y e
d j
i r
s b
s la o
e o
n t
t i
n a
t y
o i
l e
u v
r c
g a
i m
g t
sn i
a e
c d
ei k r
a v
i n
d s
a i
o n
d y
b gm u
e ro u
h n it f
a m
t s
c a
r a
co i
a i
o l
h s if o
f u
s wic i
t d
n g
a o
e e
ye i -
v k
r F
gd M
t s
R od i
r c
r f
E i
i l
f j
s c
L o e o
e i
dm b
f' e
r O
o f
p o
or s
a a
s f
h o f
e y
t t
r r
s t
n e n i
A a
n o
mk n
i i
a t
e d
i as l
i a
s dt p
n t
i i
n f r d
b e
i e
n f s o n i
el u
f i
i m
s
- d. a t i d
s t
i o
n s n
np l
a F
eo a
eg ee ta mi C
D dl o
t iy e
g C
v rt pn i
i s
e f
e e
pf oa o
e l
t d
h oa er u
t r
s ve n
se f
p ec o
sc o
pe dn n
ah i
i u
ta ea e
t e
s n o h f vl ed a
ri u
at t o e o h u l
Eg Tg O
- y 5
O f
sr o
ea o
ir e
n co r
np m
u
- s em s
u k
e a
s qt e
e e
t t
ir n
m r g as f n u a e
h l
e t
ei m
)
r qo l
v a
gu eg t
n e
i t
t e
ad d
o a
a h
r c
t t
es ayt i
(
s t
f vk de n
o as n a f
n y
a i
o c
sn r
a s l
o a
u mi un h d i
i t
q l
a o
rt no t
pn c
p a
ea i
t n nl ni a a e a f
f s
i m
o d
d o
e
,A f
i da og nR o
n d
i M
o v
et s e P
i t
o b n v
e i
Or o
mi sf o
a r
i ta n o r
f c
p c
r c
et ee i
m s
sd t i l
s b
t f
s a
O en nn e
u d a u
e r
p H
a?
d i
l Q
er d u s gs A
e o
eqn nt r
e c
t af s
p ao o e t
n n
s e
sd i
eet mc a
i d
s s
Oe rbr an a
d d
e Hh p
i o
e pk Qi xe u p
r s
er g ic n
t s
l a
o fi d b ef h n i
C n
or n a eh n s o rt s
r a a s a
o ni l
t dc otc n
et se sl i
r h e t e Outn at l
i l
e t
a a o
c o
l i
c e o o o H
a er hl n
Ns Gg Qsp Rp o
C O
6 O
n r
o e
i h
s t
i o
ve s
r u
r t
o o
n n
o r
y e
i i
a c
m t
a y
v i
c e
l c
o i
l dy p
n p
i e
lo s nh su m
Pu ap G
em i
so S
c Gi m
d r
i Segs t
e n
o co a
h h Sl h
ac c
et t
i eh r
es a
h dh psy p
,e k e o
t r
nt eb f c au p
gagtycd a n ms f
n s netre t e s
i p
i i
c sr o
iGof A
s O
e t y a ae t
d vSmsr v nh se t
e a s po eo ef n
r e a
fh pct c r c ca r
n r s n
ot iRew u
u h
y e
f r
s o a
sosNr o c c P
e nf u n r n slv?
k l
ve ooct soe c eos ar st t f si o
ts?
b i
f e c t
i cot st f
niy at c n e
r e e pl ost wf e
a f
jr e ne s ninf at of bur ef a et r st e l
ope nee baoa d
dcs edd or r eeh po a e en er ee dd h udd i
t r
h u ee t
hh smt se t
t t o vv qen esh ef eysot e e ee oo r
rf ucarew r r r nt r r a
a n
aica s yd d e app r
t l
i s
t e t
l on t
iepdl al mm awnu i
al hcdk uae hcii h
W-W-
W-O If
u y
7 f
O d
ir e
a )e e
l n
a cc e
i
,t e
s yc c h
a t a n t
nr a d
g e
a pm i
n t
t t r a
r s
e e no i
ef s
r cr r e
t nr e s
e uup cl s
b f
a ot a d
cr
)
t n
l t
e e n
u nl n i
o y
o ef e s
e C
h m (r g e
lp s
(
i t
t p
a s
asd i
h e
v c
n e ms i
r t
a d
o t
o eo c
l i
,t a
r u
s Gip p
o r o r
Op i
h v
Ser o
e A
s r
f h p tc ot e a
s d
o d
n e
e c
e n
o s
sh a r
a t l t
o uip n
i n
a p
cwe o
a r
p (r n
o i
e f
l r
i P
d p
ci F) r h
o i
e sRe f
s e
n h
pnEi s
n s oLl n
o t
e c:
f t iI r n o: nt Fd o
a i
e al Diu a
t eo n gl e
aprCg c
hi i
t s a
i t k oiv r c p
l t e ea m
ar Cp.
I h r Wfo O
8 O
?
?
t e
n c
u o
k er s
s r
it s
a r
o i
i c
l r
y u
a?
t o
y e
g f
e y?
s )e l
l sr a
a r
cs ee u
s d
ne r h t
g e
ec dw c
i t
t d
a n
m s
ic ae f
?
i r
sa s
d o
r
.l s
g m
)
n f
n t
p
- e. e n
i n
s a
n o
y i
d cd (d
c o
n t
i i
e e
C o
s i
k
,n yr e
l t
is u
s ea t e o
(
c e
c h
o r
s f v P
o i
i c
v n
t e
ao r
a r
n ei sc G
p e
n o
a r c S
e ei t
w y
r e
e m
h o l
t n
e f
Op e
op a o a
v i
r t
p n
u s
l d
l c
p A
o lp y
t s
u i
d s
e c eo n
j d
n e
m ta n vn e
n r
i n
a t e or o
o a
c o
o i
g r o t
i l
t t
n g
ca pt s
i af mc d
v i
d d
a n
a e
e e
e f o f
d r
l i
i P
s d
d s
e e
ng nr e
f e
e e
o n oo e
o n?
i i
s n
n i
sl sd ia s
t sa n
n vn vo n n t
i e
i o
o ea eg or fe r
rt r
i i
i r
o s
s s
s u n
dr dc f
i i
i vc e
v v
e r
l l
i a
e e
u ut e
b od oe rd ir r
r n
n o
e s
s Wu Wm s a D
t I
I I
irC O
1{l j
9 o
o t
no n
y t
n o
t i
i i
a o
t l
i r
a b
i e
t d
a d
u s
n l
l t
i is o
e a
if n
s e
m v
o e
a r
n m
c e
e e
b o
c n
u c
r o
s~
s e
u
)
i t
s s
s r
o e
e s
n s
i y
s r
o e
i t
s o
r C
m e
a f
(
f m
a ok b
3 o
s e
n h
o s
t 9
c p
c t
/
a C
o na n
1 u
o r
n eb e
3 t
or
/
o k w m
3 n
p f
a e
t a p
h e.
m n
d r
A c
(
ed o
o n
i a
b c
s e
i d
o n )n ds c
s p
n r
o e
e o
i l
i a
p si uu m
d i
l s
os b
n p
i ev h s u
m y
a a
vi P
d r e si p
o d
l r
n t
C u
n r
e i
t o
c a
on mi o
s i
f t
l n
t r
n o
i a
yt t a oo r
e t
si c n e
h la n s
e lp ne p
t n e ys l
r u
a u
d ed om bm d
p f
l i
e gn i
i t e s
r e
e m
h s
u il o
p dp s
p oe dm oo c
m o
nd ai pC S
r P
o m
0 1
O s
n o
i t
a c
if i
do M
l a
i t
m n
e t
A o
P n
Of o
o ita n
c o
i f
i i
s d
s o
u m
c l
s a
i i
D tn d
e t
e o
l p
i a
-m t
f e
o D
n g
s S
i o
n ir r
O O
p C
O o
b
1 O
1 o
t F
t o
D
)
e n
F u
sD C
n gl a
l y
ini wC iw
(
t r
t l
n a
a nd t
r y i
o oi G
c v n oc e
d o e nn d
i s
i ord e
t pi au cM b
l y
c u
s t
c e
s oq aR n
f ec s
d va ge rS o
a ni r
t S
t e ef o
a i
f 0 h
t e
cg v e r9 t
a l
t e
t g a9 i
c a
n ja a a e1 d
t i
f n
bm i
t m
y e
e no i
i a
l o
m isd a
r5 d
d o
a l
ud o1 na t
a sie q
t eo M
d is e
ce mg s
ro ar an n
ma c
o eu mla l
s a
u t
F ms e a c rJ ot i
a r
n do e n c n n
a oher "t r
e e Ot C
i e
ef p
r m
t v a e
o o
t eael nl 4
s Sa a
P y
h h sp f
0 u Rd i
t t eo c
ar 1
Cn f
n mt e dg r
f o u
o e
aaaw ye of Af t
v n
u hrh o l
a t
ed r
g t. p t
ve eo i
i o
q t
c a o
e dr
.r i
i l
t t v tt t i
ep ea pit r
s co ed F
t ce xn er s
l u
e oynl jp e
c ea bp 6 a
,t c
g nr a or u t u o a 9 v s
a t
t un oq 9
n a el s n
Fs e
m i
1 g
Da e
D l
mgau s S.
sn a
Cw 5 e D
U ai e ee wd yn b
1 e
t rl a
n t
r a
b no ai se r o t a o
s at o pd it uv a
d n gt C
en i
r t se i
yuo ie uc ss u n en s
g c
r c
erf bv Aj ri e
n i
e lou a t
e c oe ur b
it n
n a
Ppr o cd Sp o
I I
ve O
lE
n 2
O o
1 it e
n c
i e
t v
s c
e u
a s
s O
r rp r
p e
H e
n Q
v h
i t
n d
n m
e o e
a g
di n
it s
h vn u
t i
n d
oe y
y a
r r
l t
a v
d o
n P
)
l pe t
g r
a r
i n
e dp e
t a
c o
r r
n t
l r
i c
ut o
e g
l n
on a
(
c c
e o
we e
t sn t
d u
s n
a i
'e u
r a
,i o
l
)
t c
a t
F e
s S
it c
c v
a e
Ra s
s c
p Ef a
n e
h t
m l
f x
Lo r
e i
i e
o c
h w
id r e f
i t
l i
o c e
l o
a
(
n e
w n
M c
n s,a o
h i
d l
i i
t Oi h
t t
e n
la p
Or a
r t
i t
o Ho d
e a
)
e n
t p
n d
l Qm e
n s
u n
u r
ei o
u c
o t
o o
h h e s
a m
l o
c l
i f
P
(
t c
p v
e m
y f
y hi l
e c
t b
e o
c s
t i
ia i
t b
l n
s n
'n wle o
i e
n i
o os r
v o
P u
t n
t a
i i
s q
sn o n d
d G.
c s
e se io n
e a
S t
t u
r id c
F mi cs a
ta h
c nw l
t f
l s
e mc u e v
r c
e e
o iD g
oa ni e
r e
j i
a v
Ce o
a t
r l
l m
g c s h
o e
u a
s t
n 'n g
g m
c a
e D
t m
io h
a a
i u
a r
e t a
,ins t
l o
s s
s s
r sd a si e
e e
oC ye oi t gma d
d d
s r
l g
r i
i i
g v
v v
n o
Fmi g
a n
e c t
o o
o i
Doi r
r r
r f
i lCo CCm P
P P
B tav s
O e
or lE P
1!
1 3
s y
1 o
s ie t
v O
ni e
f ot H
ia da y
Q sg ns si al l
i t
a r
n i
i ct mm in a
a s
h m
ee r
s t
t n
e e
no o nm c
v eC e a m
n gd
)
e i
t ea t
n n
n nhh h n c
i i
t u r
t t o
u t
a of c
?
s t t s
bo na e
(
e s
e oh r nt s
b e
r t
s e
c u
ik n
l r
o a
y r
nt F
go a a
o o
nie Dm l
i a
g a
t n
m isf C
c e
a dse h e r cc f
h e
nev a ni i
A b
ori t
i t
o g
R d
p pt e ec d
n sme yb a
)
n P
l r
r M
u eiv of p t
i n
t y
o r
o e
r r ee t
l f
oi a o
e o
r c nf c
oh p a
a t
c t
r ct m
ot a
a on eis
~
(
n t
rt e
y t
d e
ut o a
t c
o n
y r
n ode e v o
n n
a r
h a c pe o
P e
s m
o t e n 4
e c l
t f
u t
t i
lwdt c
a 0 ;a o
a n
y a
l l
r 1 sp e
a l
n r
l p
e auo e
o F
p?
m oop gi m r
t ni i
i f n r
oo e
gcm s
e i
v e i
s a
si la p
Fl l
i t t i
a uct u
a yt o
4 Do r
t s e
yal c
i m
ee g
0 C gh s
1 d ce a
f a u F
f eg af v i
D D
sq D
o h
eia i
f o ci e
eo C
t n r ch n aa e
l r
t l
o h
s a pl n
a a
l o
emge t
i mr ssw C
i d d d
g nr i
eoiu f
t l
l a
ue u
F t f ct f nl a n
ol o
D ar aa aa o l
t el i
t l Wa W
C S p pf Spg e
ta s
c v
n e
o O
~
lE C
~
4 de 1
O s
d a
n b
s 4
a s
e 7
?
s e
c n
1 o
n n
it i
o s
c 1
i l
t e n a
G s
ita do r
e p
R u
i lu ui s f
g gi f
d q
c a
e o
e e e t
s t
r cd s
o n
)
f n
t o
t p
i a c n
o d
n i
t f e
r e
n pi o
o r
c e c r
p l
l i
(
t c e u
n a
s s
a cp c
c i
n e
n as t
d y
i i
m ct c
e o
t l
i n
e n
i r
it v
a t a i
a l
a t
x sl f
i i
f s
c c
e p
e e
s i
l r
if A
r br o
e i
d i
o d
y o
af t
s c
o r
f b
a e
M o
s o s d
n t
a r e e
,s n
i a
w pv f
e u
l O
a l
i i
u f t d
c i
e n
g s
oc o
t b
t i
e c
l e
a ej m
e t
R o
sb a
s r
t o
g uo e
p n
P c
b f
y sy f
a i
l e
o f
r f
i t
o c
e a
t a
p e
f ki t
l n
p a
ad d
s a
e t
u s
mis e
n d
o S
i f
b s
t o
4 n
e i
s n
7 u r
v u
a es s
y r
i sr 1
a u
d c
l e
d m
c n
s P
ut 1
n i
i t
e a t
h f
D n
da e
d n
t o
i dm i
is ul e
w y
s la nc Go m
a t
y e
o ei r e
c f
G n e yg t
t r
a n
a n
oy t
e s
i y
e t
s t
t l
e e
t ag af s
e n
h y
i l
f ir u a c
s t
a ge gs n
l i
S ih en o
o l
l i
r o
Oo Ro P
C W
t f
s s
n e
o o
O s
r C
U P
m 5
O f
y o
r 1
c a
l s
n e
l i
?
w mre y
ou n
r r
o y
t q
o e c c
m i
t e
t t
a i
r nef c
s l
r a
e o
i d
e u
P "e e t r
e r
4 q
7 o
G.
npa re i
i 1
S t
le4 c p
s n
d 0o 5
e 1
s i
e i1 s 0
c G
O d
h u x s i
g a
e R
H e
i t
1 tc l
a
)
1 f v Q
l e
s oi a
d a
l fi t
r t
c oh ec p
e n
c t
n n
i t
ne s
a y
g o
ay j
f o
h f
l c
mch i
b a
p t
r n
i l
ni e o t
o y
a d
t i
(
r s
o ew d
s r
l t
s h
u i
f ul ut t
p o
n s
n s
rqa gl n
r i
l e
c e
a e
e t
a o
l ee r
h n
t c
n ei a
pr r
r e
i I
t o
f e ch r
t o
e n
l a
G ada n e c
n s
t u
n c
c c
r y
etem a v s
n u
?
e i
i t
t d
'e u
s t
f t
i i
na pt c
e e
e i
e eie e a e
n i
d f
e s
b s
v o
a gcv ci f
i t
s s
l i
oit ct e
y t
f n
e l
M S
n e
a c
sc a
r l
l e
e asa a
d o
a a
h yao Fu o
c h
Oi i
t g
n l
a d
a t
s l
i dhd Rq a
e w
e e
t i
d t
o ut a Ey h
A m
n t
e h
i d
t R
r t
r Lt s
t e
e swf i
i f
e e
g P
o l
r i
v rl o da d
c e
a n
f t
t o
ea et y
n i
i a
o t
d i
t i
P c
hr e pf l
t r
i c
n u
e o
ke t e s o
m f
e r
y a
u c
e t
j t a l
r p
s a
m a
sc r
l o
b uae el e
i i f
c d
rt n
O rml v a b
e e e o
f m
c f
t n
n r
e o
a a
o y
o e d
a e
i l
t r
f f a t
t b
n i
r v e m
Fp s
a dig f
t o
d s
i n
s Rf i
r a
e t
o t
y s
d ec a tG e
n t
a f
Ea n
s al s
c u
l i
c s
oo 3
y n
e d
a p
Lt b
pi g9 4e y
r m
s t
l
,t r
e a
p s
g u
odn9 7f a
u a
h f
n n t
i iD S
rain 1 l
a o
a io pr
- 1. S m
c l
t e
s i
a f f n c
o u c t
t h
r 1
y t
oei r
n t
r t
g i
s n
dd Go l
e a
t e
ya i
e Rf w
e f
F dp a
ee o
F msn m
y g
m a
R am t
e aoa dt e
c E
ei s
R e e kn ea t
a a
e L
t r
sg a
n E
al e
i l
er t
e s
s h
al rom oo s
t e
e t
a t
L ts t
ewr pr s
d d
d d
si f
e r
y i l o
ycg;t o u c
s i
i i
l l
f e rr r
v v
u u
f n
aa s ps n
i o
o o
o a v i
l l
o ol e a s
o o
r r
t a a
P C
P P
W W
Sh Payw n
it I
i s
n s
O n
if o
o e
r C
D P
=
n 6
O o
s 1
t e
c h ed e
k t
a l
ia f
wt m.
e r
4 s
nge 7
h e
oinb 1
n c
s skn i
i h
e 1
l ia r
a c
e amc w
G d
i t
R p
i n e h
c u
mot c
a d
G r
oia a
p e
s cim o
s s
ci rs f
o s
i f
r
)
o et pe a
p y
s f d e "s pi t
o t
l t
an s
r a
n dynl t
t p
n i
o er c
s oa u sa n
A it s
sr e
e
(
t t u aa e ye r
h y
r t
r s
l r
l c
n e uh a
u n
o n
t e n
c t
bg o
g i
v u
a t
n deoi a
t d
i l
rst c
k urt a yd c
e u
a i
l t
t e e
n g
a oosi nt f
l e
i i
t ia e
e R
f f
M hfixn ai r
id sdea t c d
o n
s e u
ro o
d t
e s t
s n
M o
e s aq c s d
a a
is u uhe n a e
Ola st h u
s y
i l
c a
i i
t e
v est d
f e
c ue dn d
c i
i n
D n qi o e a lo e
i o
y ain p
o t
tt t
c P
enid ok m
r s
a P
o mhae A
l r
t crb ei t
e r
p R
f a
v b
t eeir t
f P
s o
lu at cc e n o
f e
n g
hens da t
a e
u o
t t
e vua fl h
s s
R di f p d
s i
t s
tl e a
e t
f s
eaab f
i tso e
n o
t ti u
n l
ait n
e a
i c
no
,g r
n c
t c
in et 4n y
r o
s s
i e
datl 7i e.
u n
i i
D t
n uoe
- 1. d m
c s
h i
n iqpv n
i t"
e 1
h v
c a
t i
t e
e a
nt el eahe Gt n
r t
i t
s e
w re mh c
Rr t
t t
t n e
m p
n y
c e
n d d e
c m
t a
e t
c n
tdud e n a
n o
an oi s u t
e r
i U
f t
f o
s t
p n
i f
s a c n pt s
o
,c n
d g
o o
y ys a i
c e
c s
i t
iO c
r l
r s
n o
pr n
u i
l l
e oHt e u
o o
o o
nh m
PQit n c P
C C
N I
t s
a s
n O
e o
o r
r C
P T
7 g)
O ne e
e.
1 l
it a
b i
i u
ds
(
u d
s d
e l
l v
ch u
e n
i nt o
u i
i s
ed d
,f w
s he i
n o
d i
i l
t s es a
k na mle o
s ps o e s
ri g
r i
r k
o om ed l
d l
t f
s a
vd r
k ry n
i t
ea et o
e s
pf c
dy i
)
l i
i a
r f
c t
e e
n o
olt t
n e
g bi s
t n s
o n
h h
se i
c c
dt r
g) e r i
o t
t
(
l i i
n s c u s
s a
uw w
l i
r n
f u
o s e u c o
o c
he s
s s s s n
ey o,s l
t s
a o
o rl s
i s
a e
a m
c sd s
d n
r e e
c k
d c
i r
s sn r
a f
f s
e e
yo a
a f o i
i i
t t
ar d
R v
p nf t p i
l i o
n at lo nla m
k s
i l
i l
M a
v au o
ec da t
d ca en e
e n
Ol t p c
ib t o i
a t
ic n
fio w
c i
l i
o f
a kp ad mi t
i t
e S
e o
c o
rn n
a il l h
l d
g a a pa u
t f
o g
bt e
ys oc P
n s
s o
g cn na l
wt i
i i
h n
no t
l o c a
a t
og a
eit i
fo e
o h
hn h
ga td r
i n
g t
ns c
al ae o
T i
u cs y
w ou d
f c
oi e
t
(
i ol v
s l
v e
n e
l s
u a
e l
s i
f o
ci o
yc a r u
t a
c e
c s
s ns c
w nl ra ay a
s e
r i
s dl n
e o er e
a i
ia i
b i
t t os D
v p
un o
sg af y O
i i
ta m
ga s
sy isl i
t h
t o
sk v
nt sia i
e s sn t
i c
as e
of l
l e aa a
a i
r hr r
ca cb y e
u s
o d
d el r wla H
q f
e en nao e
d d
ote s
ve ct v
n o
ni o
c e
oe dianl l
t o
h f
i f o pd rw uhu pt ocg a
c t
o a t
e e
a s t
S M
Sa Pm me Wtee t
r i
b r
n s
I s
O a
n o
u r
o Q
P C
8 s
y 1
O d
i l
s l
lu ys u
e f
o la v d
la h
ni t
s t
n e
e Ac a
s d r e
i ne yj d
ce t
or c
rb e
su a
av o o s
e t
l p
l ri gs ad u
of m
nf un e
a l
o ga b
j i
i at t
l mn a
ne e s o
e t
ec Rl t
n em vn a
e ee eo s
bn r
h g d
)
m pu t
ni t
o a
n at n
q k
h e
n s t
f o
cn o
o i
r s
ir e
c o
c v
en d
i
(
n s
o n
co eh m
id e
nc dl t l
n t
o an a
ua a
a e
n r
t l
na i
t e
r c h o
a me h
o s n
i i
pa t
c c
g t
i a
a a o
mt wl i
i s
f t
l d
nm c
ob u
i r
i o
o a ea n u c
o l
p a
M cd n
h p n
c o
t s
d e m
o 'n l
i Oi a nr t
ni p
it ao a
ol ao o
t n
a ri n
,c l
l t
es e
i t
n n e
t g mt e e d s v
n t
n e
n i
o ei a e mm sm d
i P
d v nm e n om o
t n
i l
t f
co oe ao co t
o cv ct t r t
a r
n an dt si cC v
o i
o n s nn s
f i
l s
bt ay oe ma ys f
i ol e
l i
s n
n c
o ri s r n
el s e ne e
u o
e h
ov v
o c
rd h
i i
t p
mt ce i
s a
e c t
i s
o e
ns l
i n
h c e e md D
e t
Ca h h e
e c i
m wt o n nm tx a a
ye Ca oa ep t
b r f n s
n e
oi r n e
em a
o d v ed o
t o ri cs u e ei at ee C
i v s ud l
t s
ca d
nf s
n eo u
s n s e e qn s
s t
en e a l
i i
t s c e
n i
r a
dt v
s d me d e d rs l
ic ei L
ad i
l e
r e a sd vt ud lud i
o p
onc oiu on i
f s m ae r oc Rb Pca WG Cu Ai s
s s
la n
O o
o r
o G
P C
e 9
r O
1 o n t
ma r
y o
r h
e n a
ys ba r
r h
o ah yt r
p t
ar oo m
pb me h
)
e me t
t t
n g
ed s ar n
u n
t o
nl o s, i
c r
i iiO t
c u
k n dH
(
d s
i o
nQ i
s s
rt o
k f
n s
s oys cd n h n y
e i
o spo r a r
ic n
e oi a s l
i t
t n
o v sd oa rl i
a e
eo n po l
u a
c l
l o
mg i
q o
eh c if e
g p
ee l
r b
d t v i
a f
e alae ri d
n_
e v
t t
pog o a o
o g
i f t i
t t
a a
ega i
M r
el a
r t
c ye c a r
e it ct v n u u
v n
ae a aq l
g a
a f
Oi a na d o i
u osm f
l i
t n
a q
r ut n
o u
ese ge C
n a
g 'n t
v e
n e
a og ot ri t
t n
a d
sin f c o
a f
ss o dj e i
o v
esl l
P P
o eb mimd eo s
rp f
y m
nma y
n n
r o
a r
o o
er r
e s
o s hi a i
t l
n o
sCn t
p n
a s
d o
o m
o gi ei imni z s i
t b
d t
i i
e s
i e
e n md d s u s
T s
t s
r s
a o on n a s o
h u
o e
rt Ce pa i
i t c l
a f
r pr r xt msa c
s p
o u ae n e ao x
r e
,a s
a e
pig g
l
-r e l
l t t i
o pf csp e an D
G r
a n s
e non vie r
a o
eio o pla e
e c dt d ov v
s b
i i
i t
t vd a ri l
O yn onr upu ita H
aa r ou opq t
l Q
Mp Pcd Cae itn s
s a
n O
o u
r o
Q P
C
'lI
0 n
s dn e
o c
2 O
i i
c t
p a
f n
c t
o a"
e n
o
.A t
t s
dn o
e e
R t
io m
s r
P p
ui p
e e
e gs c
s e
a t
h
.s t
n ei t
t rm a
s n
t p
o o
u m m e
c q
y o
c n
o e
c s
n e
eC d
i e
l o
e d
a o
c ie p
C w
vh g
c i
)
t t
t s ot n
o n
el r
t a
n o
h ba p y r
i oc t
g o
b n
n p
i i
(
c u
p g d
f o
d e
d e
o hy ue d
s n
i l
i s
t n
o n
st ot v
a s
i o
r e
n e c
n we e
s e
f P
i e
oa i
r
" p d
n c
ta e
is f
i t
t x e
n o
t c
t a
a c
d ae r
i i
a s
i l
t a
f en h
e e
i u
s r
th d
c s
r id q
a t
s n
o p
i t
i o
e e
eh np s
p f
t e e n
a f
M d
f d
t a
a sp md o
A n
t a
s s e i
l O
ud e n c
u e
s d
w c-t i
e g
r t
i r
n n
n a-e b
h u
o e
a o
sn t
d-s s c
c t
ii h
o h
e nh t
u c
t P
p f
ys ye a
s s
a t
cf o
'n p
f e
o f n l
e t
ee g
e i
l o
D if od d
o m
~
n r
i e
pf y
u s
i n
n e
b d t
o l
l w
e c
s a
o m
i t
i f
nd e e n
m c
is e
e r
a i
i s
s t
en n u s
ms c
t o
p u
n a
ma da d
t t
t oi a
c e
s en e n e
e C h r
f s
e a
ai d
d t
s p
t o iD D
tt nd n
a t n e
e sc en e
r n
i o
l g
r n
a yt ma m
p ei t
e t
t o
mt i
a o
s co l
mn m
r ul ir e
e m
a op oe o
p l
l cr v
a o
pe ct c
p x
or a
t t e e e
a p s G
di h
e e
n a r
r l
e u Se S
b dh d
ih a y
l l
h t
t s o e
sq Rt R
y u
u n G f
ee od o
o a
rd Cn C
a y
S Pa Ao A
M Wn W
a i
s t t a e s
n l f o
O o
e a r
C RS P
.