ML20155J581
| ML20155J581 | |
| Person / Time | |
|---|---|
| Issue date: | 11/05/1998 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-3056, NUDOCS 9811120210 | |
| Download: ML20155J581 (195) | |
Text
...
Y..
. ', _,..lN.
s'.::,.
l 9
j.,;k
."0.
- g,,. ' I*1.......,.
r.f..'..
y c
..,c.
s......;.,,,,,.
^
^
^ Y -ll ' ^ Qf ofj., t I % 4 _ry y, W. *
.L
.
- l._ { ', % l
'v,
' ; y'. ~
w V-y.
,[.,',.
.I 4
,t
. tm L
,..1
,g-lg. '. : g... "
.,4 4 ; q,
'.y. (.
} hj,j;.
- .- 1 f
3 f
' ' ' L
~
s.
J.((;..'_
l
.r l, jQ,
.'.Q.,',...: ' bl}l' C f. _ {N j:
),* I. N.
S%.h',
f j ' l " F M* ' ~ ~-
'
- G ' ^\\.
- s. ,,_ '-
.g
,5 y
[..
. ',.. g
- 5 I
(
'i,y i.),
- (,g ;gg.C $..j ].. '].g;.
./
,} '.. ' -
k
. (7
,, c.(
' z.j., ms y ', >..f,M. 'i.i Q p,, p,;
j 7: ',.,-
... b. ' 'N ; y ' q Q.[:.. ;,; p.*fW f.Q: 9:; y,4,g;:?
j 3
[
b
- f. ' '.,'.E;) '.
3' l [.
i g
1 ;
is
.,].
- ' 5
'.. ; j f.: j,l *}.,. f;M.( $.[,h],}%. ' '
E
^
- \\
f.
x
_,,_~[_,
g( a w:
y -
s
,. Y,.
.m.
,...,'1((. ', b,'.
, [.['
......'E.
.QM.[
v V-
"{
g
?
,. f
.2.*
f
..;,, ' s n
-\\...:.'.-
~..~
- ^ " '
- E-j ' }-
~ 4
.i... : ~
5' 8
- ..' l ? - S i.,. c.'. '. ',.. [ '.{.'
. *,.,j
(.,'.
o I d2
.;p.7.{g.7q jj M
J TRO4 s-
=
r ;.i m e c.ci p m O.. m s J
aeruny(ACRS)Oarornat Rlfs$
- [t.
g G '
5 i
TO BATIn fi y
. 2%
Mt M/S T-2E26 U
c.
.I 415-7130 T3
, j.
g gg I
.'4
'.'i., 1 i Id
.r
,.L,.
^.e. & ;.: O o..
[ ;I. e 'yh
$$[%%fy 5!
5" ; ' -
E ^
'1.
. l: k, f, e n:.
i'[
(,
'W *[
.j,fh ', $*;,'.;?
?...'.
"'NV
))
7, '
.u
.( ;,3 3 y ;;5 :p;;{'.;,.g.jpg 4
- \\ ' f..l h;')
.?. 7
\\
f ;.
..'l*',
'h ? ; ;. 'f ;
- [.U.N 57'j, ;,
- .
yg, A l.f j ( [ t '.hp. ],'.%./ ; i,; ;, Qly 1,
,g.>.
E fG
- y j, Q., p
- (3}Q >6,' j.k, dip 5
- y., ;.. ),)_k.,,....],.,7_-Lllj I; -
,.,,,..;r
~
- y '.
./ y y
.g.
l t'
'.M ' h.- ' '. ;
).
.,4 h.
- ..',y
- ni ' f.i ^ i n b h p.*,
L,,3 i.., ;
,jo.
....'y:f,z g y p}., q g'g.
~. n.y.' w :; -
.y
' (
.'_[,
L. i
'.', 3.g?'h 'lf*Q,;.;];?,':
V '* [, ^*by -
l
.l
. 4; g_.g.:7.y
.Q, f
A
_ ;_,,[
.\\
/
y.:),
?[h. '"
, _._' ",. l 7:
I :.. *, l p." '
.,._.;.o,,:.
- ~;
?-
' ;:l
. ' ' i;, ;l',y,l j,ffn '[.:.ly,lx.f..l}'$l.n h..i &<
- l,j yf
?
f l!
, g,.p.f. t. -
a
... ~
pp,
.w s;
.~...-a;.
- y y,,.
.,3 ;
.(
. ' ' /,"fp,.y. g Q,.;4af.3. $ g(;,:
>L
.o
. l.
7gv A;ig6ln, I.,
f j
a J'
=..4,
.. l
- ) ].. Q y[R..(..[p.[,$g' 7.n,,%gy.
.t
'.. G r:t.m:. ';.:,
..r ;., yp;p,g.yg g Q
..c 4. ;. g 3
. :., v.?,; g[..:ww::.n.,. +
..n. ~.
- o. Lj:.
h...
. ~..
.c/
o g
...,1,. (.... ; '; - 7,.? * - 3*:j.y. S ;.
4.aQ.
p e-
'g gg j.g.1
_.._ 3 '.p....,. ?. f.i" f.,:.&k.Q,. t Rip;.)d'i, 4
a.M
. m.2
.,.n....
7 :f.} t;Q q ; g, w.y_
G.m
.t '.
'.,...:.. c
,,' ~ 6. ;..[;> C, :.y :. : 3 9 s ';;g_.>:;, in
.i {[ t
.~
_[.
._4,. p.a 4_q,w* l;,y<
(;; 'l
.+
i Ev v-gq 5 -v..%.
'l s,
y ;; ; f*;.,
4 i.f, g
v p;m p:nv@'gh I
., 4
.. 2m a
4'
- - 5.I.Q.};H :.Qfl..
., '-,.....'Oh
'd,
- c..,
~.
-_4J
,y
_..e.
t.g,
. '. -..~.. ;. -.,qq a }
.i'.,
s.
. syw
"..'y'.
y,K.R *_ "Ki Q Q'.
.2)' ;.'[z
- i. '...
J..
r, o :- :...
~';
.4 4
... ; _.;.,,. ' &l %,. _
s.
... ;." T ; ;,
});i.Q. ;, ^ ' j ( ?. : <.
j,s; N.
. Q.,g y ye.'. j.
L hf
..}
.N
' W y ;.. -
- 4l.
t o
'- * ., m;,,., - -
j.: p.';." t; ' 9[.j f h, f[ y, j
'.a.
- .3
'q 4f
- ,q J,.,
3A
- (
f
, ', * : l f,
- *g.
~ ;^ ~ : J,3.; '
.[ ' [. :_. Q j.; s. :_ u '.l;f. gg;. q;,.g. )..gg,g.ygg y _.y g _
~.. ;.
- h..[g,A' fh$.
s
-?'3,;i
- , n~ - m,- - nn: -: - n n
= : - m ww w,n w,
V -
- L V.
I. y '..
f-f 4
3 e
r.
N
~
hL dh hhhW N
e wRQU?W M > de,3. 3 nw e.
~ -n,
S-e k*
^ '
- Q L
P, ~q +%g 4-M w-
,m. r
.,,,. TRANSCRIPT OF PROCEEDINGS' Wi n,g ~,~jOFl?ICIA M..~
=
m
. m.
M.,z., w. -,EM ~
O;, -
kNUCLEAR REGULATORY' COMMISSION?
,s...
a.
.a s
y nn, n
ev5m.h.,.
A,
'v
. v
~.,
MS,ma MS[,, ~, 2ADV.IS.DIIMCOMNIITTEE ON! REACTOR 1 SAFEGUARDS
'W n
w M rp.g r (p.
pk
['i:[
lf,p
%,p wgw'$;fkSh[:&
n,
~
4 s
3 h
Y g
,1 f
b h-.k nj a
~ '
d; 4
.mp.hd/E
'N1 m~4
=,.
', du s u n'a,h.' y!.
,'. > t w L
[p g
%rv
,i
-n 1]a i,..
w's.'=
+1 4
+
NF @u,f sM
- i.J s M u
.1 m
'f(.,[...
.,n W
P*
t mn -
Title:
t.
^
YMEETING:2457TH:ADV.ISORY;
.nA m, c w
j'/yM(m ms p p,,' M r%,, m. $v ~
4t
/
a-
.x
(
. W 2
Bl%e $,.y @ ",')
W < g g,4 + JCOMMITTEE N,lR E C T O l W -
s.,,
d[M s
p
+
o Q& NNN' y
't
~
" ~
- s i
c
-.. s a, w q M % n ^ ' > C,',
- 3<.
w e.y
+
~ ~ ~
M F.
LSAFEGUARDS:(ACRS)4 m,
4 mn.g4 @gr %f/g ?
n
.. n y;pw p
1 t@
/ g ;,.
r a N
h;y m$O #b-
- D
..~
1 3
me;-.
y m
y@f;q$.
EFM*
. m W, W
RETURN ORIG INAI s
e
- E.$b@f Mygp M-< ' '
ty t J
~-
r m
14,
.s A
'RO4 (ACRS, e
%p@.~$$
' f.d@e ('j -g%
', ~ '
N#
4 '. =f. s^s '
TC BJWHITE le,e oe A
n
~ ~ ".
d.@h );&
~.
is N
Mpe p&gf$ W WP, %@ k M
3'
/
y/ C ~., E.,. g i
.,j,...s_.
i
%g s
,.y i
s,
x,
y W he; i _
W*
i, 415-7130 h.
Yh Y 3k THENYS!
,Yh v.
- m. i qN.
a e
s, ae x:
a
&u.. 4,$m&Q Q T (w vW M:;p,&+n wY:% r
,' u
,. )f *. - s.
n:
s;i s
,c' 1
w i.
+mu ;
&s& '
'M ' & >
'. ~.
r yp) c vis J. ^- N
- c.u.= w-t ' :.i _ ei o ;g~
w x':m, n
n,o m A
c 8
w
? w
<m-4y-f' p g M M.v Jg'm' Dock'm a i'et"No.CFLR,' l%.1 t., :,
^.v 4
a
? QG st u
-~. +
>~r^
4:v 4
?. sh t &
C43 y &,,
4@g,~s. e xy '
m
,u
. m sgir w' cu,.
n,,
,b,*
m yw p>
ip nca 4 @-N v
~
g, p
?./
)
7" y
a' N -
W.
, ?,d.ja.k,
g 4
',.1, m' \\
"v s
x v
+
r.r s.' y-).
-v l
. '_.f.
w v-l.&
y ;
$[,.5-f - - ' 7 Iv.. L n
r i
-fw O
/
j
'g
""C
~f N 'J,
/
\\.-
1 1'.t W%&
l hWk P 1 Qxhl a
y.,
^ W y
.,. k t
I
- m...
Yb 4,
i G g k
mh;gby%h:%g
'n r
gg,gf H% ' Q
., /
Q.
- in
', v --'"'
N' c
12
- .,.y d
m wy w
- c' - l w;"
.,. s p ;;n r -
rt
,m >
, "ac y"
p yp y, H 4 -84 y: h. ' ',,
p..- V yme
., 4f' 9'sw: r
- y~ =&. r,
s
- W g. ;;
- k
- s t n ry +
op h;m y;&, J m W pu D 'r w&.me ~
r nM
.o ay n c a. V '.n
- >+1 m ey m' sm
?
.c
-z )l:
v 0>
, a: ?p> ^
1 r
Qg%yd:nmgh%m:w q;rp, r,,
y ff%24th Li
'a',
t ' ' ' w* m./
i(
<3 r
M x.
r:-
b m
m x,~
- xs s t
r s
'y' ?
2s 7 @e!N;).C>w; '% -iWorlfOnler/NQ(!- MSBl-300?5K.
m
~,
e q
a m.A
. w4 1
s
~
Q i,f, h;h, v..> < glC n91ln, if+,
. c,3 ia L h
.-c t
$f,hN %),
}< Q
[
s
.M P
.?fi:)Z.+g. s ;y 4n; f Zs v
- ) D +-
g OE
.# 5 r.?
- 4
' N. :.
%Q.hA~):' w v
5 jq s
sh,kY(.h,m), @,}J' Y+. +.9'i
'Q,
~
~
a yy v.
.,./
+
p@g y;m,v :%
d
) R-e
,4(
.j e.
xid wp.w3m^ g
-f u
7;z
~
,t
,3 3.f Y f5
- P'
. '8=.Iv C
- c. ',,
'o,L
- &k%
- y '.sf.. T C;g...
- s. m.Q '%
- g.s = V '
i a
r, i
3
& M}Zi t I
'a-V L O'$gyCAT'I'O.N:3
[
$S/h..h 4
$'p# 70.j%y W gh.
W
- R ckville,MD' 1
4 um x
- y "!...f.8 ' N,2
' ~
,l t L%y, d
9811120210 98110ea p
$%@%U@"/ S L'
i
[$MM,2 PDP ACRS PDP T GO%
+
W me q
m m@aA merS qw ;l 1 DA,TEi
'ahursday, November 5,1998
' : PAGES:173.a318-m M;L,4 mM, Q:g%w.x~ m.xg;;s..
m... m, g np. eM WP;Q" 4.
Ad V
' N ' L1025 ConnecticutAve;,NW, Suite 1914, n
AQ - t --
T a !x#
M.y.
m
).
yn, y +n r %g %r%yNr m, m
' yJ q:p +^
' D..
- a:'
a iWF 7_ ANN RILEY &, AS.S. OCIATES,6LTD.
1 WX'QSeC,. ?g. - m *m %.
' a:
% W ashin p36 m
a 7
7; e -
g-e.- <
dh~N%igrM;p d [~ [ n ~ h b N 8 d> + y d b M
w&%estq a.~.
V % W 3* n g r y A ; f "l, 4. s0H4,%8wQ0T E 88 u m' 6
pu
-a y P;
- 4 M y k*
- r gy ' f p f *
. +
1*
-b
m, l
l-
.t t-i p
l i
t l ':
DISCLAIMER i
j i
i UNITED STATES NUCLEAR REGULATORY COMMISSION'S l
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS l
)
l l-NOVEMBER 5, 1998 The. contents of this transcript of the proceeding.
of the United States Nuclear Regulatory Commission Advisory f'N, -
- Committee on Reactor Safeguards, taken on November 5,
- 1998, as' reported herein, is a record of the discussions recorded at-the meeting held on the,above date.
l This transcript'had not be:en reviewed, corrected and edited and it may contain inaccuracies.
c.
h' i
1 l.
i.
I
~,.,.
173 1
UNITED STATES NUCLEAR REGULATORY COMMISSION 1
(T 2
-ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
\\~s!
3.
4 MEETING:
457TH ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5
(ACRS) 6 7
U.S. Nuclear Regulatory Commission 8
Conference Room 233 9
Two White Flint North l
10 Rockville, Maryland 11 Thursday, November 5, 1998 12 The subcommittee met, pursuant to notice, at 8:30 13' a.m.
14 MEMBERS PRESENT:
/~%
( )
-15 DR. ROBERT L.,
SEALE, Chairman, ACRS 16 DR. GEORGE APOSTOLAKIS, Member, ACRS 17 DR. JOHN J. BARTON, Member, ACRS 18 DR. MARIO H.
FONTANA, Member, ACRS l
19 DR. DON W. MILLER, Member, ACRS 20 DR. THOMAS S.
KRESS, Member, ACRS 21 DR. DANA A.
POWERS, Member, ACRS 22 DR. WILLIAM L.
SHACK, Member, ACRS 23 DR. ROBERT E. UHRIG, Member, ACRS 24 DR. GRAHAM B. WALLIS, Member, ACRS
'25 ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202)'842-0034
174 1
PROCEEDINGS
[^}
2
[8:30 a.m.]
i- '
.3 CHAIRMAN SEALE:
The meeting will now come to i
4 order.
i 5
Are you ready over there?
l 6
This is the second day of the 457th meeting of the 7
Advisory Committee on Reactor Safeguards.
During today's 1
8 meeting the committee will consider the following:
proposed 9
Commission paper on options to revise the enforcement 10 policy; proposed final safety evaluation report on 11 Westinghouse Owners' Group topical report on risk-informed 12 in-service inspection of piping; discussion of items for 13 meeting with the Commissioners; and the proposed ACRS 14 reports.
,m
(
)
15 This meeting is being conducted in accordance with 16 the provisions of the Federal Advisory Committee Act, and 17 Mr. Sam Duraiswamy is the designated Federal official for 18 the initial portion of this meeting.
19 We have received no written statements nor 20 requests for time to make oral statements from members of 21 the public regarding today's sessions.
A transcript of 22 portions of the meeting is being kept, and it is requested 23 that the speakers use one of the microphones, identify 24 themselves, and speak with sufficient clarity and volume so 25 that they can be readily heard.
/)
ANN RILEY & ASSOCIATES, LTD.
(,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 i
(202) 842-0034 f
175 1
Today in our letter-w. citing session, which begins
-/.
2 at 3:30 this afternoon, we will be primarily concerned with 3
two topics:
the source-term letter, the report on that we 4
heard yesterday, and the letter regarding the enforcement 5
policy that we're going to hear about shortly.
6 I would also mention that the network is scheduled 7
to crash again for a few hours at 7:30 tomorrow evening, so 8
if you want to get your words into perpetuity, you better 9
plan on making your comments earlier.
10 I hope we won't be here at 7:30, but --
11.
DR. POWERS:
During our 7:30 to 9:00 o' clock 12 session we won't have computer assistance.
13 CHAIRMAN SEALE:
That's correct.
14 Are there any other comments that we should make
()'
15 before we want to begin the meeting?.
11 6 Okay.
The first topic this morning is on a 17 proposed Commission paper on options to revise tne 18 enforcement policy.
It turns out this is a topic which I 19 guess has already been released for public comment, or at 20 least I saw something about in "Inside NRC" this week.
So 21 it's been released one way or the other, even if not 22 officially.
The industry knows about it.
23 It is an attempt to address the concerns for those 24 regulatory actions which perhaps do not contribute 25 significantly to safety but do perhaps impose significant ANN RILEY & ASSOCIATES, LTD.
Court Reporters
. s.
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
. - _ ~ -.._..- -.--.-..-
..-----.~_._
176 1
. resource requirements on the licensees and on the NRC, I 2
might add, to respond to the reports from the licensees.
)
3 I would urge us.to listen very carefully.
The one 4'
concern I have is that while the ultimate objective may be 5
to reduce regulatory burden, the intent should be to provide 6
a redefinition of requirements'which meet the needs of the 7
Commission in its enforcement activities and find out then 8
where the reduction in burden falls.
An arbitrary 9
percentage reduction in burden is the wrong way -- in my H10 mind, anyway -- to approach this subject, and I hope we'll 11
' hear about that in our presentations this morning.
12 I guess we'll begin now with our presentation.
Do 13 you want to introduce yourssif and take us from there?
~
14 MR. LIEBERMAN:
Right.
I'm Jim Lieberman, the LI 15 director of the NRC Office of Enforcement, and it's my L %.)E
- 16 pleasure to be here to talk to you this morning about the 17 changes the staff has proposed to the Commission to deal 11 8
-with the issue of nonescalated enforcement.
19 To set the stage for this I thought it would be 20 appropriate to go over the basic framework of the 21 enforcement policy first, and.then get into the nonescalated 22 actions.
23 I would like to just make a comment too at the 24 beginning.
The objective of the staff and I think the 25 Commission is not to reduce regulatory burden; the objective
(~%
ANN RILEY & ASSOCIATES, LTD.
. (,,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 L
t.
177 1
is to reduce unnecessary or unwarranted re, datory burden.
[/\\I 2
And I think that's very important, because as, you know,
\\ss 3
people are familiar with the Senate hearing we had this past 4-summer and the concern about losing 700 FTEs, and as the 5
message gets translated to various individuals, it seems to 6
be the goal is to reduce burden.
And it's very important 7
that the agency has a mission, and that's to protect the l
8 public health and safety, and we want to do that in an 9
efficient way, in a way that will not create unnecessary 10 burdens to either the licensees or to the staff.
But it's 11 the right burden that we have to reduce.
12 And in the area of nonescalated actions, as I'll 13 be talking about in a few moments, these are what we call 14 the level 4' violations.
In themselves level 4 violations do
(~h ~
(
).-
15 not have a significant consequence or even a potential 16 consequence for the as-found condition.
On the other hand, 17 the level'4 violations are what I call the fundamentals, the 18 things that give us the reasonable assurance, the support to 19 give us confidence that systems work.
So they have an 20 important role in the system, albeit not the same degree of I
21 significance of other requirements.
i 22 Okay.
23 CHAIRMAN SEALE:
Could I just ask one thing, just 24
.to give us an idea of how large these level 4's, which are l
25 really the ones that'are in the spotlight here, loom in the p
ANN RILEY & ASSOCIATES, LTD.
- A
~
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 l
(202) 842-0034
~-
. ~.. - _ _
. -. -. ~.
178 l
l' overall process?
Approximately what fraction of all of the
-2 violations are in the level 4 category?
3 MR. LIEBERMAN:
About 1,400 -- it's not an easy --
4-I don't have an easy answer, and am somewhat embarrassed 5
with that, because -- but we just haven't really kept 6
these-type statistics.
There are about 1,400 or so level 4 7
violations, individual level 4 violations.
i
'8
' CHAIRMAN SEALE:
Per year.
I l
9-MR. LIEBERMAN:
Per year.
We have about 100 i
l' 10 escalated actions.
Now escalated actions may contain more 11 than one violation, so we don't have a one-to-one 12 relationship of how many escalated actions versus how many 4
13 nonescalated actions or how many individual violations are 14 in escalated actions versus the nonescalated actions.
Lp
'15 The rule of thumb I use is less than ten percent l
16 of the enforcement actions are escalated actions, or about 17 90 percent are the nonescalated actions.
18 CHAIRMAN SEALE:
And escalated actions is a code 19 word for 3, 2,
and 1.
20 MR. LIEBERMAN:
Exactly.
And that I'll be getting 21 into.
Okay.
l.
22 CHAIRMAN SEALE:
Okay.
The purpose of the NRC enforcement 1
24 program is to further the safety mission of the agency as a 25 tool that the agency has to use to -- one, to encourage
- l
- [~]'
ANN RILEY & ASSOCIATES, LTD.
&_f Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 L
Washington, D.C.
20036 l'
(202) 842-0034 l.
L
179 1
licensees to comply with the requirements.
We use it as a
()g 2
deterrent that if you're in noncompliance, the agency may
(
3 take action against you, there's a certain degree of 4
aggravation, and the desire to avoid having a penalty, avoid 5
the notoriety, and impact on the money markets, insurance 6
companies, the adverse public attention, and that may 1
7 provide an additional incentive to comply with the 8
requirements.
Obviously licensees have a lot of inherent 9
incentives to comply with requirements, for safe operation, 10 protection of their employees and the public.
But this is 11 another emphasis to remind them to do the job right.
12 Second, and probably more important, is that we 13 recognize that in endeavors as complex as nuclear 14 powerplants, that there's thousands of requirements, there's
()
15 lots of opportunities for failure, and we want to use -
s-
)
16 enforcement to provide an incentive to licensees to identify
-17 and correct violations before there's a significant impact 18 on the public.
And throughout the policy you'll see places 19 where we exercise discretion and provide recognition for 20 licensees to identify their own violations and take actions 21 to correct them.
22 DR, APOSTOLAKIS:
I found that very interesting, 23 what you just said, that there are thousands of 24 opportunities for violation.
Doesn't that tell us something 25 about the regulations?
Should there be thousands of l
I l
/~S ANN RILEY & ASSOCIATES, LTD.
' (,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
L l
180 1
opportunities?
,7 2
CHAIRMAN SEALE:
I think that is what they are Aj 3
addressing here, George.
4 MR. LIEBERMAN:
We want to draw a better
-5 distinction between them, and appreciate that my role is 6
enforcement.
I don't establish the requirements.
My job l
l 7
is, given the requirements, is to have a' structured process i
l L
8 to provide confidence that the licensees are dealing with 9'-
the requirements and, if not, to have an appropriate
(
10
. response to them.
11 And having said that, many of the opportunities l~
' 1:2 for failure is in the area of procedures.
We expect 13 licensees to adopt procedures to implement requirements.
In 14 fact, that is.one of the areas we are looking at for change, r%
(
15
.which of those -- when should we be taking enforcement s
I.
16 action when a licensee fails to meet a procedure, especially 17-in cases where -- if, as part of the corrective action for 18 the particular violation, they delete the procedure.
19 Some procedures, you know, if you fail to meet L
-20 your procedures, you also fail to meet a specific l
l 21 requirement.
In other cases, licensees adopt procedures to i
22 give them greater assurance that they will meet the 23 requirement.
So that is an area that we are looking into.
24 Okay.
When we talk about enforcement, we are 25 basically talking about three things, notices of violations, 4
("'g ANN RILEY & ASSOCIATES, LTD.
(_,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 L
Washington, D.C.
20036 (202) 842-0034 i
i
r 181 1~
civil penalties and orders.
And orders, we have orders to 2
impose civil penalties, we have orders to modify licenses, 3
suspend licenses, revoke licenses.
We also have wrongdoer l
i 4
orders.
And the wrongdoer order is an order to a 5
non-licensee.
(
6 Generally, the agency's requirements and'our i
7 enforcement program focuses on licensees, and licensees are
-8 responsible for the acts of their employees and agents.
The 9'
only time that we take action directly against a i
10' non-licensee, be-it a contractor or employee, other than the f
l 11 area of Part 21, is when an employee or contractor
{
j 12 deliberately violates requirements, when they know something 13-is wrong and they do it anyway, that is the wrongdoer.
14 Other than that, we generally hold the licensee responsible
[
15 for their actions.
16 The first step of the process requires a l-17
' determination'of the safety significance, safeguards and i
18 environmental significance, and I just the shorthand safety 19
' significance here.
When we look at safety significance, we i
l 20 consider three aspects, the actual safety consequence of the
~ 21' particular violation, the potential safety consequence of l
I 22.
the as-found condition, and a third area which we call 23 regulatory significance.
24 When we look at a potential safety consequence, 25.
that is where we consider risk PRA information.
The
.'q ANN RILEY & ASSOCIATES, LTD.
.(,,/.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036-
-(202) 842-0034 lf:
1 182 1
regulatory significance covers three things.
- First,
[
i 2
violations associated with integrity, false information,
's /
3 willful violations of the requirements.
Those are important
'4 because the heart the regulatory process is candor and trust 5
with the licensee.
6 The second area of regulatory significance is 7
associated with impeding the process, not giving us reports, 8
giving us inaccurate information for licensee amendments, in 9
response to requests for information, not doing safety 10 evaluations in the area of 50.59 and other areas, things 11 that will -- can impact the regulatory process, which 12 becomes important in itself.
13 And the third area, and this is the area of 14 controversy and my friends in the nuclear industry probably f-s i
(s-)
15 differ with me on this, as well as UCS.
Is where the 16 consequence or potential consequence of the as-found 17 condition of a violation, or set of violations, are not of 18 great significance in themselves, but the root causes of the 19 surrounding circumstances of a grouping of violations become i
20 more significant than the individual consequences of those 21 violations.
22 An example I like to use is, in doing a 23 modification, they don't put a valve in appropriately.
24 Assume this is a high risk system.
They don't have their 25 hole points right.
They don't do a good post-modification O
/
ANN RILEY & ASSOCIATES, LTD.
'w.-
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
_ ~..
183 1
test.
The surveillance that they do on the tech specs is
! fT 2
weak because they don't have acceptance values, and this
%,,)
3 system is called upon to work, and the flow that comes out 4
of this system is less than that is described in the FSAR, 5
but adequate to preserve its performance safety function 6
and, thus, is~ operable.
So, in that particular case, there 7
is no consequence, and from the as-found condition it was 8
But it was somewhat fortuitous that the system 9
worked and provided the function, and if the conditions were 10 slightly different, it would be a more significant issue.
11 And that is an example we call regulatory significance.
12 Now, this is an area the agency is focusing on.
13 We are working with varicus stakeholders.
We owe the 14 Commission a response and we expect to do this now in early
{
15 1999.
Q) 1 16 But, anyway, so there's three areas that we look 17 at when we consider safety, the actual, the potential and 18 regulatory significance.
19 Now, given a look at significance, the process l
20 then looks at severity level, and we have four severity l
21 levels, severity level I being most significant, level III j
22 being a significant concern and level IV being defined as 23 more than minor a concern, if left uncorrected, it could 24 become more significant.
25 And then we have a category we call minor l
/T ANN RILEY & ASSOCIATES, LTD.
! (,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
1 l
i '
184 l
)
L 1'
violations.
They are so insignificant that it is not worth
(~
2-
'the agency's effort to focus on.
I call them NITS, not V) f 3
important to safety.
Now, there may be a requirement to 4
have'a fence 11 feet high to meet safeguard requirements, t
5 and this' fence is 10 feet high 11 inches, clearly, a 6
. violation, but, in'all~ honestly, the fact that it is one I
L 7
inch short I. don't think is a very significant issue.
4 8
DR. BARTON:
Jim, how are these categorized in the 9?
inspection report, as a minor violation or as a deviation?
10
-How would you see this written up?
11 MR. LIEBERMAN:
Okay.
The policy says that 12 normally minor violations are not to be discussed in an l
13 inspection report.
If they are discussed, they should be l
_14 described as a minor violation.
()
ISE Okay.
Now, levels I, II and III, we call L
16 escalated actions.
We escalate them, I am not really sure L
L17 of the derivation ~because we-have been using this term for i
.18 moz+ Lnan'20 years.
I believe it-is because they escalated 19 from the region to headquarters and: headquarters is involved 20 in the level I, II and III violations.
l l
21 Now, to help us be somewhat consistent when we i
22 apply the policy to, you know, many different types of 23-violations and different activity areas that we regulate, 24 the policy provides examples in eight supplements and I have 25 listed the titles of the supplements.
What might be an 1
i ANN RILEY & ASSOCIATES, LTD.
\\m Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 (202) 842-0034 i
=-
185 1-example ~of a-level I and a' level II, or a. level III or a
(~'i 2
level IV. violation They are somewhat-general, but tLey V
3 serve as a guidepost to help us be consi' tent and the 4
guidance we have issued staff is that if it fits one of 5
these_ examples, you need to coordinate the case with the' 6
Office of Enforcement.so we maintain a degree of 7
consistency.
.8 CHAIRMAN SEALE:
Could I ask, are you going to 9
talk about, just for reference, are you going to identify 10 where the fine fits into this overall structure?
11 MR. LIEBERMAN:
Yes, yes.
I am heading towards 12 that.
13 CHAIRMAN SEALE:
Fine.
All right.
Okay.
14 MR. LIEBERMAN:
Okay.
If it is a level IV
~
'A
(
)
15 violation,_we generally issue either a notice of violation 16 or what we call a non-cited-violation.
A non-cited 17 violation is a violation that we are not citing.
It
'18-occurred -- but I will get to that in a moment.
A notice of 19 violation sets forth the requirements.
Tech spec XYZ 20
' requires you to have three widgets, and then we say contrary 21 to the above, on a given date, you only had two widgets.
So 22 it states the facts supports the violation.
23 In response, the licensee has to reply giving the 24 reasons for the violation, or, if they deny the violation, 25_
- why they believe the violation should be denied.
The
/]
ANN RILEY & ASSOCIATES, LTD.
(,,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
186 1
corrective actions are taken to prevent recurrence and when f(i 2
corrective action will be completed, and when the item was V
3 restored to compliance,_and when the long-term corrective 4
action will be completed to prevent a recurrence.
5 Now, if a licensee has identified a level IV 6
violetion themselves, taken appropriate corrective action, 7
it is a non-recurring type matter, and, normally, if it is 8
not willful, we will' issue what we call a non-cited 9
violation.
The inspection report will describe the event 10 and conclude, it says this violation is not being cited 11 based on section whatever the policy and therefore is being 12 treated as a non-cited violation.
And this is where we are 13 going to be changing considerably the policy, and I will be 14 getting back to that.
. s-~.)
('
15 Okay.
Now,'if it is a level I, II or III 16 violation -- if it is a level III, we consider civil l
17 penalties.
If it is a level I and II, we normally have a i
18 civil penalty.
And we consider four factors, the past 19 performance of the licensee; the corrective action -- What 20' did the licensee do in response to the violation?;
21 identification -- Did we identify it, did the licensee 22 identify it?; and discretion -- after we apply these three
.23 factors, is there -- is this a situation that warrants us 24-taking more action or less action?
25 For example, we may have a very significant safety l
l G(s,/
ANN RILEY & ASSOCIATES, LTD.
Court Reporters
.i.
1025 Connecticut Avenue, NW, Suite 1014
)
Washington, D.C.
20036 (202) 842-0034
1 I
(
187 1
system that we would call a level II violation and the
[b
~h 2
~ licensee did a good job in identifying the violation and L
3 taking corrective action.
It was an old design issue, it 4
doesn't reflect in current. performance.
We might say this 5
is good and not issue any citation.
On the other, hand, we 6
could have the same violation, it might have lasted a long 7-time, a significant risk.
There might have been a number of 8
opportunities to have identified it,-and the licensee missed f
9 it.
We might have a higher civil penalty.
10 This is illustrated in the naxt slide.
I
)
11 apologize for not having color slides.
We call this our 12 metro map.
The first line, the green line, is minor 13 violations.
And, generally, we don't take enforcement 14 action for minor violations, we don't write them in the (A) 15 report.
If_it is a level IV violation, we look at, you
-16 know, identification.
Is it repetitive, corrective action, 17 willful?
And if all the answers are yes, non-cited or, if 18 not, we get a violation.
19-In the escalated process down at the bottom, we 20 first look to see, is this the first level III violation in L
21 a two-year period, or is it a level II or a level III, or is 22 it willful?
If it is the first non-willful level III, then 22 we go with the green line.
And there the only question we 24 ask, is corrective action pru,:pt and comprehensive?
If the l
L
-25 answer to that is yes, then, generally, we would not have a I
O' ks) -
ANN P.ILEY & ASSOCIATES, LTD.
Court Reporters 1
1025 Connecticut' Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 i
I
188 1
civil penalty, just a notice of violation.
If the
- /I..
2 corrective action was not prompt and comprehensive, then we Q) 3-have a base civil penalty.
4 If it is not the first level III violation in a 5
two-year period, or if it is a willful violation, or if it 6
is a level I and II,-we begin by asking, should the licensee i
7
'get credit for. identification?
Did NRC find the violation, 8
or did the licensee find it?
Was the licensee looking for 9
it, or did the violation find the licensee?
If it is 10 appropriate for getting credit for corrective action, we go l
Ell back to the green line and we look at corrective action, and 12 if it is prompt and comprehensive, again, they won't get a 13 civil penalty.
If it wasn't, then they would get a base 14' civil penalty.
!()
15 If the licensee should not get credit for L
.16 identification, we found it, and we found it and licensees 17 had an opportunity to find it first, because sometimes we l
18 are just lucky, and it is not necessarily that it is 19
. appropriate.for us to get credit for that, but if we 1
20
-conclude that the licensee should not get credit for it, we 21 go to the red line and we look at corrective action.
- And, 22 again, if it is prompt and comprehensive, they get a base i
a 23 penalty.
If it wasn't prompt and comprehensive, they get a 24 double base penalty.
(
25; Now, there's little "D's" with circles here, and 1
- ("')
ANN RILEY & ASSOCIATES, LTD.
(_ /
Court Reporters c
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
- -. _. ~ - -. _ _.. -. -. -.
189 1
that is discretion.
So in each of these decision points, we
/j 2
have the opportunity to exercise discretion to make sure we
~QJ 3
are taking the right' enforcement action.
For example, if 4
you have a series of level III's, and each time the licensee 5
identifies it and the same root cause, meaning that -- and i
6 let's say in the past we haven't given them a civil penalty 7
because we thought they did a comprehensive job in 8
corrective action, at some point you say, well, you know, 9
the corrective action just isn't being thorough here, they 10 are not responding to opportunities, and we would conclude
.11 that a civil penalty might be warranted.
And so the policy 12 provides various reasons to exercise d'iscretion.
13 CRAIRMAN SEALE:
Now, just for emphasis, this is
.14 the way it is now?
I15 MR. LIEBERMAN:
That's right.
This is the normal t
16 policy now.
Now,.what I want to switch to, the Commission 17 paper.
You have a-draft -- you have a copy of the draft 18 Commission paper, I believe it was signed out -- I know it 19 was signed out on Monday, so a formal paper should be coming 20-through the system any day now.
Okay.
21 There are clearly issues with non-escalated 22 actions.
The number.of Level 4 violations have 23 substantially increased in the last year or so.
We believe 24' that increase is caused as result of a reaction to the 25 Millstone situation and situations at Clinton and Crystal i
i O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
190 1
River, Maine Yankee, where we probably should have been
(g 2
doing a better job in focusing on noncompliance.
V 3
Maintenance team inspections, A&E team 4
inspections -- there's various reasons why the number of 5
compliances have increased.
6 At the same time we recognize that increase in 7
number of violations is not consistent with the performance 8
improvements in the industry as a whole.
If anything, I 9
think what it tells us is we made an abrupt change in policy 10 on treating non-compliances, that if we had done that back 11 in, say, 1989, the violations would have been a lot higher 12 in 1989 and we would have seen a gradual decline or a 13 decline in the number of citations reflecting the 14 performance of the industry, but the fact of the matter is
[)
15 there was a large increase in the number of violations.
1
)
16 We recognize that there's a cost to responding to 17 violations.
Obviously you have to take corrective action, 18 but there is a certain cost in preparing paperwork, doing 19 the reviews.
I have heard industry folks say that that cost 20 can be ten, twenty, thirty thousand dollars per response.
j 21 We also recognize NRC fines are only a small 22 percentage of the fines in the licensees' corrective action 23 programs and they are not always the most important items in 24 the licensees' corrective action programs and that creates a 25 potential safety issue in itself because licensees are 1
/s ANN RILEY & ASSOCIATES, LTD.
(_,!
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 l
t
191 1
required by our regulations to respond to a notice of 7Y violation in 30 days.
If they desire to have credit and get N
2 N~
3 a non-cited violation, they have to develop the corrective 4
action and present their plans for corrective action before 5
the inspection period is over, and that requires the 6
licensee in some cases to divert effort from maybe more 7
significant things to the NRC issue.
The NRC issue almost 8
creates an artificial priority for responding to the issue, 9
rather than having the safety significance of the underlying 10 issue set the standard for when a corrective action should 11 be taken, so that is an issue that we are concerned about.
12 We recognize that not all requirements have the 13 same degree of nexus to safety.
That is another issue.
14 Finally, we feel it is important to integrate
()
15.
inspection assessment and enforcement.
They are not 16 separate fiefdoms, each group doing different things.
There 17 should be a greater nexus between them.
18 CHAIRMAN SEALE:
Let's say a utility finds a 19 problem.
Do they generally go ahead and suggest a 20 corrective action program and all of the other things that 21 are involved before they bring it to your attention, or do 22 they let you know what the problem is immediately and then 23 go away and do their corrective action planning and of that 24 sort of thing?
25 MR. LIEBERMAN:
I think in most case the
/G ANN RILEY & ASSOCIATES, LTD.
(,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
192 1
licensees' corrective action programs are generally working 2
without our involvement.
They have a variety of dif'ferent 3
types of corrective action programs.
Criterion 16 of 4
Appendix B requires a corrective action program for 5
safety-related items.
6 There are license conditions that require in many 7
cases corrective action-programs, say for fire protection 8
and some other areas, and even when there's no requirements 9
most licensees put all types of deficiencies into corrective 49 action programs and they identify them.
They prioritize 11 them.
They deal with them without our involvement.
12 Now on top of that we may. find some issues, and we 13 bring it to their attention, and then they add those to 14 their corrective action program.
(
15 CHAIRMAN SEALE:
Yes.
Those are the ones where 16 the schedule can. bite you then?
4 17 MR. LIEBERMAN:
Exactly -- and okay, so what have 4
18~
we done to look into this issue of non-escrilated actions?
19 The first thing we did was we arided some formality
'20 to the appeal process.
In the past, licensees appealed just 21 to the Regional Administrator based in the same office that 22 issued the violation and we changed the process to provide 23 that a copy of the appeal should be sent to my office so I 24 can make sure the response is coordinated with the l
25 appropriate program office.
l l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters s
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
?-
1 193 1
We found out that in the year or so that we have f')).
2 been doing this about half the appeals we have modified in 3
favor of the licensee's position, either based on the facts 4
or the requirements or the sanction, a small percentage of 5
the overall cases, but that at least has helped to add some 6
rigor to the process.
7 The second issue, and probably a very significant 8
- item, is we issued an Enforcement Guidance Memorandum.
9 My office is responsible for keeping the 10 Commission's enforcement policy.
We have the enforcement 11 policy.
We also issue what we call the Enforcement Manual 12
'that provides assistance to the regions and guidance in 13 being somewhat consistent in applying the policy and 14 standard formats and things of this sort.
s(q 15 In between issuance of the manual, supplements to 7
l 16-the manual, I issue a memorandum which I call Enforcement l-17 Guidance Memorandums to provide the staff guidance on how we l
'18 should be enforcing our requirements.
19 EGM 98-006 was issued in July of this year.
It i
20 did not really create any new requirements.
It basically
~21 emphasized the existing policy and we made it clear that we 22 wanted the existing guidance to be met.
23 The first thing we focused on -- making sure 24 non-cited violations, the guidance for non-cited violations 25 were followed, meaning the licensee identified and corrected 4 /h ANN RILEY & ASSOCIATES, LTD.
\\,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
194 1
the violation, it wasn't recurring, et cetera, which would
[}
2 give credit to the licensee.
L./
3 If we already had the information on the docket 4
from an LER or report or it was in the inspection report,
)
5 there's no need to ask the licensee to give us another 6
formal response.
That is part of the unwanted, unwarranted 7
burden.
No sense giving the response if we already have the 8
information in the docket.
9 We told the region, which is consistent with i
10 existing guidance, but we never said it so clearly, that we 11 didn't want our inspectors to be mining the licensee's 12 corrective action program for violations.
What that means 13 is that if the way we discovered the violation was just 14 searching through records in the licensee's corrective
~s l I I
15 action program, and then restating it as an NRC violation,
<_J 16 that's not appropriate, but if you are at a morning meeting 17 and you find out that last night there was valve had a 18 problem with the licensee identified and then you go do an l
19 inspection and that issue is a violation, that would be i
i 20 appropriate for a non-cited violation.
l 21 If you then go into the records of the licensee to l
22 find out whether this is a recurring issue, and you find out
[
l 23 that four months ago that same violation had occurred but l
24 the corrective action was inadequate -- for example, let's 25 say the only thing the licensee did was restroke the valve,
(~N ANN RILEY & ASSOCIATES, LTD.
( _)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
195 1-they.hadn't recognized they had the wrong lubricant and that
(
2 was really the cause, so the inspector discovered that or 3
identified there was inadequate corrective action it would 4
'be appropriate to issue a citation for that inaccurate 5'
corrective action.
It would not be appropriate to issue a 6
violation for that valve failure that occurred two months 1
7 ago because that the licensee identified.
8 If going through the corrective action records you 9-also discover three or four other violations which the 1
10 licensee had identified in similar areas and taken 11 corrective action for, again that would not be a basis to 12 take enforcement action.
13 A fourth item we gave guidance on is when you have
'14 similar examples of the same violation -- a similar j )
15 violation with the same root cause/causes, we have said 16 various things over the years.
Our current guidance is we 17 should group those violations into one violation with 18 several examples.
19 Finally, we said that the standard for corrective 20 action is reasonableness.
Our inspectors can always come up 21 with another thing the licensee can do and the standard is 22~
it should be reasonable commensurate with the safety 23 significance of the issue.
24 Well, our current data shows that these changes
'25 have resulted in a lowering of the number of violations --
ANN RILEY & ASSOCIATES, LTD.
t
,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
____g.
_.__m l'
196
- L strike that.
L
~T 2
[(Q The number of' findings that our inspectors have o
3.
identified,.'the number of notices of violations that we have
{
l 4
- issued, and the number of notices of violations that require 5
a response -- it's still premature because we have only been is 6
taking' data for two months on this to form a view as to 1
7-where the trends are going, but they appear to be in the 8
right direction.
I
'9
.Now the philosophy of the new approach is that we
}
(
~10'
' recognize that Level 4 violations are not in themselves a i
'11.
significant safety concern.
They are important.
They need 12 to be addressed.
They go to the fundamental issues in many 13 cases, but they are not of a significant consequence and we l
14 also recognize,.as I said a few moments ago, that licensees
+
l./~\\
(
)
15
' generally have' forward corrective action programs'and that 16 generally they are effective, therefore doing the process l:
17l
_that we have been generally doing, requiring a response, may p
18.
be an unwarranted burden.and may' interfere with the safety 19 priorities of licensees for the reasons I have stated.
l=
20 We have been using more or less the same L
21 enforcement process since. prior to TMI and clearly the
' 22' performance of licensees has increased substantially since
- l..
23 that time.
V 24' We want to maintain focus on compliance, 25 compliance is expected, but we want to increase our
[
j' 5
A ANN RILEY & ASSOCIATES, LTD.
k,,)
Court Reporters 1025 Connecticut Avenue, hW,' Suite 1014 j-Washington, D.C. 20036 L
(202) 842-0034 i
l
l-197 1
attention on the overall effectiveness of corrective action l[ l 2
because we want to make sure licensees address the 3-precursors before more significant events occur.
We want to 4
integrate with inspection assessment and we want to make 5
sure the changes that we are proposing now to non-escalated 6-enforcement won't prejudge.the outcome of the integrated 7
oversight efforts that NRR is working on, and we think we L8 need to preserve the ability to track and trend 9
noncompliances to assist us in determining adverse 10 performance issues.
11 What we plan to do is.have three types of 12 violations, as we do today, violations which we issue the
.13 normal enforcement actions with, the noncited violations 14' that are written up in inspection reports but won't require T
)
15
? response, and.the minor violations which are not normally 16 written up'in inspection reports.
17 Level'4 violations would generally be considered
. 18 noncited-~ violations, regardless of who identified it,
- 19 regardless of whether it's recurring.
The major key here is 20 thatLthe licensee has to provide us evidence that they put the matter into their corrective action program.
So if the 22 licensee puts the matter into the corrective action program to prevent recurrence, we will close the matter out as a 24 noncited. violation, and we generally won't inspect the 25 corrective action for.each individual level 4 violation, but ANN RILEY & ASSOCIATES, LTD.
I./
Court Reporters s
1025 Connecticut Avenue, NW, Suite 1014
-Washington, D.C.
20036 (202) 842-0034
....s
...........r t
i i
198 1
rather will take a sampling of these noncited violations, as
(
)
2 well as a sampling of licensees' violations or deficiencies
~ hich they. identify to look at the overall effectiveness of 3-w 4
the corrective action program.
5 Now we would issue a level 4 noncompliance, and 6
this is an area that we have differences with the industry 7
on.
I'm sure NEI will address this later on in their 8-presentation.
The four areas where we think a level 4 9
violation should be issued even though it was just a level 4 10 violation:
one, if the licensee doesn't take acticn to 11 restore compliance within a reasonable time.
That's 12 basically allowing the violation to continue.
The valve's 13 in the wrong position.
You have to put the valve in the 14 right' position.
The procedure is inadequate.
Before you
()
15 use that same' procedure again, whether it's tomorrow or next 16 week or next month, or the next refueling, you have to fix 17 that procedure.
So you can't allow the violation to 18 continue without abating it.
That rarely happens, but it 19 does occasionally.
20 DR. POWERS:
In the language on your slide you 21 have this phrase " commensurate with safety," and you had it 22 several times in your presentation.
What the words say 23-literally is that the action when measured has a measure 24 that's equivalent to a measure of safety.
How do you do 25 that?
-O.
ANN RILEY & ASSOCIATES, LTD.
$mj7.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
199 1
MR. LIEBERMAN:
Obviously there is judgment, and 2-hopefully the words are not misleading.
3 DR. POWERS:
I mean, it says --
4 MR. LIEBERMAN:
Something is in the wrong
-5.
position.
Do Lu fix it immediately?
Do you fix it, you 6
know, the next day?
You have to fix it as soon as
'7 possible
- as soon as practical is probably the better l
8 term.
Sometimes you may have to take some compensatory 9
measures in the meanwhile.
You know, you may not have your 10 replacement valve, you know, fire barriers, it takes time, 11 and you post fire watches.
12 DR. POWERS:
You're giving me examples of how a 13 situation can be corrected, but this phrase " commensurate 14 with safety" is intended to say something about the relative
()
15 ordering of these things.
It has implied in it quantitative 16 measures, which I don't think you have, so I'm trying to 17-understand how one sits down and looks at, you know, an 18 action and decides in a way that other reasonable people 19 would agree with that that action has or has not been 20-commensurate with safety.
' 21 -
MR. LIEBERMAN:
Ummm --
22 DR. APOSTOLAKIS:
Do you actually mean if the 23 licensee takes actions according to the safety significance
[
24 of.the violation?
25-MR. LIEBERMAN:
As we discuss this, I'm concerned O-ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025. Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036-(202) 842-0034
--.. - - - =
200 1-that the phrase may be a little ambiguous here.
We don't
/
2 (s) want the licensee to make his decision or make a judgment and say well, you know, this is not very safety-significant, 1
4 this is 10 to the minus 7, we know the valve is in the wrong 5
position, but we're not going to fix this for another three 6
months because, you know, it really is not very 7'
safety-significant.
We would consider that willful 8'
noncompliance.
9 The licensee has to comply; the licensee cannot-10 normally continue in noncompliance.
They have to take 11 action to fix things.
At the same time, everything can't be 1
12 fixed in a nanosecond, you know.
It takes time.
And so the 13 licensee has to take reasonable action to restore I
14 compliance.
IT 15
'Q)
And there may be some judgment in there.
Do you j
16 fix it this immediate second or can you fix it in a few 17-hours?
Do you have time to get a notice of enforcement 18 discretion from NRC?
Should you be seeking an exemption?
19 Do you fix this procedure today?
Well, I'm not going to be using this procedure until next month or the next refueling.
20 21 Well, considering the safety significance then, there's 22' really no need to fix this procedure today, you can wait 23 till a month before the refueling to fix it.
24 That's really the thought here.
So maybe I have j
25.
to focus on these words a bit more.
We don't want a P
ANN RILEY & ASSOCIATES, LTD.
- (,,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i
Washington, D.C.
20036
.(202) 842-0034 s
201 1
licensee to know there's a violation that is continuing and
' t'"'\\
2 not abate it.
I appreciate you raising that issue.
h]
3 DR, POWERS:
I would be interested to see how you l
4 look at the words'but also understand in thinking about 5
those words that when a judgment is made by the NRC of the 6
relative speed with which compliance should be restored, 7
that it's very useful -- very desirable that if other i
8 reasonable people looked at that judgment, they would agree.
9 MR. LIEBERMAN:
Well, that's a good test of
{
10 reasonableness.
11 DR. POWERS:
Yes.
12-MR. LIEBERMAN:
And this has not been a problem --
13
'this has been inherent in the enforcement policy since 1980, 14 at least 1980, where licensees have two types of corrective
().
15 actions.
The.first corrective action is to restore 16 compliance with a particular item.
The second part of the 17 corrective action is to prevent recurrence.
And rarely if 18 ever do we have an issue that goes to the first part.
19 DR. POWERS:
Right.
20 MR. LIEBERMAN:
So this is something that 2
21 inspectors and licensees using judgment have worked this 22 out.
So it has not been controversial.
23 DR. POWERS:
It's really the second step.
24 MR. LIEBERMAN:
Right, which is the cost and the-25 focus.
(,,_)N ANN RILEY & ASSOCIATES, LTD.
Court. Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
[
l 202 1
DR. POWERS:
Yes.
2 MR. LIEBERMAN:
Okay.
3 DR. APOSTOLAKIS:
I have a comment that's a bit 4
more general, and perhaps it would not be appropriate to be I
5 addressed to you, but I have the suspicion as I follow your 6
presentation that a lot of these difficulties have a 7
fundamental cause, which is not your problem, and the l
8 fundamental problem seems to me to be the fact that we have 9
regulations that go down to details that perhaps we should 10 not be regulating.
i l
11' Then what do you do?
They violate something.
12-It's a major industrial facility.
They have to operate.
L 13 They violate a few things.
As you say, they have priorities 14 they have to set and so on.
And yet these are violations.
r%
( )
15 In the books they're violations.
l.
16 If you asked anybody, you know, if I have l
17-regulations should they comply, sure they should comply.
l, i
j 18 But then reality comes in, and then you have the burden of i
19 having escalated actions and enforcements and so on.
In f.L 20 other words, the system now says well,. gee, you know, not 21 all of these were important, so now we have to prioritize.
{
22 And what I'm asking is why. don't we do that before 23
. we declare these things violations.
Why doesn't the system 24 say, for example, as it says at some very high level, that 25 there are certain things that are regulations that you have ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 l
I l
203 1
to comply with and there are certain other things that are
(~'/)
2 mere recommendations?
\\_
3 CHAIRMAN SEALE:
Well, in a sense, George --
i 4
DR. APOSTOLAKIS:
Like a goal, a goal, the 5
. critical-word is a goal.
If you're above the goal, you're 1
6 not committing a crime, but on the other hand it's something 7
that you have to L:1. ink about, and that way a lot of your 1
8 burden will be taken away.
9 MR. LIEBERMAN:
Well, there's no question it would 10 make my life easier.
11 DR. APOSTOLAKIS:
Who is the person or group that 12 would make your life easier so we can ask them why they 13 don't do this?
1 14 MR. LIEBERMAN:
Okay.
The NRR is looking at more
~()
15 risk-informing the regulations and focusing on which 16 requirements are more important than others.
17 DR. APOSTOLAKIS:
So we should ask them.
18 CHAIRMAN SEALE:
Well, I wasn't going to bring it 19 up, but now that you've opened it, I have a couple of 20 questions to ask you.
As I read through your draft paper 21 and so on, the words " risk-informed" showed up several times 22 in the writeup, and yet whenever I looked at your proposed
.23 new approach and so forth, I didn't see where there was Ony 24 significant new risk analysis.
25-MR. LIEBERMAN:
That's right, Chairman.
And the
' f's, ANN RILEY & ASSOCIATES, LTD.
w /.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
~.
l 204 l
1 final paper won't have those words in there.
l [~}
2 CHAIRMAN SEALE:
Well, that's interesting, because l O I
3 that means I'm going to have to modify my draft letter then.
4 MR. LIEBERMAN:
And the rebson for that, it might 5
be in there once, the risk benefit of this approach is not 6
really from the policy, because what's a level 4 isn't l
7 changing, but rather from the point of view of the licensee 8
on having these violations and the priority they have to 9
deal with the violations from an overall safety risk point 10 of view with their other issues.
We're not taking l
11 licensees' management time away from addressing this issue, 12 assuming they're addressing this issue of more significance.
13 So they have to focus on the NRC issue first.
That's a 14 safety benefit.
,s (v) 15 CHAIRMAN SEALE:
Yes.
16 MR. LIEBERMAN:
Or the risk benefit.
So the 17 impact may be safety -- I think it contributes to a more 18 risk-informed regulatory process.
The policy itself doesn't 19 really address risk per se.
20 CHAIRMAN SEALE:
Well, we're going to have to move
'21 along here --
22 MR. LIEBERMAN:
Okay.
23 CHAIRMAN SEALE:
But I do want to -- the way I 24 read it is the risk analysis that's done now is essentially 25 the process by which the inspector, the risk specialist in
[^)\\
ANN RlLEY & ASSOCIATES, LTD.
'\\_
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 1
- m. _ _ _... _ _.
205 1
the region, and the regional administrator make and then
()
2 concur in the decision that something is a level 4 as V
3 opposed to a level 3 or a noncited in the old mode.
And 4
'that's the level of the risk analysis that has been done, 5
and in essence what you're really saying is that there is 6
another element of discretion now based on these criteria 7
that are on this slide, that is, whether or not it's a 8
violation because of those last four willful and recurring 9_
noncompliance and those kinds of issues.
10 MR. LIEBERMAN:
Yes.
They'll always be 11 violations, but whether'we issue the notice of --
12 CHAIRMAN SEALE:
The notice of violation.
That's 13 right.
14 MR. LIEBERMAN:
And that's going to result, and I
()
15 would presume if we have 1,400 violations or notices of I
16 violation today, it will get down to maybe 100 --
i l
l 17 CHAIRMAN SEALE:
Okay.
18 MR. LIEBERMAN:
Or some very small number.
At the 19 same time, one of the items we have in the tasking memo is 20 focusing the policy more on risk.
We modified the policy a 21 year or so ago to emphasize that risk is an appropriate l.
22 consideration.
We've always considered risk, but this makes 23 it more formal, both to increase and decrease the severity l
_24 level.
l
~25 The policy wasn't that clear about the decreasing
[]-
ANN RILEY & ASSOCIATES, LTD.
' (s l Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
206 1
of severity levels.
In the main we've modified that to make
)
it clear that we can go both up and down, and we do that.
A 2
3 hydrogen recombining system not being functional would be a 4
level 2 violation by the policy.
But becetse of the role of 5-the hydrogen recombining system we lowered the severity 6
level to a 3 or 4 depending on the circumstances.
l 7
In other cases we might increase it.
8 DR. POWERS:
How do you know?
I mean when you say 9
gee, it is a hydrogen recombiner system and I know that 10-that's not very risk-significant, how did you know that?
11 MR. LIEBERMAN:
Well, I'm getting input from 12 others, from PRA point of view, from --
13 DR. POWERS:
When you go to the PRA, is it PRA as 14 in the great god PRA, the conventional wisdom that exists
' f")
t 15 out here, or is it the PRA for this specific plant?
v 16.
MR. LIEBERMAN:
A combination of both.
We have 17 SRAs in the region who give us input.
We have people here
~18 in NRR who glve us input.
For example, there's a case we 19 just issued, enforcement notification, yesterday involving 20 San Onofre, where we -- I don't have all the facts in my 21 head, but we lowered the level from a level II to a level
- 22 III using --
23 DR. POWERS:
I believe we have gotten that, a 24 description of that particular incident here.
25 MR. LIEBERMAN:
And so we are using these
'Pg ANN RlLEY & ASSOCIATES, LTD.
k/
Court Reporters s
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
.-ar
207 1
judgments every day.
But policy clarifies that.
What we
(~')
2 want to do is, in the supplements, in the examples, have N>
3 better examples of risk-informed issues.
We are not looking 4
for numbers, but better guidance to make sure that when we 5
have an escalated action with a civil penalty, or if we have 6
a civil penalty, that those cases and the more safety 7
significant cases.
8 DR. POWERS:
Now, when we go into the appeal 9
process, suppose that you said, based on risk, this is a --
10 I am not going to raise or lower it.
You came neutral on 11 it.
And the defending party comes in and says, no, I have 12 done my risk assessment and I think you ought to really 13 lower this because this is a relatively low risk 14 significance.
Is that a basis for appeal?
if'v')
15 MR. LIEBERMAN:
That would be a basis for appeal.
16 I kind of fear the litigation of that.
17 DR. POWERS:
Yeah, I can see we get a combination 18 of -- combat between experts, you will never resolve this 19 thing.
20 MR. LIEBERMAN:
But one of the examples that we 21 have in the policy as a level III is when you have equipment 22 of indeterminate quality where you need a substantial effort 23 to determine whether it is operable and the impact of that, 24 and it is fortuitous you are in compliance, you know, 25 there's judgments.
We will never get rid of judgment.
We I
/~'\\
ANN RILEY & ASSOCIATES, LTD.
(_,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 l
(202) 842-0034
I 208 l
1 don't want to have a prescriptive enforcement system and the
/N 2
answer is either A or B.
In enforcement lore, 1-o-r-e, in
- d 3
different government agencies, in the criminal process, when 4
sanctions are very prescriptive, it never works.
So you
~
5 always have to have that room for judgment.
And I think I
6 licensees should appreciate.that, too.
7 DR. APOSTOLAKIS:
Maybe also when inspections are 8
very prescriptive, it never works.
9 CHAIRMAN SEALE:
In that regard, one of the
.10 concerns I had in regarding through ti.is, too, is that you 11 are modifying a process that a lot of piople have had 12 experience with in your inspection corps and so on.
And it 13 is my impression that when you try to modify behavior on 14 something that is pretty well ingrained that general
()
15 criteria sometimes are not very effective in gaining the 11 6 attention, if you will, 17 It would seem to me that some examples, 18 specifically, of items that are on the borderline within 19-level IV as to whether they should be, in fact, an NOV 20 issues or whether it is considered to be a non-cited 21 violation kind of activity, would be very helpful mostly --
22 at least at first, to the inspectors and to the regional 23 people.
Because general words sometimes mean different 24 things to different people.
And then you get into this 25 question of inter-regional consistency and things like that.
l l
l
,((
ANN RILEY & ASSOCIATES, LTD.
Court Reporters t
1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 (202) 842-0034
~
. _. -.. - -. ~ -.... - ~ -
209 1
So I would urge you to try to come up with some 2
examples for people that.they can use to help them in making i
3 that distinction.
?
4.
MR. LIEBERMAN:
Now, that is a good point.
In t
5 addition to what we plan to do, is require all notices of 6
violations to be reviewed by my office.
1 7
CHAIRMAN SEALE:
It would help you then.
l 8
MR. LIEBERMAN:
And that would help us develop 9'
those examples, too.
10 CHAIRMAN SEALE:
Sure.
111 MR. LIEBERMAN:
Because we are serious about doing 12 this, and we are changing a mindset.
We are changing a 13
. practice --
14 CHAIRMAN SEALE:
Yes, exactly.
()
15 MR. LIEBERMAN:
-- that inspectors have used for a 16 long time.
17 CHAIRMAN SEALE:
And that process may provide a j
11 8 skimming, if you will, of those borderlines and sharing that 19 in'a1useful way could be very helpful.
20 MR. LIEBERMAN:
It is not going to be easy to 21 change a. system.
But this is a very abrupt change, and many 22 people favor this because it is going to save us resources,
'23
.as well;as licensees, and we can use our resources, rather 12 4.
than drafting citations, out doing inspections.
.25 CHAIRMAN SEALE:
Yes.
ANN RILEY & ASSOCIATES, LTD.
(_f
. Court Reporters p
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
..--. =
. ~ ~ _ ~. -... - - - - - -. _ -...
l l
210 1
MR. LIEBERMAN:
And focusing effort in the more 2
.important areas.
So a lot of us are very positive about 3
this.
It is still --
4 DR. BARTON:
Before you leave, there is a 5
statement here, "Use smart sample in inspecting licensee's 6
corrective action program."
7 MR. LIEBERMAN:
Yeah.
8
'DR. BARTON:
What do you mean by smart sample?
9 MR. LIEBERMAN:
Okay.
We are working with NRR and 10 they'have.the lead in developing the inspection guidance.
11 But let's say there's 12 non-cited violations at a facility 12 over the past two years.
We may be taking four or five of 13 those and then looking for similar type infractions the 14 licensee has identifi'ed in those areas, and maybe taking
()
15 twice as many of the licensees and-focusing on those rather 16 than every one-of ours.
As a sampling, smart was just --
17 DR..UHRIG:
I wondered if'that anything to do with 18 risk.
19 MR. LIEBERMAN:
We want to, in the sampling
=20-process, we want to sample those which are more significant 21 from a risk point of view.
But there may be some less 22 significant that we will look at, too, from the overall 23 series of requirements.
Failure to do post-modification 24 testing.
In itself, that may not be risk-significant, but r
25.
if there is a pervasive problem of not doing
, (N.
ANN RILEY & ASSOCIATES, LTD.
- N,)
Court Reporters s
L 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 4.
211 1
1 post-modification testing, I mean that is important to give
(~)}
us reasonable assurance that design work was done right and
~
2 3
installation and engineering work was done right.
And we 4
need to have robust programs in those areas by licensees to 5
give us -- give them confidence, as well as give us 6
confidence.
So is that really risk-significant?
I mean one 7
could argue, but it is important to make sure that is being 8
done.
9 Okay.
We know that the area of minor violations 10 is a controversial area, but we think it is important that 11 we should continue using minor violations because it helps 12 us to focus our inspection away from things which are less 13 significant.
We know we have to develop inspection guidance 14 to be consistent with this new approach, and making sure our s
[w) inspection effort focuses not only Appendix B issues but 15 16 other areas, transportation, health physics, safeguards, 17 other areas that legally Appendix B doesn't apply to.
18 We recognize that non-compliances may reoccur 19 because the licensee hasn't dropped everything to fix this 20 particular to prevent recurrence, because from a safety 21 point of view it is not worthy to do it this week.
Maybe 22 doing it next month or six months from now is appropriate 23 given everything that is on the licensee's plate.
We are 24 going to document the non-compliances in inspection reports 25 so that when someone looks at the issue six months from now, i
[~)]
ANN RILEY & ASSOCIATES, LTD.
x_
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
i 212 1
nine months from now, if they select that particular issue
(T 2
to sample, they will understand what the concerns were at V
3 the front end.
4 We plan to continue to use the plant issue matrix 5
to trend and track non-compliances.
We are going to provide 6
an appeal process.
Initially, we are going to only apply 7
this to reactors, because they have the robust corrective 8
action programs.
9 We are going to continue our efforts to develop 10 thresholds between level IV's and minor violations.
We want 11 to have this process driven by the assessment needs of the 12 agency and -lot the enforcement needs of the agency, because 13 in the past, enforcement drove many of the inspection areas.
14 We want to have the overall assessment process drive it.
rr (V) 15 We are going to be reconsidering the issue of 16 citations based on procedures, which I talked about at the 17 beginning.
We want to integrate -- I don't know if you have 18 had briefings on the discussions of changes to the oversight 19 program with using performance indicators and the green band 20 and white and red band.
We want to have greater consistency 21 between those activities and the enforcement activities, not 22 using enforcement as a consideration for assessment, but, 23 rather, having the results of the assessment process 24 influence enforcement.
25 And I guess the last slide is just a summary.
We
[~T ANN RILEY & ASSOCIATES, LTD.
i
(_,l Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i
Washington, D.C.
20036 (202) 842-0034
213 1
think this will reduce unnecessary burden.
This addresses
()
2 the risk issue by having the impact, the result more V
3 consistent with safety and risk based on the timing on the 4
licensee's actions, and it continue to permit us to carry 5
out our oversight responsibilities.
I guess there is a typo 6
there.
Because the bottom line is unnecessary burden, and 7
we don't think the agency loses its ability to do the job by 8
making this change, and it can save resources for a lot of 9
folks in the meanwhile.
Any questions?
10 DR. POWERS:
I congratulate you on a fine 11 presentation.
I certainly enjoyed it.
12 What I am a little distressed about is I got the 13 impression that the words we have read now have been 14 changed.
We have a document here and we have looked through
!q s
15 it now.
That document has been changed.
Do we have the
\\J 16 corrected document?
l 17 MR. LIEBERMAN:
I presume ACRS gets copies of all 18 the Commission papers.
Is that a fair statement?
If you 19 don't, I'll make it happen.
20 CHAIRMAN SEALE:
The problem I have is I mentioned 21 earlier that the words -- the documents I had had several 22 references to ri sk-informed and so on, and it would be 23 worthwhile to double-check to see if in fact there is a 24 later version that we might want to be able to look at in 25 order to -- see, I have a draft signed by Mr. Travers --
/ ~'
ANN RILEY & ASSOCIATES, LTD.
k,)N Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 m
214 1
MR. LIEBERMAN:
I'll send the latest version over 2
to Noel -- I'll leave with Noel a copy of the final --
3 CHAIRMAN SEALE:
Okay, fine.
4 DR. POWERS:
Because if go through searching for 5
specific quotes or something or basic information and we i
6c get --
7 CHAIRMAN SEALE:
Yes.
We don't want a tempest in l
.8 a teapot.
9 MR. LIEBERMAN:
This paper was on a fast track 10-
- and --
11.
CHAIRMAN SEALE:
Sure.
12:
MR. LIEBERMAN:
-- as you can imagine, and I had a 13 lot of help, plus we had two different EDOs.
We wrote one 14 paper to satisfy one EDO and then we had to do a crash
()_
15 change to satisfy the next EDO.
16 The basic thrust of the paper stayed the same but 17 some words were changed.
. 18 DR. POWERS:
The ACRS had agreed to help get these 19 things through on a fast track.
One of the problems we'll 20 face is that what we review before the meeting may have been 21 changed.
We just need to make sure we have the right one, 22.
that's all.
23 MR. LIEBERMAN:
I appreciate that.
No problem.
-24 CHAIRMAN SEALE:
Any other questions?
l
. 25 DR. APOSTOLAKIS:
Well, yes.
I am still i
f3 ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 I
s h
g
215 1
struggling to understand something, which again may not be
/~%
2 U
within your powers to change but I would like to have your 3
reaction to it.
4 You mentioned as an example earlier the height of 5
the fence.
Do you have any other examples of such detail?
6 I mean I have seen in the past a NUREG that recommended that 7
the length of a cord of a telephone should be 27 inches --
8 CHAIRMAN SEALE:
That's voluntary so it's --
-9 DR. POWERS:
That is a different thing than a 10 regulation'that says that it is 11 feet high.
I don't know 11 that there is an 11 foot high regulation, but there are 12 things like that in the regulations.
- 13 DR. APOSTOLAKIS:
So there are regulations then 14 that are real regulations at lower level and then there are
(
):
15 what? -- recommendations or suggestions?
16 CHAIRMAN SEALE:
Well, NUREGs.
i 17 MR. LIEBERMAN:
Well, take 50.46.
The cladding 18 has to be 2200 degrees, less than 2200 degrees.
19-DR. APOSTOLAKIS:
Yes.
l 20 MR. LIEBERMAN:
Let's say it is 2201 degrees.
I 21 mean whenever you have a limit, there is always going to be 22 a slightly over that limit -- the nature of the process.
23 You are speeding down the highway.
The limit is 24 60 and you go 61.
A policeman exercises judgment.
25-.
You have to have a response, you have to be able (g
ANN RILEY & ASSOCIATES, LTD.
'()
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 (202). 842-0034 l
....m 216 1
to make a notification when the 15 minutes of a certain type l
2 of emergency.
You make it in 15 minutes and 10 seconds.
I 3
mean judgments are made and'the cost per minor violation is i
4 to allow the inspector to exercise some of that judgment and 5
not be afraid that the IG is going to come down on him and, 6-boy, you know, you cross the T and dot the I.
7 It also allows the regional management to focus 8
the inspector on more significant issues.
Our process-is
'9' complex.
It's lots of issues in it.
Our licensee needs
~
t 10 frequently procedures to be able to implement requirements.
11 For us to be satisfied that the licensee is doing 12-the job right so we can have the requisite reasonable i
13-assurance we need confidence procedures are being done and 14 hold points, records are kept.
We are not there all the
.()
15 time.
Records become important to.give us confidence things 16 are done right.
17 The Zimmer case, and I remember the Zimmer case --
18 that plant that never operated.
The best I could tell, that 19 plant was built pretty good but they didn't have the 20 paperwork to demonstrate the plant was built right because 21 for whatever reason they didn't focus a lot on the quality 22-assurance records.
_23 Records have an important role in the process.
It 241 doesn't necessarily.make the plant safer, but gives us the 25 assurance the plant is safer -- is sufficiently safe, and so l
l T )T
(~
ANN RILEY & ASSOCIATES, LTD.
l Court Reporters l.
1025 Connecticut Avenue, NW, Suite 1014
. ashington, D.C.
20036 W
(202) 842-0034
t.
217 1
some of-those requirements are important just because they j
2 are there so we can do our job right, the licensees'
-3 managements can do their job right, and so sometimes you 4'
need that detail.
)
5 In other cases you don't need the detail.
6 DR. APOSTOLAKIS:
Well, it seems to me that a lot t-b
- 7 of these problems come from the fact that we have speed l
(
_8 limits.
i t
9-CHAIRMAN SEALE:
George, it is part and parcel of 10 the same problem we have when we-talk about whether a change 11 in risk is negligible and whether it is positive or 12 negative,-and you know where that decision and you know 13 where that decision --
14 DR. APOSTOLAKIS:
It's the same problem.
15' CHAIRMAN SEALE:
Exactly.
16 DR. APOSTOLAKIS:
In 1.74 we had shades of gray.
17 CHAIRMAN SEALE:
So why --'I mean we know where it 18 has to be fixed.
j 19 DR. APOSTOLAKIS:
We do?-
'20
' CHAIRMAN SEALE:
Sure.
21-DR. APOSTOLAKIS:
What is that?~
22 CHAIRMAN SEALE:
I mean it is a policy matter and 23 it mayl require some legal relief.
224.
DR. APOSTOLAKIS:
Oh, I'm sure it does.
25" CHAIRMAN SEALE:
Yes.
I - ~'i ANN RILEY & ASSOCIATES, LTD.
! '[\\s) 4 Court Reporters l
-1025 Connecticut Avenue, NW, Suite 1014
. Washington, D.C.
20036-(202) 842-0034 4,,,-
--m,.
--.,-.rm-.,
,,n-,
g.-,,
,,-, -.,,,4m,.,
,..r
218 1
DR. APOSTOLAKIS:
But would it be inconceivable to f)
2 have the regulations in the form of a goal and an upper Q/
3 limit?
Would that be such a revolution that --
4 DR. KRESS:
I like that concept, George, and you 5
know we have a thing called components and systems important 6
to safety.
I 7
I view what you are saying as why don't we have a 8
comparable thing called regulations important to safety?
9
[ Laughter.]
10 DR. KRESS:
I don't know how you draw the line but 11 some of them would be below the line and some of them would 12 be above.
If you are below the line, you know, you would be 13 awfully soft on --
14 DR. POWERS:
If you had a regulation that was not
,m (Q) 15 important to safety, it wouldn't be a regulation at all.
16 CHAIRMAN SEALE:
That's right.
17 DR. APOSTOLAKIS:
No --
18 DR. KRESS:
Define what you mean.
19 DR. APOSTOLAKIS:
No, that's why we are removing 20 unnecessary regulations.
21 DR. POWERS:
Where are we removing unnecessary 22 regulations?
23 DR. APOSTOLAKIS:
We are saying we are.
We are 24 saying we are.
We have the noble intention.
25 DR. POWERS:
Can you show me a regulation that's
(]
ANN RILEY & ASSOCIATES, LTD.
(s,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 e
.. -....-...-._.-.- _ - - ~ -
219 1
being removed because it is unimportant to safety?
2 DR. APOSTOLAKIS:
Not yet.
3 DR. POWERS:
Can you see one that is on the block?
4 DR. LARKINS:
The criticality margins in 5
containment is a regulation we recently removed and in fact 6
that was an enforcement issue and so that needs to be 7-addressed.
8 The QA -- there's requirements to approve QA 9
changes.
There's various things that we are looking at in I
10 the risk-informed Part 50, but as your point about goals, 11 either something -
this is.my view -- either something is j
12.
important enough to be followed and it should be followed, 13 or it is not that important. If it is not that important to 14 be followed that it's-up to the licensee to decide whether
()
"15 they should follow it or not, then it shouldn't be a 16' requirement.
'17 DR. APOSTOLAKIS:
No, that's not what I meant.
18L MR. LIEBERMAN:
But if it is a requirement, the 19 good licensees -- we don't have to worry about the good 20 licensee, they are going to do the right thing -- it's the 21 poorer licensees that need to follow the requirements and 22 maybe they have cost considerations and they chaose not to 23 follow it.
24' We frequently regulate, maybe we shouldn't but I i
'25 think we do, to the lowest common denominator.
)
()
ANN RILEY & ASSOCIATES, LTD.
V Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
220 1
DR. APOSTOLAKIS:
No, my point is you are
)
2 addressing'the question of whether a particular quantity 3.
should be regulated and I grant you that.
4 Let's say it has to be regulated, but it seems to 5
me that real life tells us that even if we regulate in terms 6
of speed limits, in real life what happens is you have a
{
7' goal and On upper limit.
I have heard over the years from i
1 8
very many people that when the speed limit is 55, oh you can 9
go to 60 and it's okay, but if you are above 60 you are in 10 trouble.
If the limit is 65, up to 70 is not that bad, so 11 people automatically form these, you know, upper bounds and 12 it is the same thing with core damage frequency.
We have 13 discussed here 10 to the minus 3 and above is bad.
14 So I was wondering whether we should recognize 7
Y 15 that and, for example in your temperature of the cladding, V
16 say we would like it to be below this, but we absolutely i
17 don't want it to be above that, so you have an upper limit, 18 in which case you know the whole legal system comes down on 19 you and the fence -- we would like it to be 11 feet, but, 20 you know, the absolute minimum is 10.
21 DR. MILLER:
Isn't that what is already being i
22 done?
23 DR. APOSTOLAKIS:
That is what I am saying.
24 DR. MILLER:
Well, it's already being done, t
l 25~
DR. APOSTOLAKIS:
No, it is not being done.
It's i -
t
/
ANN RILEY & ASSOCIATES, LTD.
(
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 l-
221 1
1 being done de facto --'not after proper scrutiny.
You have 2
heard those words before.
3 MR. LIEBERMAN:
The security levels in the policy 4
tries to address the different levels of significance --
l 5
DR. APOSTOLAKIS:
I understand that.
6 MR. LIEBERMAN:
-- of the violations and this new 7
change will say those violations which are not as 8
significant'as others, we are going to leave that more to 9
the licensee to deal with and so you are going to have some 10 of that, but there are those things which are clearly
.11 important and we want licensees to deal with more promptly.
12 Those are the more significant ones.
.13..
Many licensees have administrative limits.
If we 14 say the requirement is."X" they will operate at a different
()
15 level t'o give that margin ---
16-DR. AFOSTOLAKIS:
Sure.
17 MR. LIEBERMAN:
-- that you are referring to.
18 DR. APOSTOLAKIS:
In fact, yesterday somebody told j
19 me-from a plant that-if the allowed outage time is "X," when 20-they plan, they plan on using "X'over 2" -- so if they have 21 three days, their planning will be to complete the repair in
.22 a day and a half.
23 MR. LIEBERMAN:
Because at some point you get to l
D 24c that limit.
25 Now what we have to do because we have been g
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington,' D.C.
20036 j
l (202) 842-0034 q
L 4
- - ~.. -. -..- -.-..
222 l'
accused of since licensees are required to have procedures,
[E')\\
2 if they have that procedure "X over 2" and they violate "X
\\_
)
i 3-over 2" and then we turn around and issue a citation'for 1
4 that, maybe we shouldn't be doing that.
We should be J
5 issuing a citation at X, not a procedure where they have l
6.
. established to give themselves margin.
l 7
DR. MILLER:
According to this new approach to 8
maybe have the inspector will feel more comfortable in using
)
L i
L 9
judgement on issues like the speed limit -- gee, we'll let
.10 him go 70 instead of 65 -- is that part of this policy you 11
.are going to --
~
12 MR. LIEBERMAN:
It will --
1 13 DR. MILLER:
Inspectors having a little more 14 judgment?
()s 15 MR. LIEBERMAN:
It.does that but it also has.the 1
~16 negative aspect is much less too, because all we are saying 1
17 is, hey, there is a non-cited violation here, fix it.
18-We are not having the notice of violation, we're 19 not -- I think it's going to improve the rapport between 20-inspectors and licensees.
21 Now if the inspectors got their satisfaction by
~22 issuing tickets, well we are not going to do that,_you know, 23 and so that.is going to have to change some mindsets -- like 24-for the issue, if the inspector wants to help improve the 25
-safety process by pointing out issues to the licensee, we l
L l
i
( ( ').
ANN RILEY & ASSOCIATES, LTD.
LA/
Court Reporters y
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
223 1
can do that in a less so-called adversarial way.
'O 2
CHAIRMAN SEALE:
I am going to have to supervene V
3 here.
4 It's always interesting that George starts these 5
things and thea leaves the room.
6 (Laughter.]
7 CHAIRMAN SEALE:
You have certainly pointed out to 8
us that as we consider this issue in more detail we may want 9
to ask you to come back and share with us your thoughts on 10 this broader issue, and if you have any broader thoughts on 11 this issue, we would be.very interested in hearing'them, so 12 if you'd contact Noel if there is anything you would like to 13 add, we would appreciate it.
14 MR. LIEBERMAN:
Okay, thank you.
().
15 CHAIRMAN SEALE:
Thank you -- an excellent d
16.
presentation, and we do need that revised version.
17 MR.. LIEBERMAN:
He has it right now.
18 CHAIRMAN SEALE:
Okay.
Now we have a request from 1
11 9 Ms. Ginsberg, from-the Nuclear Energy Institute.
20 I must apologize to you, and I will blame John 21 Larkins.
He made an appointment for me to go see the new 22 Commissioner at 10 o' clock and so I am going to have to bug 4
23
.out of here in a few minutes and turn the gavel over to Dr.
~
24 Powers ---nothing personal, I assure you.
25 MS. GINSBERG:
Good morning, Mr. Chairman, and f'
ANN RILEY-& ASSOCIATES, LTD.
s, Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
224 1
thank you very much for the opportunity to present to you
/)
2 the industry's process thus far as we have developed our D
3 framework for a revised enforcement process.
Much like Mr.
4 Lieberman, we have focused on a number of general guidelines 5
by which we will assess whether our process meets our 6
objective, and also focused in large part to date on the low 7
level.-- what we are describing as non-safety significant 8
violations.
9 We began our process by trying to define for 1
10 ourselves what we believe, what the industry believes
)
l 11 enforcement should do.
And, obviously, a lot of the things i
12 that you have seen here have been covered in our comments, 13 the industry's comments, over the past at least five years.
14 The first bullet relates to the agency's overall
.s i
i 15
. general move to risk-informed, performance-based regulation.
Q) 16 We believe enforcement should take advantage of -- look 17 toward and take advantage of the opportunity to become more
-18 risk --
19 DR. POWERS:
Explain to me why you believe that.
20 It is the most peculiar statement it seems to me, 21 enforcement ought to ensure safety.
That is certainly 22 something that I understand it ought to encourage people to 23 do the right thing, but to reflect an agency's move to 24 risk-informed, performance-based reg 11ation is peculiar to 25 me.
I
(
('Sg ANN RILEY & ASSOCIATES, LTD.
(,j Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 l
(202) 842-0034 n
225 1
MS. GINSBERG:
Okay.
Perhaps I can restate it and 2
make it a little bit more understandable.
By that, we mean 3'
-focusing on safety significant items, and the way the safety 4
significance is determined is, in part, based on risk 5
information and risk insights.
Does that help?
6 Moving to the second point, as Jim has already 7
described, we believe, consistent with the NRC Office of 8
Enforcement's view, that there is a strong basis upon which 9
.to rely on corrective action programs that licensees have 10 established, and that that provides a strong basis for-11
. treating level IV's without enforcement -- without the 12 enforcement overlay.
13.
And the third point we think is extrecaly 14 important and that is that in the past, enforcement actions
()
15 have not always been as understandable, as easily or as 16 objectively evaluated a they should have been, and we 17 believe that there is an opportunity here to incorporate 18-more understandable and objective criteria into the 19 enforcement process.
20.
DR. POWERS:
The previous speaker indicated to us 21 some difficulty when we had an objective criteria based on a 22-number.
I think his words were it didn't work.
Is that the
- 23 kind of thing you are talking about, and are you contesting
-24 the previous speaker's contention that a number doesn't
-25 work?-
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
226 1
MS. GINSBERG:
No.
If you are talking about, for
{/}
2 example, a core damage frequency number or something along 3
those lines, I think we would agree that a specific number, 4
as a generic matter, may not be appropriate because we think 5
that there is an opportunity to customize enforcement, as 6
you will see on the process we have developed in the context 7
of developing severity levels or assigning severity levels 8
by plant-specific information, if that is available.
9 I think we were -- and it really goes to this last 10 point, we were more talking about the kinds of terms that --
11.
in fact, in our paper, we provided comments on the paper we 12 have not seen yet, to the Commission, terms like the 13 enforcement policy currently includes, and I will just give 14 you a list of them that I happen to have written down, (A) 15 significance, less significant, more than minor 16 significance, repetitive, isolated, relatively isolated, 17 regulatory concern, programmatic breakdown, regulatory 18 significance and management weakness.
Those all lend 19 themselves to very subjective analysis, and one of our 20 objectives is to ensure that the enforcement process does 21 not -- is not allowed to continue to be as subjective as it 22 has been in the past.
23 DR. POWERS:
Well, I am still stuck on objective 24 criteria that are not numbers then.
I just don't know what 25 you mean.
ANN RILEY & ASSOCIATES, LTD.
s_/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
227 1
MS. GINSBERG:
Well, as -- okay, perhaps as we go 2
through, I will -- it will become clearer.
I think in our 3
Table B we have some criteria, and they are not numbers, 4
they relate to-whether a system can perform its intended 5
safety function.
But perhaps if you hold the question, it 6
may become clearer in a moment.
7 In terms of the areas for change, what we would do 8
would be -- our process would propose, would be similar to 9
that which Jim Lieberman has just described for what are 10 similar to, not exactly the.same as, I should say, because 11 we would redefine severity levels I, II and III and, 12 essentially, process only items, violations that rose to 13 that' level of severity through the enforcement process.
14 Things that are non-safety significant, which is to say the
' (G) 15 equivalent perhaps of level IV's, give or take the 16 definitions, would not be processed through the enforcement 17 process.
They would be rolled back into the corrective 18 action programs.
19 DR. POWERS:
Now, when they define level IV 20 violations, they have a nice term in there, and let me -- I 21 am not sure I can reproduce it exactly, but it is something 22' like "and if left uncorrected, could become more 23 significant."
Okay.
Is what you are saying, that when I 24 encounter a level IV, I don't do anything about it until it 25 progresses to the point that it becomes -- crosses this ANN RILEY & ASSOCIATES, LTD.
e Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
i
'228 1
boundary into a level III?
{'~)
2 MS. GINSBERG:
No, a level IV or something of n,)
\\
3 non-safety significance would have to be entered into the 4
licensee's corrective action program.
i i
l 5
DR. POWERS:
Now, as I understand it, level IV's j
6 are -- it is not that they are not safety significant, it is 7
that'they are minor safety significant, and if left 8
uncorrected, they could become more important.
So why isn't 9
IV included up there?
10 MS. GINSBERG:
Because what we would say is that 11_
anything below this level would be put back into the 12 corrective action program, that it is not of a significant 13 enough safety significance to be treated to enforcement-14 action.
)
15 DR. POWERS:
So it is not non-safety significance, j
16 it is a low safety significance.
t 17 MS..GINSBERG:
Yeah.
And perhaps non-safety l
18 significant is really too broad.
Most of our papers say 19 little or no safety significance.
I think for purposes of 20 the slide, that was a bit of a shorthand.
21 DR. POWERS:
This is an area where shorthand will 22 get us into trouble.
-23 MS. GINSBERG:
Okay.
Well noted.
These are the i
i 24 ares for change, and what I described as redefined levels I, t-l 25 II and III would effectively be those violations that either j ]s ANN RILEY & ASSOCIATES, LTD.
,e Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 L
Washington, D.C.
20036 C
(202) 842-0034 L!'
I
229 1
have actual safety consequences, potential realistic safety
(
2
. consequences, and this third category would involve certain
(
3 well defined significant violations that perhaps would 4
involve integrity issues, to the extent that deterrence is 5
an issue, would capture some of those.
An example might be 6
licensed operator inattentiveness, a 50.7 violation related 7
to discrimination, that. sort of thing.
8 DR. POWERS:
Okay.
Now, when you say potential 9
safety consequences, or even actual safety consequences, and 10 we are being objective, how do I do those?
11 MS. GINSBERG:
It so happens I have an answer for 12 you.
13 DR. POWERS:
Good.
14 DR. APOSTOLAKIS:
The last bullet, you didn't b
15 discuss.
pr.
16 MS. GINSBERG:
Sure.
Pardon me.
17 DR. APOSTOLAKIS:
You didn't discuss the last 18 bullet.
19 MS. GINSBERG:
Yeah.
I had mentioned this before.
20 This really is a -- really relates to the bullet that we 21 talked about before, and the subject of bases are these --
22 and, really, the concern is that the terms are undefined, 23 and so they allow for both inconsistency and lack of 24 predictability.
25 The other thing that the industry's proposal is C,/'(
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
230
'1:
intended to do,.is; intended to avoid the opportunity for the
'(
2 enforcement process to become a surrogate for or redundant O-3
.to the performance' assessment process.
We believe these 4
-processes should be complimentary, but they certainly should 5
not perform the same function.
And one of our objectives is 6-to make enforcement event driven, not a back and forth look l
7 at licensee performance.
8 DR. APOSTOLAKIS:
So what is regulatory 9
significance in your view?
Do you know?
10 MS. GINSBERG:
Well the answer is no, we don't 11 know.
Mr. Lieberman and the industry -- the Office of 12 Enforcement and the industry are in'the process of trying to l
13 discuss that very issue.
But our -- I think our view would 14 be that regulatory significance has no place in enforcement, IT 15 and if you look at the process we have developed, we think l MJ
. 16 we adequately cover the violations that would occur without
-17 the infusion of regulatory significance, without the overlay 18 of it.
-19 And, again, because what this does is this tends 20 to allow the NRC to aggregate small items that in and of
- 21 themselves have either little or no safety significance and 22 come up with a broad programmatic conclusion that we think i-23 is-not well placed in the enforcement process, but, rather, 24
_better in performance assessment.
i-25-DR. APOSTOLAKIS:
Well, let's take an incident or a
i
()\\
ANN RILEY & ASSOCIATES, LTD.
(,
Court Reporters
- ~
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 4
-w
+
l i
231 1
.two from the past and just to try to understand what you
('Ng 2
mean.there.
Because I suspect that a lot of the differences V
3' come from the fact that the agency does not have as a sole 4
objective safety measures, as we understand them.
They are 5
a federal agency, you know, the public expresses concerns 6
and so on, covers on magazines, and they have to respond to 7
these things, because they are a federal agency.
- 8 Several years ago, somewhere, they found operators 9
asleep in the control room.
In terms of actual safety 10 consequences, potential safety consequences, or well defined 11 significant violations, where would that belong?
12 MS. GINSBERG:
Well, I am sure Mr. Lieberman is 13 smiling, because we have had this discussion before, and we 14 have --
I ')
15 DR. APOSTOLAKIS:
Not with me, though.
.V 16 MS. GINSBERG:
No, with you, indeed.
We have 17.
developed what we describe as Table C, and that would be the 18 certain well defined significant violations, and it would i
19 certainly fall under here.
We are not -- the industry is 20 not walking away from those kinds of violations and saying 21 they should not be subjected to enforcement treatment.
22 DR. APOSTOLAKIS:
It is hard for me to see how you 23 can have well defined significant violations that would L24 cover, a priori, somebody falling asleep.
25 MS. GINSBERG:
Well, --
i e"
ANN RILEY & ASSOCIATES, LTD.
k,%)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
232
-1 DR. APOSTOLAKIS:
Then you will have a huge list 2
of things.
I mean --
3 MS. GINSBERG:
That is correct.
4 DR. APOSTOLAKIS:
They shouldn't fall asleep, they 5
shouldn't eat, they shouldn't --
6 DR. POWERS:
Well, it seems to me that the example 7
that the previous speaker brought up that might be 8
appropriate here to discuss is suppose we find that records 9
are not being adequately kept.
I have difficulty imagining
{
10 a PRA telling me that the existence or lack of existence of 11 records has any significant bearing on the core damage 12 frequency.
But the previous speaker said those are 13 important to us, we can't be inspecting everything all the 14 time.
7 15 Now, is that a -- that seems to me to be a 16.
regulatory significant item that is difficult to ascribe to 17 safety.
Is that the one that you want to get rid of?
18 MS. GINSBERG:
Well, I think the way we would 19 capture that would be that that would be considered a 20 nonsafety-significant item such that it would be rolled back 21 into the corrective action program.
To the extent that 22 there was some impact from those records or that was 23-discovered in the course of a performance assessment 24 process, the performance assessment process would look at 25 overall recordkeeping.
[\\
ANN RILEY & ASSOCIATES, LTD.
'\\_
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
_ _ _ _ _. _., _ - _ _ _ _. _ _. _ _ _ ~.. _. _ _. _ _ _. _,..
233 l'
DR. POWERS:
So --
2 MS. GINSBERG:
An individual -- our point would be 3
that an individual violation of a recordkeeping requirement 4
.would not necessarily rise to the level of regulatory 5
significance.
6 DR. POWERS:
So my strategy is pretty clear here.
7-I' keep ru) records, because all I'm going to do is get a 1
8 nonsafety-significant notice, something that's not a 9
violation anymore, and I roll that back into a corrective 1
1 10 action program that does nothing, because I can never.get --
- 11' it never rises up to anything.
l 12' MS. GINSBERG:
Well --
f 13 DR. POWERS:
Is that -- I mean I'm just asking.
14.
MS. GINSBERG:
No, I don't think that's the way it
(
-15 would work.
I'think the way it would work is you woul'd put 16~
it back into your corrective action program.
The corrective 17.
action program would deal with it on the --
18' DR. POWERS:
Why?
19 MS. GINSBERG:
Schedule.
.I 20 DR. POWERS:
Why?
21 MS. GINSBERG:
Why?
Because from a 22 performance-assessment-process perspective you would want to
-23 make sure that your corrective action program is solid.
24-DR. POWERS:
Why?
'25 MS. GINSBERG:.You've got a great deal at stake to
)
ANN RILEY & ASSOCIATES, LTD.
.,.'\\
~
Court Reporters 6
1025 Connecticut Avenue, NW, Suite 1014
('
Washington, D.C.
20036 l
(202) 842-0034 i
234 1
ensure -- licensees have a great deal at stake to ensure
)
that their corrective action programs are solid.
We said at 2-3 the outset that one of the bases upon which this treatment 4
of level 4's is premised is in fact that the torrective 5
. action program is effective.
It doesn't serve licensees 6
well to in fact let their corrective action program fall.
I 7
just don't see that happening, and I don't think that's a 8
realistic -- certainly not a deliberate action, 9
DR. APOSTOLAKIS:
But I do think though that there 10 may be things that are of a regulatory concern.
I mean,
-11 when it gets out that the operators were sleeping --
12 MS. GINSBERG:
Um-hum, 13 DR. APOSTOLAKIS:
It creates a terrible image for 1
i 14 this agency and the industry, and that's of regulatory
(
15 concern.
I mean, it's of concern even though the safety 16 significance may not be high.
So you're saying you have a 17.
Table C that will cover those.
18 MS. GINSBERG:
Yes.
19 DR. APOSTOLAKIS:
Okay.
20 MS. GINSBERG:
I am saying that.
l 21 DR. APOSTOLAKIS:
We'll look at Table C.
l 22 MS. GINSBERG:
Okay.
Let me do this as quickly as 23 possible.
I will focus on the level 4's, and in the process l
24 of that try and identify for you where our areas of l
25 agreement and some disagreement or question with the NRC's
' Q(,,/
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 r
._._.._,_ _ -=_._ -- _._.._ __.
l>
l 235 1
proposed -- elusive but proposed approach is.
{
if 2
What we would do is we would -- and let me make a 3
disclaimer at the outset that this is not in proper 4
flow-chart form.
Please ignore whether it's a square or a 5
diamond.
That is not intended to be meaningful.
6 We would start by saying what is the event or 7
circumstance, determine whether or not a regulation or other 8
requirement has been violated, and then at that point, 9-assuming that what you've found, and for a moment just let's 10 deal with the little or no-safety-significant items i
11 violations, we would say at that point that it is identified 12 by the -- for example by the NRC, no enforcement action i
13 would be taken because it would be put into the corrective 14 action program, and our objective would be that the -- or i
'r
.(
15 our thinking right now would be that the inspector would log 16 that in some way, not create a narrative the way the NRC 17 currently has proposed, but simply track the same identifier 18 that the licensee-uses in its corrective program.
l-19 DR. KRESS:
Do you have a set of criteria to i'
20 decide yes or no in that box?
21' MS. GINSBERG:
This box?
L L22 -
DR. KRESS:
No, the nonsafety significant.
(-
23 MS. GINSBERG:
This.one.
It turns out that it's 24 essentially a negative definition because it wouldn't fall L25 into one of these categories.
And in fact you may evaluate f
ps ANN RILEY & ASSOCIATES, LTD.
\\j Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
.m..
.m.
236 1
it based on one of these categories, and it still would fall
/N 2
into what we're describing as a nonsafety-significant U.
3' violation.
.It would be similar to what are currently level 4
4's, give or take some redefinition.
5 DR. KRESS:
So the safety significance depends on l
6 the. consequences, actual and potential, when -- and other 7
considerations.
8 MS. GINSBERG:
Yes.
Yes.
9 DR. POWERS:
And so now we've -- with that box, i
10 that nonsafety-significant infraction --
11 MS. GINSBERG:
Um-hum.
l 12 DR. POWERS:
Box we have now created two 13
. documents.
We have' created an inspection report, and then L
14 we have created this off-the-record log.
./
i 15 MS. GINSBERG:
No,-I'm not suggesting it would be
.\\d i
16 off the record.
I'm suggesting it could be an attachment 17 to -- or the idea here is not to include a great deal of l-18' narrative by the inspectors, for two reasons.
One is it's 19 duplicative,.because the licensee has already identified in 20 its CAP a description of the item and assigned it either a l
\\
L 21 number or some other designation.
22 DR. POWERS:
If I'm a member of the public and I 23 want to know how this plant that's next door to me works, I 24 have access to this corrective action program so that I can 25 find out all about these nonsafety-significant infractions?
f ANN RILEY & ASSOCIATES, LTD.
- l_,
Court Reporters
.1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
237 1
MS.-GINSBERG:
I don't know the answer to that.
I L[~
2 will find the answer to that out, whether that material is
(
3 public.
I don't know the' answer to that.
I doubt it is.
4 DR. POWERS:
I doubt it too.
5 MS. GINSBERG:
Right.
Right.
I doubt it is.
6.
DR. POWERS:
I mean, that seems to me a fatal flaw 7
right there.
8 MS. GINSBERG:
Well, but the point is whether or 9
not this material is captured in the inspection report via a 10 narrative description or a number designation, the treatment 11-of these items would be the same, which is that they would 12 be rolled back into the corrective action program, no 13 enforcement action would be taken at that point, and so you 14 have an enormous savings of resources and licensee and
()
15 1RRC -- in terms of licensee and NRC attention on 16.
nonsafety-significant items, items with little or no safety 17 significance.
18 For purposes of discussion I just wanted to 19 explain to you what our view is on the NRC's current l
20 proposal, and we have submitted comments on the proposal i
21 that if you'd like I'd be happy to get to the Committee.
22 Basically --
23 DR. POWERS:
Very useful.
r 24 MS. GINSBERG:
Okay.
Be happy to do that.
25 Basically the industry's views are that with i
' [)
ANN RILEY & ASSOCIATES, LTD.
- \\s /
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
. - ~ __,. -.. _.. _.. _
238 1
respect to sort of this process to the extent that it is j] -
2 consistent with the NRC's proposal, we support it and we
-%J 3'
think it is the right way to go.
There are, however, some 4
concerns that the industry harbors about the exceptions that 5
the NRC'has identified.
We've discussed briefly the 6
industry's concerns about including a narrative within the 7
inspection report.
8 In addition, one of the items that we are perhaps 9
most concerned about relates to the NRC identification of a 10 recurring noncompliance, and basically our concerns here are 11 twofold.
One is that repetitive is just -- it's just too 12 subjective a term.
It's not well defined.
13 DR. POWERS:
Tha.t's the one that just flabbergasts 14 me.
I mean, if there's a term that I think is not p
(
15 subjective, it would be repetitive.
I mean, there's one, 16 there's two, there's three.
I guess I don't understand why 17 that would be classed as subjective.
i 18 MS. GINSBERG:
Which is it?
Three, two, four?
I 19 mean, repetitive could be two.
It could be four.
It could 20 be six.
21 DR. POWERS:
Yes.
22 MS. GINSBERG:
Is it repetitive in the same 23 subject area?
Is it all in engineering, or is it repetitive 24 because it happened in engineering and also in operations?
'25 It lends itself to some subjective analysis, and we don't
(~'s ANN RILEY & ASSOCIATES, LTD.
\\/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
~_-____. _ _.
f 239 i'
l' think that that is helpful, particularly --
l ' []-
2 DR.' POWERS:
As far as I can tell from what I've l\\J l
3
-heard in your presentation and what I heard in the previous 4
presentation, subjectivity is an unavoidable feature of this 5
entire enforcement process.
There's going to be some 6
subjectivity someplace.
7 MS. GINSBERG:
I guess we would agree with that,
(
8 but the question is how much and where.
And we don't --
l 9
particularly because these are nonsafety-significant items, 10 I don't think we think this is appropriate.
11 The other thing -- the other reason that we object 12 to the fourth exception is that basically it serves the 13 purpose of a mini-performance assessment.
It allows the 14 enforcement process to be used as an evaluation of
- . V)
'I
.15 programmatic issues, and we think that those evaluations are L
16
.better placed elsewhere.
17 That essentially describes our position on the 18 level 4 issue.
If you have time, I'd be happy to provide 19 for you the -- an overview of essentially the top portion of 20 our process which we've tried to develop.
l 21 DR. POWERS:
Let me just poll our Members a little 22 bit.
I myself am fascinated by this, and would like to take l
23 the time.
We are going over our agenda and will pay for 24 that later, but do the Members have any objection in going 25 on through this?
i I
e
/"'
ANN RILEY & ASSOCIATES, LTD.
(_)s Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
=.
i 240 i
1 Now let me caution you that at the conclusion of
(}
2 this presentation I'm going to ask the NRC if they have any
%)
3 response to what ther've heard.
So it's more than just a 4
little bit over.
We'll probably go to 10:30, I would think.
5 DR. APOSTOLAKIS:
How about Table C?
Table C is 6
the one with --
7 DR. POWERS:
Well, I suggest you progress ahead, 8
and if I start getting nervous about the time, I'll wave my 9
hand or something like that.
10 MS. GINSBERG:
Okay.
That's just fine.
Okay.
11 The caveat with Tables A, B and C is that the 12 industry is still working on these and we tried to be very 13 clear here where we have ongoing efforts underway.
14 Obviously, what we have tried to do here is, with respect to rx
.(
j 15 actual consequences, identii, the condition, and then -- as 16 we are going to go through the supplements and use some 17 other information to develop what we think would be examples 18 of levels I, II, III and what would be a non-safety 19 significant item.
20 This item -- this table is perhaps of greatest 21 interest, or maybe B and C, of greatest interest to this 22 group, and the reason is you asked initially about what we 23 describe as more objective criteria.
And what we have tried 24 to do here is set out more objective criteria that are not 25 numerical.
On the top, obviously, we have what the SSC is, i
s
[]
ANN RILEY & ASSOCIATES, LTD.
(,,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 j
(202) 842-0034 l
241 1
the level of risk, and then on this side, what we have done V(T is try and identify the varying conditions that might be 2
3 involved in a particular violation.
4 I should note that one of the things that our 5
process does is, for example, if you were to have a level 6
III, our process allows, as you discussed with Mr.
7 Lieberman, the licensee to come in with plant-specific risk 8
information to allow the NRC to customize its decision 9
regarding the severity level based on the very specific 10 information that that plant produces.
11 Now, the famous Table C.
12 DR. APOSTOLAKIS:
So, let me understand Table B, 13 since you put it up there.
14 MS. GINSBERG:
Okay.
[)\\
15 DR. APOSTOLAKIS:
First of all, there is a typo.
N.
16 MS. GINSBERG:
Thank you.
17 DR. APOSTOLAKIS:
"Once train of multitrain," that i
18 is the second from the bottom, one train, i
19 MS. GINSBERG:
Oh, I beg your pardon.
20 DR. APOSTOLAKIS:
But -- well.
One train of a 21 multitrain system cannot perform the function, so then we 22 declare it as what, as high risk or significant, depending 23 on what?
24 MS. GINSBERG:
These categories are AEOD defined.
25 What we tried to do was -- we originally started with a look
("N ANN RILEY & ASSOCIATES, LTD.
(~)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 l
(202) 842-0034 l
242 1
at' the maintenance rul'e and the' cut of the set that would be
- (~'Y 2
'used through the maintenance rule.
What we found, however,
- g 3
was.that different plants included SSCs, you know, there was 4'
a variance among the plants, licensees.
So what we did was 5~
- take a look at a more, what we thought was objective set, 6'
and go to the AEOD set that would relate to risk 7
significance.
8 DR. APOSTOLAKIS:
So whether the system has two 9
trains or three trains will be reflected on the category 10 there?on whether it is high risk significant?
11-MS. GINSBERG:
No, this reflects the system.
12 DR._APOSTOLAKIS:
The system.
13' MS. GINSBERG:
And I am_ going to look to my 14 technical folks here.
But, for example, the RPS system, as l'
15 I understand it, would be here.
ll 16 DR. APOSTOLAKIS:
Oh, the significance of the
(
17 system as a whole, i
i 18 MS. GINSBERG:
Yes.
Yes.
L 19; DR. APOSTOLAKIS:
But then again, the_ issue of 20 subjectivity comes'up.
One train of a multitrain system.
21 If the system is one out of two, there is a certain 22 significance to that.
If it is one out of three, there is a I
L
-23 different significance.
I 24 MS. GINSBERG:
And I think that may be where the
{
25
- issue was joined.
But we think that that is a---
that is a ii - O ANN RILEY & ASSOCIATES, LTD.
\\s/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i:
Washington, D.C.
20036 (202) 842-0034 1
mm
243 1
' technical discussion that is-appropriate in this context.
T 2
DR. KRESS:
The vertical line between high and 3-
~ just risk significant, is that drawn at a particular risk
~4 achievement worth?
5-MS. GINSBERG:
At a particular?
1 6
DR. KRESS:
Risk achievement worth.
I mean --
7 MS. GINSBERG:
I think it is based strictly on the 8
set of-systems that AEOD has come up with.
9 DR. KRESS:
That is the --
10 DR. APOSTOLAKIS:
Probably.
11 DR. KRESS:
Probably then it is.
12 MS. GINSBERG:
Okay.
Now, -- I'm sorry, go ahead.
13 DIL _ POWERS :
When you thought about what category j
._14 you would put things in, let's take your high risk 1
()
15-
.significant SSC, the very bottom category, and you said that 16 you have one train of a multitrain, and multitrain you can 17:
mean as few'as two, and let's take something for which there 18 are only two.
Cannot perform intended safety function under l
l 19 some circumstances.
Doesn't that mean your plant is now~in
.20-violation of the single failure rule?
21 MS. GINSBERG:
My technical people are telling me
-22 yes.
23 DR. POWERS:
So that is not safety significant.
24' MS. GINSBERG:
Okay.
Moving to Table C, what I l
25; didn't do.was I didn't, for purposes of developing the p
a
... f')\\
ANN RILEY & ASSOCIATES, LTD.
} (_
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
.~
)
i 244 1
slides, I didn't list Table C.
You have in your package.
1 IT 2
And Table C is currently undergoing close scrutiny by the
%.)
3 industry, as you might expect, it is a subject of great 4
concern to the industry as well as, obviously, to I am sure i
5 people who are going to be reviewing it at the NRC.
6 What we have done here is tried to account for 7
those things which are not susceptible to treatment on the 8
grid kind of approach that we captured in Table B,
- but, 9
nonetheless, are important for purposes of regulatory 10 oversight.
And some of these issues might relate to what 11 would have a deterrent effect.
Some of them relate to j
12 integrity issues.
And what we have tried to do, you can see 13 the list on top, examples of these, 50.5 with deliberate 14 misconduct, employee discrimination, 50.9, complete and (n) 15 accurate information.
If, for example, it would impede the v
16 regulatory process, we can understand -- that is a basis 17 upon which enforcement actions, you know, could be taken.
18 And then the last feature of our process would 19 relate to the CP adjustment, the civil penalty adjustment.
20 And there are a number of items here that are I think 21 currently in the enforcement process.
We would add one 22 which relates to the performance assessment process, and 23 that would be that you may get civil penalty adjustment.
24 You may get a reduction or you may not have a civil penalty 25 at all if you are in the green zone, and that is to credit
(N, ANN RILEY & ASSOCIATES, LTD.
(_)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 o
245 1-good performance.
And this would be the one area where good 2
performance would be credited.
3 That concludes that portion of the presentation, 4
but I would be happy to take questions.
5 DR. APOSTOLAKIS:
So, essentially, your position 6
is that you would like to see more guidance and less 7
subjectivity, less discretion.
8 MS. GINSBERG:
Yeah, I think, depending on how you 9
describe guidance or criteria, yes.
I think what we have i
10 identified through Table B is a way to add clarity, 11 consistency and predictability, and we think those are very 12 important aspects of the enforcement process that haven't be 13 as present as perhaps they should have been in the past.
14 DR. POWERS:
Do members have any other questions 1
15 of the speaker?
16 (No response.]
17 DR. POWERS:
I will now live up to my promise and 18 ask the NRC if they have any response to what they have
~
19 heard, which I assume is relatively new, though, perhaps 20 there have been discussions of this.
21 MR. LIEBERMAN:
Well, we have been having several 22 meetings between NEI, UCS, Public Citizen and NRC since the 23 summer discussing these issues.
I would just like to make l
24 four real brief comments.
One, we don't think that we will L
25 ever be able to eliminate subjectivity completely, nor do we l-O-
ANN RILEY & ASSOCIhfES, LTD.
3, b Court Reporters 1025 Connecticut Avenue, NW, Suite 1014
[
Washington, D.C.
20036 j
(202) 842-0034 L
..w,..
246 1
want to eliminate subjectivity.
But what we want to do, as
(~}
2 Ellen said, we want it to be more predictable, we want it to C/
3 be more consistent, and we want it to be clearer and try to 4
-- we want it to be as objective as we can, but we need to 5
allow that room for judgment.
6 Regulatory significance we know is a term that 7
presents some difficulties, it does have that assessment
)
8 aspect to it.
It is the area we are working on.
In fact, 9
what we did recently, we provided both NEI and UCS a 10 collection of past enforcement actions and we have asked 11 them, from their perspective, which cases fall under what 12 they call regulatory significance in their views and whether 13 it should be considered for escalated space.
14 We have a lot of examples in the policy, but our
/^\\
(
)
15 concern is trying to predict every possible violation or
(_/
16 grouping of violations that may be very significant.
Prior 17 to Peach Bottom we didn't think about operators sleeping, 18 you know.
We have an example in the policy about 19 inattention to duty.
There will be thing that will come up, 20 that by any reasonable standard will be significant.
You 21 know, we don't have a requirement no elephants are allowed 22 in the control room, you know, but, you know, if that 23 happened, we probably wouldn't like it.
And that is 24 obviously not likely.
But the point is that no set of 25 standards will cover every area.
/'3 ANN RILEY & ASSOCIATES, LTD.
(_)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
247 1-We have some differences with the industry, but I
/'".
D) think we have a. larger amount of agreement.
.The basic 2
-3 disagreement in this level IV-areas, the issue that is on 4
our plate now, is these exceptions.
I think these 5
exceptions will be very rare, but I think it does allow the 6
agency the flexibility to enforce the requirements when they 7
feel it is clearly appropriate to do so.
8 This issue of repetition does have some degree of 9
subjectivity because it is not that the same violation 10 reoccurred a second time, it is whether the same root causes
~
11 occurred.
Last time it was because the guidance the 12 licensee used was unclear.
This time was because they 13-didn't give the. resources >to the workers to do the job.
14 Same result, but two different root causes.
The corrective
)
'15 action for the'first one would not have prevented the second 16 violation. 'And when we talk about repetition, we are
- 17 talking.about when the licensee didn't effectively address
.18 the root causes of the first one.
And there is some 19-judgment, but I don't think it is unreasonable judgment.
- 2 0 -
As to the green zone issue, we don't have that 21
' assessment process-in place today.
We are considering what 22 actions we should be taking in the future, and we may well 23-
. agree with NEI on that issue, but it is premature.
24 DR. POWERS:
I am going to ask you to do a little 25 lmore. duty now
/
ANN RILEY & ASSOCIATES, LTD.
(
Court Reporters
-1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
248
.1.
MR. LIEBERMAN:
Okay.
. /~'\\
2 DR. POWERS:
You have indicated that you have'had N).
3 meetings both with NEI and with UCS.
Unfortunately we don't 4
have a representative from UCS here.
5 Could you give us in capsule form their views on 6
these issues?-
7 MR. LIEBERMAN:
I would be happy to.
8 They did provide a paper t t.>.l e C o m m i s s i o n
)
l 9
earlier -- I guess the end of October -- and their basic l
10
. view was they were in agreement with the position of the l
11
-Staff as to the Level 4s.
12 As to regulatory significance, they are in 13 agreement with NEI.
14 DR. APOSTOLAKIS:
I would like to get a copy of
/N l
(
)
15 that.
j 16 DR. POWERS:
I think we have gotten a copy of it.
'l
'17 DR..APOSTOLAKIS:
We have a copy?
18 DR. POWERS:
I believe that's true, isn't it?
19 MR. LIEBERMAN.
Oh, I'll be happy to provide you a 20 copy.
21 DR. APOSTOLAKIS:
Do the members have a copy?
22-DR. POWERS:
I definitely have gotten a copy of 23 it.
Now the route by which I got it I can't attest to.
I i
24 keep poor records, see?
25
[ Laughter.]
l
- t'~
}
ANN RILEY & ASSOCIATES, LTD.
' (d Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 l'
249 1
DR. APOSTOLAKIS:
But Noel, can you make sure p
\\s-}
2 we --
l 3
DR. POWERS:
Thank you.
Are there any other 4
questions that the members would like to pose at this time?
5
[No response.]
l 6
DR. POWERS:
We'll thank the speakers very much.
7 This was -- you know, for a topic that on the face 8
of it looks like it wou11 be conducive to somnolence, you 9
all have given very fine presentations and kept us very well 10 informed on this issue, and I appreciate the quality of the 11 presentations very much.
12 I propose we take a break now until 25 of the 13 hour1.50463e-4 days <br />0.00361 hours <br />2.149471e-5 weeks <br />4.9465e-6 months <br />.
14
[ Recess.]
/~N
(
)
15 CHAIRMAN SEALE:
All right.
We are ready to 16 resume.
The next topic is -- ah, there he is -- is the 17 proposed Final Safety Evaluation Report on the Westinghouse 18 Owners Group topical report on risk-informed inservice 19 inspection of piping.
That is Dr. Apostolakis's 20 subcommittee and I guess Bill Shack is also involved with 21 some of that, so George, I will ask you to introduce the 22 speakers.
l 23 DR. APOSTOLAKIS:
Thank you.
We had a L
24 subcommittee meeting on the 28th of October where some 25 questions were raised about the validity of certain things (O
ANN RILEY & ASSOCIATES, LTD.
x_)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
_. ~ _.
-... ~ - -
)
1 250 11 that are in the Westinghouse report.
I received, I guess
. N 2
'all'the members! received copies of the letters where there if'1 x
-3 are two" things.
One is a letter from-Mr. Walsh to Mr. Wenn
~
4 of the'NRC on what Westinghouse plans to do in revising the 5;
. documents-in response both to the Staff's questions and some 6.
of the issues we raised.
This letter is dated November 3rd.
.7 Today'is the'Sth so I read it but, you know, I didn't really
'8.
have enough time: to really digest the thrust of what it 9-says.
10.
We also received an E-mail from Dr. Boslick of the 11
'NRC Staff,. ho at the request of Mr. Dinsmore I w
12
' understand --
13 CHAIRMAN SEALE:
Yes.
14 DR. APOSTOLAKIS:
-- worked on the equations, l ):
15.
specially'3-8 and equation 3-5 of the document, providing 16-
, arguments why they'are correct, and again this was received 17.
I believe on the 3rd as well at 5:26 p.m.,
so my point is 18 that'I have read these documents.
I am not sure I am really
'19:
on top of'what the authors of these documents are arguing
)
i
'20 about.
21 So today we also -- well, this was a side remark.
22 Last time when we met, I mentioned that there was a paper i
23' from the ASPE group in the journal Risk Analysis which was accompanied by a letter by Dr. Vasili on the validity of 24 25' risk importance measures that carry his name, and there was
'O.
ANN RILEY & ASSOCIATES, LTD.
\\./ -
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C. 20036 l
(202).842-0034 l
l'
251 1
a response to the reservations expressed by Dr. Vasili by 2
the ASME Research members, which response I understand was 3
coordinated by Dr. Balkey, who is here.
4-That is not as important as the other stuff today 5
because I think we are all aware of the limitations of the 6-
'importance-measures and this is not the only forum where 7
they have been aired.
It is just for the committee's -- for 8
the members' information -- but the other two documents I 9
think are more important, the letter from Mr. Walsh and the l
10
-E-mail from Dr. Boslick, so today I guess we will have to 11 discuss some of these things plus the general overview, 12 right? -- to make sure we are all --
i 13 DR. SHACK:
George,.you should point out they also 14 gave us a revised copy of the SER.
,x
(
)
15' DR. APOSTOLAKIS:
That was today -- the first' time 16 I saw it was an hour, two hours ago.
Yes, that's correct.
17' LDR. SHACK:
But they did underline the changes.
18 DR. APOSTOLAKIS:
So you have read it, I trust, 19' Bill?
20 DR. SHACK:
I have looked at it.
21 DR. APOSTOLAKIS:
Okay -- so the point is that we 22' got a lot of information the last 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> perhaps, some of I
l 23 it the last two hours, so I think the purpose of this j
24 meeting is now to understand where we stand and whether the l
25 concerns have been addressed, and what form -- I guess the ANN RILEY & ASSOCIATES, LTD.
Court' Reporters s.
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
252 1
committee has to decide what form its recommendation to the
.(~
2.
Commissi~on will take, but that is for later, not for this
's..
3 meeting, but this is an important meeting with the Staff.
4 So without further ado, I will ask Dr. Strosnider 5
to take the floor.
6 MR. STROSNIDER:
Good morning.
My name is Jack 7
Strosnider I am Acting Deputy Director for Engineering.
I 8
just want-to make a few brief opening statements.
9' We actually have three risk-informed ISI 10-initiatives underway right now.
11~
One of them is the WCAP report, which is the 12 principal focus of this discussion today, and it provides a
'13
_referencable report for licensees that want to come in and 14 use the approach laid out there.
v; 3 15 We also'have two plant-specific evaluations, one 16 for Surry and one for Vermont Yankee for risk-informed ISI 17 programs to be implemented at those plants.
18 These initiatives are an important part of our 19 larger risk-informed plan and in fact they show up on the
.20 Commission's tasking memo.
They are important issues to
.21 show that we are making process with the risk-informed 22 initiatives, and we are actually I think, we hope, on 23 schedule, maybe even slightly ahead with those three safety 24 evaluations 25 As I mentioned to the subcommittee last week, I l
(
ANN RILEY & ASSOCIATES, LTD.
.(
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 I
I i
253 1
have looked at the safety evaluations, reviewed these in the
(T 2
last few weeks, and I have some observations having done
.u) 3 that.
4 The first is that the Staff has concluded this 5
really is an excellent application for the risk-informed 6
approach, and I want to digress perhaps just a second and 7
give sort of the big picture of what we are really trying to 8
accomplish here.
9 The plants have been implementing Section 11 type 10 ISI programs for 20-plus years now, and I don't want to 11 characterize those programs as being totally arbitrary.
12 There was some thought that went into it in terms of looking 13 at different classes of systems and focusing on high stress
-14 areas, that sort of thing, but in fact there is a lot more (G) 15 information we can bring to bear on this issue now in terms 16 of operating experience, and from the risk assessment PRA 17 perspective.
- 18 What we are really trying to do here is take those 19 existing programs and focus or bias those samples, if you 20-will, toward the areas where we have a higher likelihood of
'21 degradation and where the consequences of such degradation, 22 the probability translates to higher risk, so we are trying 23 to do a smarter sampling.
We are really just trying to take 24 what we currently have and see if we can improve on it, so
-25 we are using operating experience, we are using the Q(s,/
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 L
254 1
risk-informed approach to do that.
~T 2
I think this approach really uses a good
[G 3
integrated decision-making process as laid out in Reg Guide
'4 1.174 and as in the risk-informed ISI Reg Guides.
There is 5
consideration given to defense-in-depth is an example, 6
They are augmented inspection programs which are 7
over and above what are currently in Section 11, where we 8
know that there are particularly active or aggressive forms 9
of degradation such as stress corrosion cracking or erosion /
10 corrosion.
Those programs aren't changed.
They are 11 credited in terms of looking at the risk contribution from 12 various psrJ.an; of the system, but they are left there in 13 recogniti_a unat those are important areas that we have to 14 look at.
(g) 15 Overall, it's a good initiative.
We are looking 16 again in terms of biasing, sampling, and you can look at 17 that in the sense of trying to look at some relative 18 contributions to risk from different segments of piping.
19 The results of all this is a significant savings 20 in burden on the industry in terms of manrem exposure and 21 dollars and we think it makes good sense.
22 I want to say a little bit about the efforts of 23 the Staff on this.
I think it is important that, number 24 one, we recognize this was a multidiscipline effort and it 25 had input from all the -- from a lot of different parts of
[
ANN RILEY & ASSOCIATES, LTD.
\\m-Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
255 1
the' organization including Research and many divisions and
(}
.2 NRR.
3 The observation I have, which I think is very 4
important is that when I look at the safety evaluations and 5
even more than that when I discuss these things with the 6
Staff is that I hear the traditional engineering staff --
7 all of a sudden there is some. risk terms coming into their 8
language and on the other hand the people that have 9
traditionally done some of the risk, they are starting to 10' look at things perhaps from the defense-in-depth in the 11 traditional engineering prospectus, so I think, you know, 12 what I see happening is that the Staff is really starting to 13 look at this in an integrated way, and I think that is a 14 good sign and hopefully a good example for future efforts.
0)
(
15-With regard to the future activities on this, we 16 have, as I mentioned, the two pilot plant safety 17 evaluations.
We are looking at getting those out by the end 18
_of the year and also this WCAP report.
19 In discussions with the industry, we recognize 20
'that it is important to establish a good regulatory 21 framework for implementing this sort of approach and in that 22 regard-we have been trying to figure out the most-23 expeditious ~way to implement these risk-informed programs 24 that we find technically acceptable, and we have had 25-
' meetings with the industry.
'h ANN'RILEY & ASSOCIATES, LTD.
\\s /
. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 I
256 1
We have agreed upon a list of what sort of 2
information would have to be in the plant-specific
)
3 submittals that would come in referencing this topical 4
report, so that we have an understanding ahead of time of 5
the scope and what we think is really important for us to 6
look at.
7 We have looked at things like can this sort of 8
thing actually be implemented without Staff review and 9
approval.
To get to that point really requires some 10 rulemaking, which is a longer term effort or maybe some 11 efforts on behalf of the-code, but we are pursuing those 12 areas to try to put in place the best framework we can for 13 implementing these approaches.
i 14 And one other comment in that regard is that the irN LiV) 15 staff has told the industry at least at some meetings that 16 we will entertain relaxation in schedule for updating their 17 10-year programs if they're interested, if they need that 18 time and can justify it in order to come up with a 19 risk-informed program.
And we're looking at developing an 20' information notice in that regard.
21 So having said that, as I indicated, we think this l
22~
is a good approach.
It makes sense to us.
We also though, l
23 as Dr. Apostolakis pointed out, there were some important L
24 issues that came up with the subcommittee in terms of things 25 like how we treat uncertainties, the quality of the PRA, i
/~'T ANN RILEY & ASSOCIATES, LTD.
'\\j Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
257 1
there were some specific questions about equations in the 2
topical report, and the staff -- yes, they have been pretty 3-
. busy in the last week trying to provide information, and I 4
realize a lot of it just got to you very recently.
But they 5
- are prepared today to try to discuss those issues and add L
6-whatever they can to help the Committee in their decisions.
1' 7
So basically that's just the overview I wanted to
.8
.give.
I think Mr. Bagchi, Syed Ali, Steve Dinsmore, and L
9 Jack Guttmann then are going to give~an overview and try to 10:
address some of those specific issues that came up with the 11 subcommittee.
1:2 DR. APOSTOLAKIS:
But you do agree, though, that 13 even -- given that this is a good step forward, which I 14 don't doubt, what we are adopting must be technically sound.
E[
15 MR. STROSNIDER:
Yes, absolutely.
I'm not -- I
,. A_
16l think the important thing in the discussions we've had is 17 yes,.it has to be technically sound.
There's no question 18' about that.
I think we need to look at it in the L19 perspective of what we're trying to accomplish, and I think 20 some'of.the discussion gets into,'you know, absolute values
-21 and what competence you can have in those versus relative 22-differences 1and that sort of thing.
But there's-no question 23 that we want'to make'sure that it's technically accurate and 24 that we're not losing -- making a mistake in terms of the
'25' way'we're changing.the inspection program, i
I-l-
()
ANN-RILEY & ASSOCIATES, LTD.
j \\m,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
258 1
One other comment I'd make in terms of the
('^)
2 integrated process, we do think that the expert panel is a V
3 very important part of this, and, you know, one of the final 4
tests is when you look at the progrs-'a and how it's modified, 5
does it make sense?
You know, can an engineer pick it up 6
and say yes, this really does make sense to me, both from a 7
point of view of these are the areas where I know I'm more 8
susceptible to degradation and yes, these particular systems 9
are important to the safety of the plant, and, you know, 10 when we take that stuff back and look at it, we think that's 11 what we're accomplishing.
But there's no argument at all 12 that we need to make sure that we're technically correct in l
13 terms of, you know, the quantitative or the risk aspects of 14 this, what's being done.
,-~
(
I 15 DR. APOSTOLAKIS:
And it's not really a queocion N/
16 of using approximations.
I mean, I'm all for 17 approximations, as long as we understand what these 18 approximations are.
It doesn't have to be perfect.
19 MR. STROSNIDER:
Right.
Yes, they need to be well 20 defined.
21 DR. APOSTOLAKIS:
Yes.
22 MR. STROSNIDER:
Well understood.
23 DR. APOSTOLAKIS:
Yes.
24 DR. POWERS:
Let me pursue just a second your 25 endorsement of the expert panels.
If in fact we had the
("'g ANN RILEY & ASSOCIATES, LTD.
( s/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
259 1
ultimate Apostolakis-endorsed and blessed PRA that did
['/')
2 everything that is ascribed to PRAs being capable of doing t-l 3
now but in fact they can't do, would not the expert panel 4
need disappear?
5 MR. STROSNIDER:
Well, you're looking at perhaps 6
someone who comes from a more traditional engineering 7
background, and I'd say no.
I'd say that panel is very 8
important as a test to make sure that what you're coming out 9
with makes sense from an engineering point of view.
10 DR. POWERS:
But their role would evolve from one 11 who would make categorizations to one who would endorse the 12 categorizations made by the PRA.
13 MR. STROSNIDER:
In the limit if you had that sort 14 of confidence in the PRA you would probably reach that f,)
15 conclusion.
As you indicated, I don't think we're that w./
16 f ar --
l 17 DR. POWERS:
This is not a trap.
This is a hope l
18 that the technical judgments that have to be made by the 19 expert panel for categorization are born of inadequacies and 20 uncertainties in the determinations made by the PRA.
21 MR. STROSNIDER:
I think they're there to temper 22 those problems, yes.
23 DR. POWERS:
Um-hum.
24 DR. APOSTOLAKIS:
Yes, but I --
25 DR. MILLER:
Could I go a step further on your
['
ANN RILEY & ASSOCIATES, LTD.
\\_,}/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 l
.-.m_.
260 l'
answer of noes, and.that's partly due to your implicit
[Vh 2
belief that we'll never have a PRA of the type that.Dr.
3 Powers has alluded to?
4 MR. STROSNIDER:
Well, in the hypothetical 5
situation where you had this perfect PRA and everybody could I
6-look at it and say yes, this captures all the things that
~7 I'm interested in -- yes, and in limit you'd say well, the 8
. role'of the expert panel might be diminished or changed in 9'
the sense of what you're asking them to do.
But I think at 10 this point in time it's very important that we have some 11 judgment.
And we've looked at the type of people, the 12 makeup, to make'sure that the right sort of people are on 13 these panels,.and that they can address some of the 14 uncertainties both from the materials or traditional
- f'%
_)
11 5 -
engineering and the risk-assessment perspectives.
I l-16'
' CHAIRMAN SEALE:
Never say never.
1 17 DR. MILLER:
I'm not one you'd get ever to believe i
l 18' the computer's going to ultimately replace human judgment.
19 CHAIRMAN SEALE:
But I'd like to believe perhaps 20 that at this. point we have a level of sophistication and 21 personal completeness in the PRA such that maybe 50 percent 122-of -some of these decisions are PRA endorsement kinds of 23 things and perhaps the rest rely almost. exclusively on the L
24 expert panel and other input to them as the basis for i
125
- decisions.
e l
s ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014
[
' Washington, D.C.
20036 (202) 842-0034
-,,-----e
.s-.y,-
m r-m-y-
1:
261 1
MR. STROSNIDER:
I would suggest that perhaps some (7N 2-of the people who looked at these reviews, the staff might 3
be able to address this a little bit better, because I think 4
part of the question you'd have is what kind of changes is 5
- the-expert panel making versus what comes out of the PRA.
6-And maybe they can shed some light on that.
7.
CHAIRMAN SEALE:
So we've tweaked our ears now.
8 DR. APOSTOLAKIS:
The expert parel in this case 9
will have an expert on praise and the phenomena, it will 10
. have an expert on erosion corrosion on it.
Is that 11 understood or --
- 12 MR. STROSNIDER
Yes, in general-you'd be l
13 looking -- yes, I mean, we've looked for people having the 14 materials engineering expertise to support those sort of
()
15-analyses to understand existing plant programs.
As I 16 mentioned, there's augmented programs on erosion corrosion, 17 IGSCC,-how they fit into the mix, as well as, you know, the
'18 plant operators, the system people who have the broader 11 9 perspective of which. systems are most important in terms of
~ 20 plant safety.
So you need to have.a mix.
But clearly part 21 of the expectation is that you'd have people with that 221 proper materials engineering expertise.
. 23 DR. APOSTOLAKIS:
But again the proper materials l
- 24.
- engineering does not necessarily mean expertise in 25
~ probabilistic calculations.
i
[f)
ANN RILEY & ASSOCIATES, LTD.
i.\\ J Court Reporters l
-1025' Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 y
g 3
y
-v
,r-w
,. - - -. +.
+ + =. ~ +
__.. _ _. _ m._. __
l 262 l
1 MR. STROSNIDER:
Yes.
Actually there's some
"'\\
2
[d.
discussion in the safety evaluation about the people who are 3
using the probabilistic code and importance of them having 4.
appropriate training to understand how to use that code, L
5 what its limitations are, and its application.
And there's L
6 actually a discussion in the evaluation on that point.
7 DR. APOSTOLAKIS:
So are we'done with the general 8
discussion of PRA?
9 DR. WALLIS:
I have a question.
I was just 10.
looking at the transparencies we're going to see.
I thought 11 this was about WCAP-14572.
When do we get opportunities to 12 ask questions about that?
It doesn't seem to be in the 13 presentation at all.
So what's the appropriate time to ask i
14 the question?
f) 15 DR.'APOSTOLAKIS:
Anytime you want to.
16 DR. WALLIS:
No, no, not anytime.
Is there 17 someplace at the end of all this presentation, or when?
18-MR. STROSNIDER:
I guess I would suggest that the 19 staff is prepared to make that presentation and then to 20 entertain any questions that we haven't answered.
21 DR. WALLIS:
If they don' t address the repor c, how 22 can we ask a question about it till the very end, when it's l-
!~
23
_a-new topic?
24 DR. APOSTOLAKIS:
Well, 7 and 8 are really the 25 Westinghouse report, viewgraph 7 and 8.
t (s l ANN RILEY & ASSOCIATES, LTD.
/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
263 1
DR. WALLIS:
So at that time we can interject.
/
2 DR. APOSTOLAKIS:
Yes.
3 DR. WALLIS:
Okay.
4 DR. APOSTOLAKIS:
And then 9.
f 5
MR. STROSNIDER:
Okay.
Thank you.
6 DR. APOSTOLAKIS:
Okay, Jack.
You thought you
.7 were going to have only two minutes.
8 MR. STROSNIDER:
That's okay.
9 MR. BAGCHI:
Good morning, my name is Goutam 10 Bagchi, I am Chief of Civil Engineering at Geosciences 11 Branch and I also want to recognize the sister branch Chief, 12 Dr. Richard Barrett.
Close cooperation between these two 13 branches has really brought us to this point where we can 14 say that we have done an in-depth review of the WCAP report
,m-Q
-15:
for a methodology to apply risk-informed inservice 16 inspection, i
17 I want to point out right at the beginn:ng, 18 recognizing fully there needs to be technical accuracy in
]
19 the methodology, that the whole purpose of -- let's keep it j
20 there.
Perhaps I jumped the gun a little bit.
.21 Last time the subcommittee got to know various 22
-people.
On the extreme left is Mr. Dinsmore, Steven l-23' Dinsmore.
He is in the PRA Branch.
Right next to him is L
24 Mr. Jack Guttmann, he is from the Office of Research.
He 25 was responsible for developing the Regulatory Guide and he
- /
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 i,
(202) 842-0034
-.... -... - -. - _ _ - ~
I 264 L
j; 1
is one of the task leaders for that project.
And I 2
. introduced myself, I am Goutam Bagchi.
And to my right is
-3 Dr. Syed Ali, he is also a co-leader in that project.
He L
4 developed the standard review plan.
And Steven and Syed are 5
the principals for developing the safety evaluation report 6
and doing the. review.
7' I.just want.to keep it in perspective in front of
~
8-every -- in front of the full committee that we are talking 9
about-the inservice inspection program.
It is truly p
110 risk-informed.
That is not the end of the process.
The end j
L 1 11 of the process is really developing the inservice inspection 12 program which is risk-informed.
And we are only talking L
13 about piping welds and major impact cf this program is l
'14 really on the volumetric examination of the welds.
()
15 Another important attribute to' point out is that 16
-it is all within the framework of the existing regulation.
1 17 It is entirely within the framework of ASME Code Section XI, 11 8
. inspection methods, acceptance criteria, selection criteria 19 when some flaws are encountered, all of those things, so 20
- there is'nothing totally novel about this risk-informed ISI 21-program.
22 People who do these examinations, workers, craft
?23'
-people, engineers, everybody is familiar with the process, 24 that they don't have to change.
That is an important point 25 I want to make.
/
3001 RILEY & ASSOCIATES, LTD.
l -
Court Reporters
-1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
265 1
The augmented programs, as was pointed out by Mr.
/N 2
Strosnider, is that they are not changed.
And if they are U
3 changed, they are going to have to provide justification and 4-submit a very specific application indicating the reasons i
5 why, and that is going to evaluated by the staff-in a 6
detailed manner.
l 7
And I also want to remind everybody that every 8'
system gets a system pressure' test.
So it is not as though 9
that if you didn't do the volumetric examination, you are 10 not looking the integrity of the pressure boundary.
There 11 are observations that people would have to make as a result 12 of system pressure test, if there is any leakage that would l'3 be observed.
14 And another important point is with. respect to the
/~T 15 letter to the Commission, the Commission didn't require us x)
~ 16 to come back and get a specific approval before the safety 17 evaluation report could be issued.
We do intend to send it 18 to the Commission for information, but a recommendation from 19 the ACRS goes a long way, so I am here to seek your advice 20 and counsel and approval of the staff evaluation report when 21 all the discussions are completed.
22 The reason why we tried to take this kind of an 23 overview is because of our experience with the subcommittee, 24 we wanted to be sure that we explained to you the overall 25_
process and what is the end point of the overall process, l
[
-ANN RILEY & ASSOCIATES, LTD.
(
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 l-
!I
~
266 1:
and then go into the details as you wish.
2 I won't read from the slide because you all are 3-familiar with that, and I just wanted to make that point, 4
that it is in conformance with the regulatory guidance that i
i 5
is out there.
6 DR. SHACK:
Which of the SERs is using the WCAP
.7 approach and which is using the EPRI?
8 MR. BAGCHI:
Surry Unit 1 is using the WCAP l
_9 approach, they are also identifying it with the code case in 10 577.
Vermont Yankee is using the EPRI methodology which has l
11 not'been fully submitted yet, but it also in conformance 12 with the code case in 560.
ANO 2 is the full scope -- well, 13 I should have said Vermont Yankee is in code case in 560, 14 which is partial scope.
And ANO 2 is a full scope EPRI OQ 15 methodology.
ANO 1 is a partial scope Westinghouse 16 methodology -- no, it is EPRI also.
I'm sorry.
17 And this WCAP report, we really -- it shows up on 18-the tracking list as due on 1/31/99, but our commitment is 19-the end of the year and, hopefully, we will be able to keep 20 that.
And EPRI methodology report is supposed to have been 21 submitted by October 31st, the responses to the staff
.22' questions.
We have not received those, so we can't really 23 fix a schedule for that.
l R2-4 And Jack talked about the information notice.
25 This is to ensure that people who do'have an opportunity to Y
l' I h'
. ANN RILEY & ASSOCIATES, LTD.
EN_/
Court Reporters l
1025 Connecticut Avenue, NW, Suite 1014 L
Washington, D.C.
20036 (202) 842-0034
~
267 1
use risk-informed inservice inspection get this in a time
)
2 necessary to develop a program and justify it, or we are 3
going to consider delaying the initiation of the program.
4 And Syed is going to go over the ASME process, you 5
know, what are the periods and so forth, and then you will 6
have an appreciation for what is so important about the 10 7
year interval.
And the code cases in 560 and 577 and 578, 8
once those are endorsed in the Regulatory Guide, and then 9
eventually in the regulation itself, then we wouldn't need 10 to do any kind of a detailed review, it would be pro forma.
11 And I also need to remind you that this acceptable 12 alternative process that we have laid out in our safety 13 evaluation report allows, on a plant-specific basis, use of 14 this risk-informed methodology for the life of the plant.
A
(
)
15 Now, Syed is going to go over the ASME Section XI, 16 ISI process.
17 MR. ALI:
In the interest of just presenting a 18 complete picture, we thought that we would present a brief 19 overview of the current ASME process for the inservice 20 inspection.
The.ASME process if for ASME Class 1, 2 and 3 21 piping.
In that process or program, the life of the plant, 2:2 the 40 year life of the plant is divided into 10 year 23 inspection intervals, and then each interval is further 24 subdivided into what is called the periods, the first period 25 being three years,_the next four years, and the final period C)
ANN RILEY & ASSOCIATES, LTD.
ls_/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
268 L
1 of three years, which makes the total of a 10 year interval.
2 Probably the step number 5 should go maybe before j
3 4 because once these periods and the intervals have been L
4 determined, then according to the ASME Code, a certain 5
sample of piping which is based on the past experience is i
6' selected for inspection.
The sample sizeofor Class 1 is 25 7
percent, for Class 2 is 7-1/2 percent and for Class 3 it is 8
only pressure and visual examination.
9 Coming back to then -- the next step is actually 10 step 4, where, taking the example of, let's say, Class 1, 11
. where a 25 percent sample is to be inspected, then that 12 sample is inspected during the 10 years in the manner 13 discussed or shown in step number 4.
In other words, in the 14 first three years, 16 percent to 34 percent of that sample 15 has to be inspected.
In the first 10 years, 50 to 67 16' percent of the sample, and then within the 10 year interval, 17 then the 100 percent, but that is the 100 percent of the 18 sample, not the 100 percent of the entire piping, is to be l
19 inspected.
20 The only other thing to point out is that the
-21 samples are selected in the ASME process based on the 22 traditional design basis stress levels, cumulative usage 23 factors and factors like that.
24 We will now'go over the overall summary of the 25 risk-informed ISI process.
The first step is to determine ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
. = _ _ _ _ _ _
l 269 i
l 1
the scope of the systems that are to be included and l
^\\
2
\\ ['w ]
according to the Reg. Guide and the standard review plan, l
3 both the full scope and the partial scope programs are 4
acceptable as long as the program is well defined and the 5
guidelines of the Reg. Guide 1.174, with respect to the 6
changes in risk, are maintained.
7 DR. WALLIS:
Maybe I should ask a question now?
8 MR. ALI:
Yes.
9 DR. WALLIS:
About failure probability assessment, 10 2b.
I looked at Table 353 and there's a list of 11 probabilities with no ISI and with ISI, and maybe I had the 12 impression that this was a fairly elaborate calculation, but 13 the numbers differ in quite a few instances by a factor of 14 exactly 10.
Now, why is that?
(O) 15 MR. GUTTMANN:
This is Jack Guttmann from the v
16 Office of Research.
When the fracture mechanics codes, the 17 assumptions on the effects of inservice inspection is 18 assumed that -- assumes that if a degradation occurs where 19 you get a crack, as an example, that exceeds the ASME limit 20 of 10 percent, if that starts to reach that condition to 21 exceed that, and you do an inspection and identify that 22 inspection, and then go through the complete ASME process 23 for assessing the consequences, should you correct 24 correct the defect or replace the pipe, then the pipe really 25 doesn't break, you have caught it, and then you do some
(~~}
ANN RILEY & ASSOCIATES, LTD.
l
(_,/
Court Reporters i
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
L 270 t
l l'
other samplings to make sure that you -- that the crack that
)
you found is not an anomaly and you correct the problem 2
3 before it turns into a leak or a break.
So taking credit 4
-for this inspection and corrective actions typically results 5
in about a factor of 10.
6 DR. WALLIS:
It is exactly 10.
7' MR. GUTTMANN:
Well, that is probably rounded off.
8 But it is'approximately a factor of 10.
l 9:
DR WALLIS:
So do you inspect -- there is one 10-here that says water hammer.
You have inspected -- it just 11 seems to me odd because of the difficulty of evaluating 12 water hammer loads and so on, that you can really have this 13
- factor of 10 so exactly determined as a result of ISI, when 14 it is not that precise a problem.
Isn't it that someone has
'I \\
15 used judgment and 10 has appeared as a number?
\\s,/
l 16 MR. GUTTMANN:
No, this is -- the factor of 10 is l
[
'17 an industry consensus based on a lot of analyses, and was i
18 Laddressed during the ASME research activities, and our
^
19 contractor at P&L, Fred Siminan, did quite a few analyses, L
20 and his-conclusion'has been that if you really look at it, 21 the effects of inservice inspection, the benefit only gets l
- 22
.you about a factor of 10 in improvement of failure
.23 probabilities.
24 DR. WALLIS:
So this 10 is a result of a lot of l
[
L25' calculations?
- =,G -
ANN RILEY & ASSOCIATES, LTD.
Id,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
271 1
MR. GUTTMANN:
Yes, it is.
2 DR. WALLIS:
It comes out as exactly 10 several 3
times.
4 MR. GUTTMANN:
No.
No, well, it is interesting, I 5
didn't focus on that in that report, but it is on the order 6
of a factor of 10.
It can vary, it doesn't have to be 7
exactly a factor of 10.
8 DR. WALLIS:
Now, if I submit an income tax return 9-and the numbers in one column are different from the next by 10.
a factor of 10, they would probably ask me about that.
11' MR. GUTTMANN:
Well, they ask you to -- they also 12 advise you to use the nearest dollar.
13 DR. WALLIS:
Well, I will another --
-14 MR. GUTTMANN:
They don't like the cents.
([
L15 MR. BAGCHI:
But the significance of'those things 16:
.is that risk-informed-ISI can direct inspection efforts 17 towards the areas where they would pay off.
And I think 18' they have indicated that including, for example, the service
'19 water system, the risk reduction was significant.
So the 20
. scope is really important.
It is --
21-DR. WALLIS:
Well, that's -- I am just trying to 22 figure out how complicated the analysis was.
23 DR.-APOSTOLAKIS:
We have a comment from the 24 gentleman. MR. BISHOP:
This is Bruce Bishop from O.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut' Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
. ~
. ~..
l' 272 i
" l' Westinghouse, I am involved in some of the structural
(
2 liability calculations for the failure probability 3
assessment.
The numbers do not come out exactly 10 4
typically in the probabilistic fracture mechanics 1
5 evaluation.
6 Typically, we do not take any credit for inservice L
7 inspection in the risk ranking, using the probabilistic 8
fracture mechanics results, except where'we have augmented 9
' programs.
And in some cases, okay, we have -- rather than
-10' using the actual probabilistic fracture mechanics results 11-for like wastage, and for some of those augmented programs, 12 based on the guidance and the experience that Jack was 13 talking about, based on these studies, we have, in fact, 14 applied an exact' factor of 10 to the failure probability j
(
15 without inservice inspection.
So that-is sort of a rule 16 base but that is only for -- when you are giving. credit for
-17 augmented inspection programs, they are already part of the
- 18 program that we are not changing.
19 That is not exactly calculated with a 20 probabilistic fracture mechanics code.
The number without 21.
the inspection was calculated with a probabilistic fracture 22.
mechanics code and, then, based on the rules we have for 23 different augmented programs, we applied that factor of 10
'24 based on those rules.
25 And the 10 in fact was then based on the work that h
ANN RILEY.& ASSOCIATES, LTD.
f'L%_/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
._. ~ _ _.
1.
273 1
Jack was talking about based on studies that were done at
/
2 PNL and other --
\\,
1 s
3 DR. WALLIS:
I had a question about 2A, 4
consequence evaluation.
You've got two tables here, 5
Millstone 3, Surry, CDF percent contribution by initiator.
6 These are both PWRs.
There seems to be absolutely no 7
correlation whatsoever between the two tables.
It says core 8
damage frequency percent contribution by initiator.
These 9
are tables _1, 2,
3, 1,
3 for two different plants.
4 1
.10 Absolutely no correlation whatsoever.
One might expect that 1
l 11 the rankings or the percents or something would at least in 12 some way.be related to each other, but they --
l 13 MR. GUTTMANN:
You're talking about --
14' DR. WALLIS:
Just seem to be completely different.
'15 MR. GUTTMANN:
You're talking about~ comparing two 16 plant results?
17 DR. WALLIS:
Right.
18 MR. GUTTMANN:
It turns out'that these --
19 DR. WALLIS:
They're both PWRs, they're both 20 Westinghouse.
21 MR. GUTTMANN:
It turns out that the contribution
- 22 to CDF is very plant-specific.
There are certain conditions 23-at Surry that are totally different from Millstone.
In 124 addition, the methodology, they both used somewhat slightly 25 different methodologies in that there are a lot of l
l! ["')
ANN RILEY & ASSOCIATES, LTD.
l (3/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
l l
274 i
1 sensitivity' studies were performed for the Millstones which
(}
2 did not include small breaks and leaks where the Surry 3-one --
4-DR. WALLIS:
Why are there such differences?
i-5-
MR. GUTTMANN:
Different plants have different l
6-
-designs, different -
the Millstone is a more-modern plant, 7
so the secondary effects of water spraying, a pipe breaking 8
and knocking out other instruments or equipment is limited 9
compared to that of Millstone -- no, Surry, which is an j
10 older plant.
11 DR. WALLIS:
So the details of the' design play a 12 much bigger role than the main piping and so on, which is --
13 MR..GUTTMANN:
It's a combination.
Yes.
14 DR. WALLIS:
Doesn't this indicate when we're
)
15 looking at these PRAs for plants then we have to look at 16 every plant, cannot make some sort of generic determination 17 for 100 plants based on a sample of a few plants because 18 they're all so very different?
19 MR. BAGCHI:
They should be.
20 DR. WALLIS:
Is that a conclusion I should take i
21' away from this?
22 MR. BAGCHI:
Yes, sir.
That's very solid.
23-DR.-WALLIS:
That's a bit surprising to me.
A l24
. whimsical change in the design of some plant can make all i
the difference in the world to a PRA that looks exactly like 25 i
lO ANN RILEY & ASSOCIATES, LTD.
Court' Reporters L
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
275 1-in every way like some other plant except for some little 2
thing.
3 MR. GUTTMANN:
I wouldn't call it a whimsical 4
change.
It's En1 evolutionary change as we learn from the 5
older plants to the newer plants:
the effects of flooding, 6
effects of other things as we learned.
7-DR. WALLIS:
But the naive observer might assume 8
the two plants are the same.
9 MR. GUTTMANN:.That's a possible observation.
10 DR. WALLIS:
Some. piece of his input and 11 instrumentation in some other place.
12 MR. BAGCHI:
The main reactor coolant loop results 13.
might come out almost identical.
14 DR. WALLIS:
Okay.
Thank you.
))
15 DR. APOSTOLAKIS:
I would ask Mr. Syed, would you 16 go back to slide 6, the previous one, and tell us which 17 parts of this is the new risk-informed ISI process attacking 18 or changing or modifying?
Is it number 6?
19 MR. ALI:
The first change is really it could be H2 O even step number 1, which is the scope.
21 DR. APOSTOLAKIS:
Okay.
22 MR. ALI:
For example, Surry, the plant has chosen L23
-to apply this program to class 1, 2,
and 3 piping as well as 24 other piping that might be important from the point of view 12 5 of, you know, maintenance through experience.
So that's the O\\
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
276 1-step --
(
-2 DR. APOSTOLAKIS:
Okay.
3 MR. ALI:
That could change.
Vermont Yankee is 4
applying this process only to class 1.
5 DR. APOSTOLAKIS:
Um-hum.
6 MR ALI:
So"that's a change right there.
Once 7
this system scope has been defined, I think that then the 8
next major change will be in step number 5, the sample size.
9 In ASME process there's a fixed sample size, 25 percent for 10 class 1.
Here it could be different.
11 DR. APOSTOLAKIS:
Okay.
12 MR. ALI:
Typically it's less than that.
Step 13 nunter 4 is pretty much unchanged.
So those are the major 14 changes, and I think --
-(
)
15 DR..APOSTOLAKIS:
- Well, 6.
~
16 MR. ALI:
Yes, some of the things that we have 17.
knocked out here about, you know, examination matters.
That 18 might be changed because now the examination matters might
- 19 =
be more tuned with the degradation process.
g 20 DR. APOSTOLAKIS:
But 6 also changes, right, as a 21' result of the other changes?
22 MR ALI:
Yes,-because 6'is not used.
i.
23' DR. APOSTOLAKIS:
Okay.
.24 MR. ALI:
It's not the basis of selection.
25 DR. APOSTOLAKIS:
Now I understand also that this L[i;fN ANN RILEY & ASSOCIATES, LTD.
Court Reporters s
1025 Connecticut-Avenue, NW, Suite 1014 2
84 - b3 3
. -.~ -. -. - -= -....
-..._. - - - ~ ~
277 1
new' approach looks at a broader spectrum of failure 2
mechanisms; right?
3 MR. ALI:
Right.
4 DR. APOSTOLAKIS:
The current section 11 limits 5
itself to what?
6 MR. ALI:
It does not look at the degradation 7
mechanisma at all.
8 DR. APOSTOLAKIS:
At all.
9.
MR. ALI:
All it looks at --
10 DR. APOSTOLAKIS:
Just inspection.
11 MR. ALI:
Yes, the design -- you know, when the 12
- piping stress analysis is done, where are the high stress t
.13-levels.
Now those high stress levels might be as a-result 14 of.a, you know, potential earthquake.
[
15 DR. APOSTOLAKIS:
No, but the mere fact that it 16 talks about stress levels, though, it must have some 17 mechanism in mind, doesn't it?
18 MR. ALI:
The stress level -- this is the stress
~9 level due to'the weight of the piping, due to the 1
20 temperature changes, due to the design conditions, which H21 include, you know, possible accidents or earthquakes.
22-DR. APOSTOLAKIS:
Okay.
Okay, I understand.
23.
Thank you.
24 DR. SHACK:
Well, cumulative usage factor means 25-they're including fatigue also.
t l
l
/]
ANN RILEY & ASSOCIATES,.LTD.
[N )
Court Reporters b
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L.
t
- j.,
v
h 278 1
MR. ALI:
Yes, fatigue.
l [h 2'
DR. SHACK:
That's what a mechanical engineer i V l
3 thinks of when he --
4 DR. APOSTOLAKIS:
Yes, that's what I'm saying.
Is 5
there such a thing as a cumulative usage factor for 6
corrosion?
7 MR. ALI:
That's basically due to fatigue.
8 DR. APOSTOLAKIS:
That was my point.
I suspected 9
that really fatigue was behind all this, the current section 10 11.
You're saying it's not quite right.
11 MR. BAGCHI:
High stress point as well as fatigue.
12 Those were the two things.
13 DR. APOSTOLAKIS:
Okay.
Now my colleague, Dr.
14 Wallis, raised the issue of probabilities, but I see later (Q,/
15 you have something on uncertainty, so maybe I'll wait.
16 MR. ALI:
Maybe because of some of the issues that 17 have come up and the questions, maybe I'll just quickly go 18 through this, you know.
19 DR. APOSTOLAKIS:
Right.
20 MR. ALI:
The subcommittee last time told us also 21 to concentrate on the end result, so I'll --
22 DR. APOSTOLAKIS:
Yes, the risk evaluation in 23 terms of risk reduction worth and risk achievement worth we 24 all agree that it depends on the quality of the PRA and so 25 on.
Okay.
l t
I l
/~' ()
ANN RILEY & ASSOCIATES, LTD.
\\_/'
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
...._..m l
279 1
MR. ALI:
When everything has been said and done, i
7'h 2
then.this is really the process that is used to --
(ms/
I I
3 DR. APOSTOLAKIS:
Is Westinghouse going to change 4
that word "importance"?
Are you -- it's not "importance."
-5 MR. ALI:
It's more of a potential.
6 DR. APOSTOLAKIS:
I know.
Every time we say that, 7
but is there going to be any action on it?
Anybody from 8
Westinghouse.or ASME who -- are you still going to call 9
.these high failure importance and-low failure importance in 10 the revision?
Can.you find another word, since importance 11 means something else in PRA parlance, or do you want to 12 think about'it, Ken.
13 MR. BALKEY:
This is Ken Balkey from Westinghouse.
14 When we drafted the words to go through this,-we had high
)
15 failure probability, and we stayed away -- people gave a 16 recommendation to stay away from that because even in the i
17 process using the structure reliability model and using the 18 PRA, when you combine it together there are still things 19 outside of that'that the expert panel may come up that were
'20 not'in.the failure probability calculation,.and the word 21 "importance" is more broad to pick up the-probabilities, R22 plus some other deterministic factor that the panel may have 23
.said this piping segment needs-to be in the high category. We've really struggled with that, you know, when we wrote f25~
-the-report.
g F
((~'i -
ANN RILEY & ASSOCIATES, LTD.
'\\_s/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l'
Washington, D.C. 20036 (202) 842-0034 g
280 1
DR. APOSTOLAKIS:
If you interpret the word f~').
2
" probability" not the result of the PRA but the judgment of U
'3 the panel.
4 DR. SHACK:
I should point out in the late, 5
lamented appendix to the reg guide, the staff said high 6
failure potential, which --
7 DR. APOSTOLAKIS:
Potential I think is better.
8 DR. SHACK:
Is a better word.
But --
D DR-APOSTOLAKIS:
Would you disagree with 10
" potential"?
11' MR. BALKEY:
I would -- no, I guess I would not.
'12 I would not, you know, the potential --
13 DR. APOSTOLAKIS:
The only problem, Ken, is that 14 the word "importance" is referring to risk reduction worth
(~N) 15 and risk. achievement worth in most people's minds.
That's 16 all.
17 MR. BALKEY:
Okay.
18 DR. APOSTOLAKIS:
That's all.
There is no deep
-19 technical disagreement here.
20 MR. BALKEY:
Okay.
21 DR. APOSTOLAKIS:
There is a problem with --
22 MR. BALKEY:
There was a lot of input on the words 23
.for this figure.
'24 MR. WEAKLAND:
This is Dennis Weakland from the 25-WOG.
The word " potential" to plant personnel, we have a is
[]
ANN RILEY & ASSOCIATES, LTD.
(_/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 L
(202) 842-0034
1 l
281 1
limited number of people within the plant that are PRA (G"')
2 cognizant, so importance means one thing to them, importance 3
means something other to others.
4 But when you say potential, particularly high 5
failure potential to management and management folks, they 6
immediately equate that to frequency.
So though the expert 7
panel may take something off my RCS line which has a very, 8
very low probability of failure, but if you put it in the 9
high failure potential, they immediately get very concerned 10 that is this a pipe that's going to break?
No.
It has 11 nothing to do.with its frequency of failure, but its 12 consequence of the failure.
So I'm just really reluctant 13
'when I put it in plant space --
14 DR. APOSTOLAKIS:
I understand.
)
15 MR. WEAKLAND:
That's all.
16 DR. APOSTOLAKIS:
How about making the change 17 somewhere else?
Call it high failure probability and low 18 failure probability, and then put in the title selection of 19 examination locations by the expert panel.
Make sure that 12 0 this probability refers to the panel's judgment, not the PRA
'21 result, because that seemed to bother Ken.
That's something 22 we cannot resolve right now, but certainly "importance" is 23 not the right word.
24, MR. BALKEY:
George, the reason we -- originally 25 we started out with high failure probability, okay, and then ANN RILEY & ASSOCIATES, LTD.
4 N
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 IL
282 1
the question begs itself, what numbers do you assign, what
(^T 2
is the break line there between high failure probability and Q,)
3 low failure probability?
And you're looking at a given 4
number.
Is it 10 to the minus 4, 10 to the minus 6, you 5
know, and again because of the uncertainty, we recognize the 6
uncertainty in the failure probability calculations, we said 7
we wanted to keep it more general, so we came up with the 8
word "importance."
And again that was -- but that was'one 9
of the reasons we decided not to use the word " failure 10 probability."
11 DR. APOSTOLAKIS:
You can get that question anyway 12 though.
I mean what is the number between high failure 13 importance and low failure importance?
14 MR. WEAKLAND:
George, we recognize your concern f%v) 15 and I think as an organization since we are a collective of s
16 various different types of users and different levels of 17 people, we'll take a look at it and find a word that really la suites the organization and we will move away from 19 "importance" --
20 DR. APOSTOLAKIS:
Okay.
21 MR. WEAKLAND:
-- since it seems to be a sticking 22 point.
23 DR. WALLIS:
George is not the only one.
The 24 trouble is importance is kind of like double counting.
25 DR. APOSTOLAKIS:
Yes.
("T ANN RILEY & ASSOCIATES, LTD.
(_,)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
l 283 1
DR. RALLIS:
Importance includes in our minds Q).
/~
2 safety significant, so something that separates it.
-3 MR. WEAKLAND:
Right.
It gets into the definition of the word --
4 5
DR. WALLIS:
-- it looks like significance to me 6
and it's the same word so you are putting "x" against "x" 7
which is an old student --
8 DR. APOSTOLAKIS:
That is exactly the problem.
9 DR. WALLIS:
-- student trick for getting good 10 correlation.
11-
[ Laughter. ]
12 MR. ALI:
So the process works in putting a 13 particular segment in one of these four blocks.
In Block 1 14 are segments which are high safety significant but also have
"(
.15 high failure potential or importance --
16 DR. APOSTOLAKIS:
Probability --
'17 DR. WALLIS:
Likelihood.
18-Some people are doing that, too -- they are using 19 the word " likelihood" as if it were different but --
20 MR. ALI:
Than probability.
21 DR. APOSTOLAKIS:
I think we should leave it up to 22 the Westinghouse group to think about it.
23 MR. ALI:
And then within that block then the 24 segments are also subdivided.
If there is a segment that 25 has an active failure mechanism known to exist or it's being A],_/.
ANN RILEY & ASSOCIATES, LTD.
t Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
284 5
l' monitored as part of the existing program, then it is
[( )
2.
susceptible to 100 percent inspection.
-3 If not, then that'part of the segment or the 4
sample is selected based on a statistical selection process.
5 DR. WALLIS:
Could you say something about the 6
criteria for boundaries between these blocks?
7 DR. APOSTOLAKIS:
That is what they wanted to 8
avoid.
i 9
MR. ALI:
Between 1 and 2 or between -- within 1, i
10 the line, the dotted line within 1?
11 DR. WALLIS:
The boundaries that form the "x" in 12 the middle of the square.
13 MR. ALI:
On the horizontal axis between the high 14
. safety significant and low safety significant, initially it j"h
(
)
15 is the risk reduction worth.
Greater than 1.005 is high 16 safety significant, less is low.
17 DR. WALLIS:
Yes.
18' MR. ALA.
Although segments that are between 1.001 l
19 and 1.005 are subject to further review by the expert panel 20 that they might be selected and put in the high safety 21 significant.
22 In segment 2, as I said, is the statistical 23 selection process --
24 DR. APOSTOLAKIS:
But you did not say the 3 to 4, 25 which is what Bishop doesn't like to be asked.
That is why
[~'/}
ANN RILEY & ASSOCIATES, LTD.
l
\\_
Court Reporters l
i 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 M
285
-1 they didn't put probabilities in there --
i /
2 MR. ALI:
No, there is no fixed number there.
'h' 3
DR. APOSTOLAKIS:
.There is no fixed n umber.
4 MR. ALI:
Yes, there is no --
5 DR. APOSTOLAKIS:
But even down there, there is no 6
fixed number.
That is just a starting point.
7 Now.one other thought.
Why do you use nouns on 8
the vertical axis and adjectives on the horizontal?
)
9 Low safety significance?
It's a reason?
I i'
10 MR. ALI:
I think that is because of the word 11 "importance" -- significant is okay but important 12 DR. APOSTOLAKIS:
It could have been important but l
13
'it'is not.
14 MR. GUTTMANN:
The low safety significant and this o
L (j'
-15=
.high safety significant determination comes from the l
16
. detailed. analyses --
l 17 DR. APOSTOLAKIS:
Oh, I understand.
18 MR. GUTTMANN:
-- and the expert panel, so now the 19 expert panel has defined a piping segment - -
20 DR. APOSTOLAKIS:
You didn't understand my 21 question.
On the vertical axis you say "high failure 22 importance" and down there you don't say " low safety 1
23 significance" -- you say "significant."
L
- 24' DR. POWERS:
The proper response to Dr.
1 '
25 Apostolakis is to quote him "No deep thought."
j i
ANN RILEY & ASSOCIATES, LTD.
o 1-Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 w
w --
286 1
DR. APOSTOLAKIS:
I never claimed that all my
(%,}
thoughts were deep.
2 3
DR. POWERS:
No, no, no -- I mean I think that is 4
the response you are looking for there, isn't it?
5 MR. GUTTMANN:
Since we are discussing the WCAP, 6
these are the words that are in the WCAP.
7 DR. APOSTOLAKIS:
I understand that, so maybe the 8
question was to WCAP but it doesn't deserve a formal reply.
9
[ Laughter.]
10 MR. GUTTMANN:
Our NUREG uses different words.
11 DR. APOSTOLAKIS:
These are the decision rules in 12 other words.
That is really what this is -- which is a good 13 idea actually.
14 MR. BAGCHI:
It should all be adjectives, right?
/~T i%;)
15 MR. ALI:
If there are no further questions on 16 this, then the next --
17 DR. SHACK:
Nice try.
18 DR. APOSTOLAKIS:
Now the other thing -- let's go 19 back.
It seems to me this is the heart of the approach so 20 we should really be careful here.
21 It says there in category 4 only system pressure 22 test, okay?
Now Mr. Bagchi told us that system pressure 23 test is present. everywhere.
24 MR. BA9 CHI:
Right.
25_
DR. APOSTOLAKIS:
Would it be a good idea to put
.[]
ANN RILEY & ASSOCIATES, LTD.
's.s/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 i
~ - - -.
~. -.
287 1
it everywhere?
Say statistical selection process for 2
-something-plus system pressure test?
That really would help 3
me a lot because, you know, you really have to read the text 4-to understand what this means.
5 It could be interpreted as only in 4 we do the 6
system pressure test and that is clear not --
7 MR. ALI:
That is not the case.
-8 DR. APOSTOLAKIS:
I think what has happened is j
.9_
that most people read the document and they look at the 10 table and say yes, sure, that's what they are doing, but I 11
.am taking the other point of view.
12 If I go straight to the table, I should really get
.13 a pretty good idea of what the whole thing is about.
14 MR. BAGCHI:
I think that-is an excellent comment.
,f
)
15 We are too close to reading this-thing over and over again.
16 We missed that.
You are absolutely right.
17 If we say that, it enhances the statement in that 18 table, in that box number two.
19 DR. APOSTOLAKIS:
Okay, thank you, i
20 MR. ALI:
Next we will go into one of the issues 21 that came up the last time -- uncertainties.
Jack is going 22 to address that.
23 MR. GUTTMANN:
This is Jack Guttmann.
I am l
24 Research.
-25 At the November 29th subcommittee meeting
![{
ANN RILEY & ASSOCIATES, LTD.
i(_/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 L
r
. _ -.~ -.=_ -...
288 1
questions were raised as to the uncertainties --
Q)-
/
2 DR. APOSTOLAKIS:
November 29th'--
3 MR. GUTTMANN:
I'm sorry, October 29th.
4 DR. APOSTOLAKIS:
I thought it was last year.
5 MR. GUTTMANN:
-- the uncertainties in predicting 6
failure probabilities of piping using fracture mechanics 7
codes.
A comment was also made why not rely on data without 8.
fracture mechanics codes?
9 I would like to address both questions.
10 Oh, you don't remember that question, huh?
11 DR. APOSTOLAKIS:
The second question I am not 12 sure was raised.
The first one for sure.
13 MR. GUTTMANN:
Well, you said if the uncertainties 11 4 '
are so high, assuming that the uncertainties is everywhere,
)=
15, then why.even do it?
IkJ 16 DR. APOSTOLAKIS:
Oh, that's different than saying
'17 why not rely on data?
18 MR. GUTTMANN:
Well, you said why use that? --
19 just rely on data.
20 DR. APOSTOLAKIS:
Anyway, I will agree with you 21 whatever you are saying, but the first one is really --
22
[ Laughter.)
23 DR. SHACK:
We can check the transcript later.
l 24 MR. GUTTMANN:
First, I would like to address the 25 question'of --
. /~'}
ANN RILEY & ASSOCIATES, LTD.
. (,_/
Court Reporters 1
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 i
i L
l.
. ~...
. -.... -. ~.. ~. -. _. -. _. _
289 1
DR.'APOSTOLAKIS:
Uncertainty.
2 MR. GUTTMANN:
Uncertainty will go after that.
I 3
-would like address why not rely on data alone.
l 4
If one wants to estimate the change in risk in 5
terms of quantitative CDF and LERF, data is. insufficient.
L i
'6 Data can be used to estimate the failure rates where the i
7 active degradation mechanisms exist, such as erosion /
l 8
corrosion.
t 9
Data is insufficient or not available for 10 estimating failure rates where no active degradations exist l
'11 and the piping meets the ASME standards in design and L
~12 fabrication.
The data we do have is frequently insufficient i'
13 in assessing the total cause of the failure.
14.
In the past, piping were replaced and generally I
(
15-categorized as failing due to some active degradation l
16 mechanism such as vibration, water hammer, pitting, or.they I
.17 put down just-" unknown."
l 118 The loading defects and other inputs important to L
11 9 today's_ detailed understanding of the failure mechanisms I
.20 were frequently not addressed or not documented.
Fracture 21
' mechanics technology has evolved to a state that enables us 22 to estimate failures and possible locations most likely to
[
- 23 experience failures with greater accuracy than we had 30 L
24 years ago.
25.
Validated fracture mechanics codes and data cannot
/)
ANN RILEY & ASSOCIATES, LTD.
.(_,/
Court.-Reporters 1025 Connecticut. Avenue, NW, Suite 1014
' Washington, D.C.
20036 (202) 842-0034 L
o
~ _. _. _. _ _.. _..
290 1
be separated.
The assumptions that go into the validation
()
2 of fracture mechanics codes rely on available data.
Under 3
certain conditions of interest, data does not exist and 4
extrapolation is required to estimate the failure potential 5
of a piping segment.
Under these conditions, fracture 6
mechanics codes provide us the tools to render such 7
~ estimates.
These tools are not unique to the nuclear 8'
industry, they.are used in other risk-related industries 9
such as aircraft and chemical industries.
10; The power of these tools is more than providing 11-input to a PRA, and this may address-some of the questions 12 that were previously raised with regard to the expert panel.
13 The powers. exemplified by the Surry.and Millstone 14 experience, when questioned, What have you learned from the 15.
experience?,.one' response' focused on the reactor operators 16 who sat on the expert panel.
The impressions that certain 17' piping would never fail was changed.
18 The operators finally got a quantitative estimate 19 of the.importance and the potential for piping failure.
The 20 operators are now more sensitive to the possibility of such 21 failures and the consequences to those failures, a positive 22-
. safety enhancement.
23 Another advantage of applying fracture mechanics 24' codes is the detailed insights that the inservice inspection 25 engineers get on the mechanisms that drive failures.
These
' - O ANN RILEY & ASSOCIATES, LTD.
Court Reporters ss 1025 Connecticut Avenue, NW, Suite 1014
' Washington, D.C.
20036 (202) 842-0034
291 1
are the input that they now have to consider when
(Y 2
fquantifying failure rates, another positive safety Q,
3
- enhancement.
4 These tools also assist in identifying which 5
location in a segment is most likely to experience 1
6 degradation should every location have a very low j
7 probability of failure.
Data is not sufficient to render 8.
such judgment.
Fracture mechanics codes can provide 9
insights as to which location would most likely exhibit 10 degradation.
You would then focus your inspection on that 11 location.
12 Another benefit of fracture mechanics codes is 13 exemplified by the Surry pilot application where a non-destructive examination was added to piping that already 11-4
- g
)
'15 undergoes augmented inspections.
Without the use of 16 fracture mechanics codes, this added inspection would not be 17
-identified as important.
18 Given that as a background, I would now like to l
19 focus on uncertainties.
Uncertainties is a function of the t
i 20 degradation mechanisms.
The more active the degradation l
l l
21
_ mechanism, the higher the failure rate, the greater the 22 number of data points, and the lower the uncertainties.
For
[
23 piping that experienced no active degradation mechanisms, p
24 such as hot and cold legs in the PWR, over 2,000 reactor 25 years of operating experience has shown no degradation.
But A
. [~}
ANN RILEY & ASSOCIATES, LTD.
\\m/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
292 i
- 1' until we provide -- or approve the risk-informed ISI
)
2 program, we will continue to inspect risk-insignificant 3
piping and unnecessarily expose workers to radiation.
4 If the failure potential, let's say, is between 10 j
5 to the minus~8th and 10 to the minus 11, as shown on the 6
slide, the large uncertainties in the calculated failure 7
Lprobabilities'are-a direct result of.the fact that the tails 8
of the input distributions are based on extrapolations from 9
actual data.
Here we are talking about pipin3 that doesn't 1
10
. exhibit.any degradation.
11 For example,.the stresses that cycles, the 12 stress cyclesithat occur.during the operating life of the 13-
' plant and the stress locations that would result in' failure, l
14 to get these tails to intersect, we are talking about six O
ij 15 sigmas in extrapolation.
So, where'you take the data, do-16 you take it in the middle, do you take the highest, lowest, l!
17 those can have a significant effect on how you extrapolate 18 that far to.try to do the statistical approximation of what-1 l
19 the failure rate actually is, and the uncertainties can 20 spread -- span four orders of magnitude from the median 21
. values as shown in this diagram.
22 These uncertainties are accounted for in the 23
' analysis'by incorporating an uncertainty distribution for
'24 the input variables and performing Monte Carlo sampling.
25 About a year ago, we discussed and addressed some F
,OT ANN RILEY & ASSOCIATES, LTD.
~ \\m /-
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034-i.
i w
n w-
.e
293 l
1 of the variables that.are considered to be important in l
2 calculating the failure frequencies.
Okay.
So, as I said,
\\~-}r.
3 these uncertainties are accounted for in the analysis by 4
incorporating uncertainty distribution for the input 5
variables such as shown on this slide, and performing Monte 6
Carlo sampling.
This provides a mean failure probability j
7' with an uncertainty distribution in the failure calculation, t
8 When modeling the piping in a PRA, the uncertainty 9
distributions can be incorporated in the CDF and LERF 10 calculations.
You model it in the fault and event trees, 11 and-for each number you put in an uncertainty distribution, i
12 then you do Monte Carlo sampling, and you can get a mean or 13 median and the uncertainty distributions.
14.
When modeling the piping, as surrogate components,
, ()
15 the impact of the uncertainty distribution can be addressed 16 through a Monte Carlo sampling using a spreadsheet, and that
)
l 17 is the process that the WCAP applies.
One of the things f
18 that we have learned when the -- when we performed an audit 19 analysis trying to incorporate the segments into the PRA, is
)
20 that at this stage of sophistication of the codes, it is 21 more efficient to use surrogate components than it is to 22 model the piping components within the PRA.
We had a lot of 23 difficulties separating the old cut sets and it turned out 24 that we had over 30 million cut sets.
When we performed l
25 these analyses, it took something like two days letting the 4
(
ANN RILEY & ASSOCIATES, LTD.
(_)N Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 A
_- ~ -
294 1
PC -- high speed Pentium processor just crunching,
- [~'t 2
crunching, crunching and then trying to analyze the data.
%)
3 So, our conclusion, or at least the conclusion 4
from our contractors is, at this stage, it looks like the
'S surrogate component methodology is probably more efficient.
6
.This enables the engineer to identify piping segments whose 7
categorization may be -- I'm sorry.
The incorporation of
.8 the uncertainties enables the engineer to identify piping 9'
segments whose categorization may require upgrading to high 10 safety significance, and the Westinghouse methodology does 11 this -- does perform this uncertainty analysis and they have 12 upgraded a few of their piping segments into inspections.
13 ~
In practice, this method did result in upgrading.
14 So I would hope that this somewhat addresses the
/s()
15 issue of uncertainties.
It is modeled in the analysis, it 16 is incorporated for events such as erosion corrosions.
17 Where the estimated failure probability is anywhere between 18 10 to the minus 2 and 1, the uncertainties are significantly 19 less, and we do incorporate those into the analysis.
l 20 DR. APOSTOLAKIS:
Well, let me tell you what my 21 concern was.
It is not -- and I have a real concern there, 22 I mean the equation thing is minor in my opinion, because it 23 is easily fixed.
The uncertainties that you addressed, and 24 I believe this figure also addresses, is due to l
- 25 uncertainties in the parameters that are input to the code,
/7 ANN RILEY & ASSOCIATES, LTD.
s,/ '
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 o
{-
295 1
and you listed a number of them there.
But in these kinds
(~~))
2 of things, where we really don't have sufficient data, as q
3 you pointed out, there is another kind of uncertainty which 4
has to do with the assumptions behind the code, the model 5
uncertainty, as we refer to it.
6 Now, the Westinghouse document does talk about the 7
parameter uncertainties at some point, but then they have a 8
very interesting Section 4.4, starting on page 41 of the 9
supplement, where they finally say the third type of 10 uncertainty is that due to the lack of knowledge about what 11 is the true behavior at the piping location being analyzed.
12 Now, this is what I call model uncertainty.
In other words, 13 in Praise you have to assume a certain law for the 14 propagation of cracks.
Now, these things are not derived, (n) 15 according to my understanding, from first principles, they 16 are semi-empirical, aren't they?
17 Now, some of them are better based on reality than 18 others..Okay.
Like my impression is that these models for 19 crack propagation are better validate -- or having better 20 validation than say erosion corrosion models, but I may be 21 wrong, but that is my impression from reading the 22 literature.
23 So this is the kind of uncertainty that 24 Westinghouse calls the third type of page 41.
But then they 25 go on and say this is the biggest uncertainty and represents
/]
ANN RILEY & ASSOCIATES, LTD.
()
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
__.._m L
296 l
1 the difference if the best estimate failure probability
[~')
v
. prediction and the true fa'ilure probability.
This 2
L
'3 difference is zero is and only if all the failure modes and 4
degradation mechanisms are exactly as modeled and all the 5
input. variables, including their uncertainties, are correct,.
'6
'which is highly unlikely.
7 Figure 2 from some papers shows almost four orders of magnitude difference in calculated leak probability for 8
9 six different assumptions on the initial flawed depth 10 distributions and so on.
i 11 So, then it goes on and says, I am still on page
~12 41, this last type of uncertainty which bounds all the other 13 uncertainties is also the most difficult to predict, and I
.14 fully agree, because the true state of the one. piping
)
15 location being_ analyzed is not really known.
It can only be 16 inferred from available data on'similar piping material 17 samples that have been measured or tested.
In some cases, 18 only engineering judgment has been used to estimate the
'19 median stress values and their uncertainties in SRRA input.
20 And then it goes on to give the rules that we all 21 know, you know, a factor of 20, a factor of this, which I 22 think it is a reasonable approach.
So, and then it says --
23 Figure 229 shows a range from 2 to 5 orders of magnitude on 24 the estimated failure frequency.
25 So my -- the thing that worries me is this kind of I
f
- [
ANN RILEY & ASSOCIATES, LTD.
\\,
Court Reporters L
1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 (202) 842-0034
297 1-uncertainty, not the one you addressed, Jack, because, in
)
2 fact,. Westinghouse says this is the biggest uncertainty and
-3 it bounds all the other uncertainties.
It'is bigger.
Yeah.
j J
.4 MR. GUTTMANN:
Referring to the uncertainties in
)
5 the calculationso-- let.me --' Bruce Bishop is coming up.
6' DR. APOSTOLAKIS:
Let me finish my concerns.
]
7-Within -- I mean this is a fact of life.
You are 8
not to be blamed for that.
This is the way it is.
These.
9-models are still, you know --
10 DR. SHACK:
I am not even sure I would call that 11' an uncertainty, George, because the uncertainty they are 12 really talking at: tit here'is at this segment do you.have 13-erosion / corrosion or don't you?
If I don't have erosion /
1 14 corrosion.I probably have a break probability of 10 to the
)
15' minus 12.
If.I do have erosion / corrosion, you know, it's 16
. orders of. magnitude different, but you don't really know 17 whether you do or you don't have it, so that produces a very 18 large. uncertainty in the true probability of failure and it 19 comes back to the experts' decision as to whether he is 1
20
. going to say there is a likelihood of erosion / corrosion in 21.
this segment or there isn't.
"22' IMt. APOSTOLAKIS:
No, but again I am getting i.
-23 confused here.
There is a calculation of the probability of j:
24-leak, right?
25 DR. SEACK:
I look at that only as a a
4 ANN-RILEY &' ASSOCIATES, LTD.
j s /.
Court Reporters 1025 Connecticut Avenue,1M, Suite.1014 Washington, D.C. 20036 (202) 842-0034 1
298 1
quantification of what the analyst really knows.
You know,
()'
2 if you ask me what the hell is the probability of a leak in O.
3 a pipe and I am a structural expert, I have no idea what the i
4 probability of a leak in a pipe is.
I can tell you with a 5
reasonable degree of uncertainty is this pipe subject to i
6 erosion / corrosion.
7 DR. APOSTOLAKIS:
No, no, no, no --
8 DR. SHACK:
Is this pipe subject to severe 9
erosion / corrosion.
10 DR. APOSTOLAKIS:
But if we go back to the 11 basic --
12 DR. SHACK:
-- and what the code does for me is to 13 turn that into that probability --
14 DR. APOSTOLAKIS:
But let's go back to the --
7"N (s_ )
15 DR. SHACK:
The code can't tell me whether that 16 segment is subject to erosion / corrosion.
That is a decision 17 that I have to make.
18 DR. APOSTOLAKIS:
That's right.
19 DR. SHACK:
And that is an input I put into it.
20 DR. APOSTOLAKIS:
Yes.
Now let's go back to the 21 decision rules, the selection process.
Can we go back to 22 that viewgraph?
That's right -- that's the one.
23 So here we have the vertical axis, okay, which is 24 the output of the expert panel, but it takes as input some l
l 25 of these calculations.
Now clearly I have to decide that I
[')
ANN RILEY & ASSOCIATES, LTD.
(_ /
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
L 299 1.
am looking for erosion / corrosion someplace, but then I run a
)
2
- model and everything Westinghouse says on page 41 applies.
3 This model now is empirical.
It has some 4
theoretical insights but it is essentially empirical.
In 5.
fact, there is a figure in one of the papers by Kastner 6
where he shows the actual measured rate of removal of 7
material and the estimate-by his code, and clearly there is 8_
_model' uncertainty.
9 Now if you look at tables like the one on page 89, 10 you see that there are leak probabilities that are all over 11 the place, from 10 to the minus 2 down to 10 to the minus 7.
12 These are clearly the results of some calculation.
I can't "13 imagine an expert said it'was 9.14 and 10 to the minus 7.
14 The fundamental problem there is if there is (O
15
.indeed several or a few order of magnitude uncertainty about j
16 these numbers, can I really trust the ranking?
17-So it ultimately comes down to how'I am using 18.
these numbers, okay?
Now one of the counter-arguments is 19 yes,:the expert panel will look at them and if you have an 20 expert who has read-that paper by Kastner, for example, he 21 will look'at this and say well, gee, you know, maybe it's 22-not 10'to the minus 7 -- it could be as high as 10 to the 1
l
'23 minus 5, and then I can be conservative and put it on the 24 high safety significance category anyway.
You know, that's 25.
-a way around it.
- [~]
ANN RILEY & ASSOCIATES, LTD.
\\s./
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 j.
300 1
DR. SHACK:
You won't get that difference except
/)
2 if you decide this joint is subject to erosion / corrosion and
.g 3
this joint isn't.
That is the decision that gets you from 4
10 to the minus 2 to 10 to the minus 7.
5 Once I decide---
6 DR. APOSTOLAKIS:
Well, that is not something that 7
is stated anywhere and I am not sure that that is --
8 MR. BISHOP:
But again we are talking about the 9
methodology.
The methodology is very general, but I would 10 like to support what Dr. Shack is saying.
11 If you have an active degradation mechanism, you 12 will get a high failure probability.
If you have none, you 13 will get a very low failure probability.
14 The other point I would like to make is --
!(
)
15 DR. APOSTOLAKIS:
If you have none?
What do you 16-mean by none?
You have not' observed any signs of --
l 17 MR. BISHOP:
If you have a system that is very low 18 temperature, so that you can't cycle temperature, you know, 19 thermal fatigue cycles, you know, there's just no way you 20
.can actually fail it, okay?
It's very robust.
You cannot 21 identify a degradation mechanism.
22 The tool will help you calculate a number to 23 assign to that very low failure probability, and it will be 24-very, very low because you have no degradation.
25.
If you have a good chance of having erosion /
/"'s ANN RILEY & ASSOCIATES, LTD.
- - (s,) '
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
-.,. - ~ _ ~. -. - _.
1 301 i
l corrosion or thermal-stratification, fatigue or vibration
- 2-fatigue, you will get a high failure probability.
We know.
~3 :
that and if you go through and look at some of the example 4
results that come out of the Surry calculations and so-5 forth, you will see there is a correlation between active 6
degradation mechanisms and failure probabilities.
1 7:
Now again, that is in the details of the data, but 8
that correlation is built into it.
9 The other point I would like to make is that in 10 the chart that' Jack showed I would like to think that 11 because we have been very careful about this that we are 12' calculating the upper bound on the failure probability.
(
13, Now we call that median but I think in reality if L
14 we:go back and look at the uncertainties that were assumed.
L(q j
15 in that study that was-done at PNNL versus what is actually 16 used in the SRA calculations, we have gone through.and we f
17 have tried to. err on the conservative side so that we are in.
18 fact, I believe, and I think we could go back and show in l
19 l detailed' step by step, for example, like the way we treat g
"20 there were two orders of magnitude, for example, on the flaw 21 density -- how-many flaws you have in a given weld.
22 I know in that case we are -- really we are using
.23 like the upper bound numbers, not the median numbers, and so
-24 there's two orders of magnitude uncertainty there.
We use upper bound and we use higher uncertainties 25
' /'V)
ANN RILEY & ASSOCIATES, LTD.
(m, Court Reporters
(
1025 Connecticut' Avenue, NW, Suite 1014 F
Washington, D.C.
20036 (202) 842-0034
..__.m.
302 1
on the stresses.
There are a lot of things that we do to
.I~ -
- N_)i~
try to -- because we don't know~ exactly -- we have got a big
-2 3
uncertainty band here that encompasses'all of the data in 4
the industry.
In reality, we are probably at some smaller 5
part of that, but we assume we have got this in all cases, 6
so we are always tending to err on the conservative side and 7
the fact is we had meetings with Ken's group on risk-based 8-inspection guidelines and these are the people, the-9 developers of the PRAISE _ code, Dr'. Chapman, who has used 10 this process over in the UK,.the consultants from PNNL and 11' we have talked about all these assumptions that we'used in 12 these calculations, and the basic agreement was that either 13 that is reasonable value or it is conservative.
14 I think in that process we have been very
(
11 5 religious in that, and I think we could make an argument
~
16' that we are,really an upper bound probability, i
17 DR. APOSTOLAKIS:
Well, first of all,'this is not 1
.18-what the report says, what you just said, okay, so I have a 19:
problem with that.
20 MR. BISHOP:
But again,.if somebody asked me --
1 21' DR. APOSTOLAKIS:
What is written is different, so
'22 the report should reflect what you are saying.
23 Second, I really think you guys should try to
- 24 understand where I am coming from.
I am not blaming you for 25 the uncertainties.
These exist.
The question is what do we
.LO ANN RILEY & ASSOCIATES, LTD.
' 1 \\m)
Court Reporters
!1 1025 Connecticut Avenue, NW, Suite 1014 Washington,.D.C. 20036 (202) 842-0034
303 1
do about it?
()
2 Now what Bruce just said sounds reasonable, could 3
be one way of attacking it, but I still have a problem.
4 These are parameter uncertainties, and you are telling me 5
yourself here that this is not the biggest concern.
It is 6
the model itself.
7 DR. SHACK:
But it is the model.
You are deciding 8
whether you have -- don't confuse this with the Kastner 9
model.
This is not a predictive model.
Nobody is putting 10 in a pH and, you know, they are not doing a flow calculation 11 to say that mass transfer --
12 DR. APOSTOLAKIS:
So how do they --
13 DR. SHACK:
They are using their judgment to say 14 for this system with this pH level, this material, and this D)
(
15 geometry the adjustment factor on erosion / corrosion is
.5 16 compared with all my experience.
.17 It is a way of quantifying an expert's judgment.
18
'It is not a predictive -- you know, it's not CheckWorks.
19.
DR. APOSTOLAKIS:
Let me understand this, how it 20 works.
I mean the expert comes up with a probability of 21 failure or there is a calculation that is involved?
22 DR. SHACK:
It is a combination of the two.
I 23 think the best description is it is a way to quantify the 24 expert's judgment on the severity of the degradation.
25 DR. APOSTOLAKIS:
But how?
~ /]
ANN RILEY & ASSOCIATES, LTD.
(s)
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
304 1
DR. SHACK:
Because he has to decide whether there 2
is erosion / corrosion and how severe it is, and he does that 3
. based on plant experience, based on'the geometry, based on 4
the materials.
He puts all that in, but he really boils-
~ hat all-down into a parameter that he chooses between zero 5
t 6
and one.
7 DR. APOSTOLAKIS:
But he runs SRRA.
8
'DR.
SHACK:
And then he runs SRRA because --
9 DR. APOSTOLAKIS:
What does that mean?
10' DR. SHACK:
If I decide this was reasonably 11 susceptible to erosion / corrosion -- you know, it's certainly 12 not as bad as the joint at Surry but it is much worse than 13 main steam line piping or main reactor cooling piping with stainless steel, so I am going to say it's, you know, 14 or^g.
( j 15
' conservatively
.5.
I'
~
'16 I still have no idea what the -- as an expert I 17,
'can make that decision based on erosion / corrosion, but to I
18
.come up with a probability of failure I have to run SRRA.
j j
19 DR. APOSTOLAKIS:
Okay, so on page 89 the first 20 entry says that Segment AFW001, postulated failure mode 21 Lcorrosion, and then it gives a probability for a large leak 22 without ISI and no detection, 8.8, 10 to the minus 3.
f-23 Where did that come from?
24 DR. SHACK:
The code.
25 DR. APOSTOLAKIS:
Okay, the code, so now that
(}
ANN RILEY & ASSOCIATES, LTD.
2
%,/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
I 305 1
brings me to something else.
i
(i 2
There is an assumption here which is reflected G'
3 also in the response that Westinghouse wrote and it is not 4
in these viewgraphs, but if we go to the attachment to the 5
letter by.Mr. Walsh, which is also stated in the main 6
report,.on page 4, Uncertainty Analysis.
7 They propose to use the result of the code as the 8
median, the point estimate as a median, and then a log 9
nornal distribution about it, where the parameter sigma is 10 calculated according to what is given here.
11 See, the fundamental assumption there that is not i
12 even stated and I think drives all this is that these point 13 estimates somehow are all a measure of some central tendency 14 of the distributions -- in this case the median.
i 1 )
'15 If that is true, then I can indeed rank them and 16
.put safety significance based on these.
The problem that I 17 am having is that in one case for one weld it may be the 18 80th percentile.
For the same mechanism in another place, 19 it may be the 20th.
)
i 20 There is nothing that tells us that the results of 21 the code are the medians throughout.
If that is true, then 22 everything is okay.
See, that's the problem -- that they 23
-may be all over the place -- so it depends very much on what 24 I do with these numbers.
I don't think we can resolve that issue in this 25 L
I l.
I
- - (/~T)
ANN RILEY & ASSOCIATES, LTD.
Court Reporters s,
1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 (202) 842-0034
306 1
project.
- /y 2
MR. BAGCHI:
We are putting these things into four
\\~ / -
3 boxes. 'After that we do an in-service inspection program.
4 Inspection 11 has a very strict requirement of escalation of 5
inspection sites when you do find anything.
'6 Let's not forget the process that it is driving.
7 We're just putting it into four boxes in one case.
All you 8
can say is that we might make an error in not putting it 9
into a proper box.
I will give you that.
However, there is 1
10 always the pressure test requirement.
If you're seeing some 11 incipient process. going on, it is going to tell you 12 something.
It's not ending all.
13 DR. APOSTOLAKIS:
Incipient 1 pressures is not what 14 bothers me, it's the calculations.
What I am saying --
()
.15 MR. BAGCHI-Well, I really do --
161 DR. APOSTOLAKIS:
The numbers are not all the same 17 percentile.
They're not all mean values.
They are not all 18 medians.
Westinghouse says they are, but they're not.
19 Because.that's what uncertainty is.
The uncertainty is not 20 the same in all the calculations.
So a number that is given 21 as 10 to the minus 5 could in fact be -- the median could be 22 10 to the minus 4.
But the number below it, I mean, that 23 would be something else.
And the assumption is that no, all 24
.the results of the codes are taken as the median of the log 25 normal distribution.
Here it is, stated in black and white.
1 I
!/'~%-
ANN RILEY & ASSOCIATES, LTD.
'[
Court Reporters j s, 1025 Connecticut Avenue, NW, Suite 1014
^
Washington, D.C.
20036 l
(202) 842-0034 l
l
- ~.. -. -
~. -
. - -. ~. - _ - ~. _ -. -. -.
l 307 1-And if.you accept that, then all the comparisons
()
2
- make, you know, they're reasonable.
But the uncertainty is 3
not about all these being medians.
The uncertainty is that l
L 4
you simply don't know what you have calculated, i
5 In many cases in fact. Westinghouse is right.
It's 6
a conservative estimate.
In other cases you don't know.
7 That's what bothers me fundamentally.
Nothing else.
The 8
equations and all that, we can handle.
I mean, it's not a I
9 big deal.
10 But this one I'm having a real problem, and the 11 real problem I'm having is that I do agree that this is a 12-gsod thing to do.
I do agree with that.
But I'm having 13 this problem down below, so if we find a way around it, 14 because I don't think we can resolve it in mathematical
.15 '
terms..The problem is much bigger than what we're trying to
-1 16 do here.
But is there any way, for example, to alert the~
J 17 expert panel'to this possibility so at least some of them 18 will think about it?
19=
That's the only solution I see, that the expert l
20J panel,-you know, which is a catchall solution for 21 everything,-is the only one here.
But at least if we -- you 22.
see, if we write what this attachment to the letter says, I
'23 think we're misrepresenting the problem, because when you 24 say the median of the log normal distribution is equal to 25
-the point estimate, I don't agree with that.
That's not i
i
.Q ANN RILEY & ASSOCIATES, LTD.
g \\,/
Court Reporters m
1025 Connecticut Avenue, NW, Suite 1014
-Washington, D.C.
20036 (202) 842-0034
308 1
.true, or it may not be true.
I don't know.
It may be, but
[~T
'2 it -- I don't see any reason why it should be, w) 3 So you're taking now a very nice distribution of 4
the uncertainty about the point estimates which are 5
' arbitrarily declared as being medians.
I don't even need 6
the uncertainty.
I'll go with the medians and make my 7
rankings.
8 MR. BISHOP:
George, that was to do the at-risk 9
sensitivity study where again you were sort of taking a 10 starting point and you were going to add factors on that, 11 okay, at the 95-percent confidence bound, and again that's 12.
probably reworded, okay.
But again what you're saying is 13 we're at an upper bound, okay, and we're going to add some 14 more uncertainty on that to do this sensitivity study.
~
)
15 DR. APOSTOLAKIS:
What I'm saying, Bruce, is that 16 the only way out of this that I see is a good story 17 recognizing what is going on and saying look, for our 18 purposes, the best we can do -- or you don't even have to
.19-
.say that.
For our purposes, we will do this and this and i
20 that.
And part of it -- a big part of it will be we will 21 alert the expert panel to these problems and let them make a 22 judgment, because this is too important.
23 And I fully agree.
I think it's an important 24 thing.
It's an important advance in the risk-informed 25 regulatory effort.
I think that's the best you can do.
But
( ) '
ANN RILEY & ASSOCIATES, LTD.
\\s/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
309 1-at.least recognize this problem.
If you agree with me, of
[#)
~
-2' course.. Recognize it and don't just say these are point
' %./
3 estimates, medians and up and down, because that's not true, 4
- in my opinion.
So the solution to this is how do you manage 5-it. l[t exists outside our will and ability to do something 6
about it.
The issue of model uncertainty, as you know, is a 7
big one. -And you don't'have to resolve it here.
But how do 8
you manage it?
That's my question.
9 Ken.
10 MR. BALKEY:
This is Ken Balkey.
11 DR. APOSTOLAKIS:
I'm sorry I delayed, lL 2 -
MR. BALKEY:
No, that's okay.
This discussion is 13 very germane to the whole application, and I just want to
'14 add to Dr. Bagchi's comments on this two-by-two figure,
[
15 because if you think of the process for putting together 16-
- here we will try our best to manage it.
17-Your aspect of the expert panel being aware is 18 very well because, as Dr. Shack said, the panel is trying to 19 make the final' determination of low and high safety significant, and they are the ones that have to agree we do 20 21 or don't have erosion corrosion, or we do or don't have 22 thermal stratification.
Because it's those type of things 23 that make the difference between high failure importance or 241 potential versus low.
25 But the way we have the process set up is that f
ANN RILEY &1 ASSOCIATES, LTD.
(m, Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202)~ 842-0034 r-4, r
e
310 1.
we're going -- we're capturing all the new information from
(
2 this specific plant as well as the generic -- as well as the 3
industry information, making a determination with the PRA 4-and the expert panel review to put them into high and low 5
safety.significant bins.
But all the systems will continue 6
to get the pressure test and visual walkdown, so if there's 7
something that happens that we weren't expecting, we find a J
8 leak somewhere, we're at least knowing that test is a 9
fundamental test that always remains and not changes.
10 The thing the staff recommended to us that's in 11-those boxes, if we're saying it's a high safety significant 12 segment, if we feel there's degradation there, then we will 13 do the best examination we can.
That's that 100 percent.
14 But the staff wanted some assurance that even in those cases I
'15 where we don't know if there's a mechanism there, in fact 16' maybe when the team did the SRA run they said hey, this is a
- 17 very benign system, but its failure would still be very 18 important to the plant.
It's now in a 19 high-safety-significant bin.
We're still required to go 20 through and still do sample exams throughout those segments.
211 So as the process goes forward, we will get 22-information from those examinations as well as from the 23 pressure tests, and we may come back and have to change some 24 locations.
That's the best way we have laid out trying to 25-manage to make sure as we go forward that we're still O
ANN RILEY & ASSOCIATES, LTD.
" V Court Reporters
.1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
- ~. ~..
311 1
continuing to do exams across all the important segments we V(i 2
.think that are there.
And we'll continue to do pressure 3
tests across all the class 1, 2,
and 3 piping.
4-That was the only thing I was trying to come back 5
to in trying to manage it and recognize those uncertainties 6L are there.
7 DR. APOSTOLAKIS:
Where in your report do you 8-address the expert panel deliberations?
Is there a chapter?
9 MR. DINSMORE:
Yes, this is Steve Dinsmore from 10 the staff.
That's also going to be expanded upon in the 11 submittal.
They gave us a letter which is going to expand 12 upon what's in the report, and at our request they're going 1131 to expand that description.
So it's not an 14 all-or-nothing --
1 )
-15 DR. APOSTOLAKIS:
Okay.
Well, I think we're
.16 running out of time, so let me ask here something about the 17 process.
What is the staff approving at this time, and what 18-is the ACRS commenting on?
I mean, I would really love to 19 see the revision of this chapter, Steve, before I finally 20 say yes, this is. fine.
So how --
21 MR. DINSMORE:
We're commenting on this.
22-DR. APOSTOLAKIS:
On the SER.
23 MR. DINSMORE:
Right.
And they're commenting on
-24 chat.
25 DR. APOSTOLAKIS:
But how doas that affect the t
O' ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 i-
312 1
Westinghouse Owners' Groups, how does it affect the 2
utilities, and are we putting a break on the process?
Are
, j(_-}
3 we following, you know -- or is there a time that 4
Westinghouse will have to go back and revise this report to 5
come back for final blessings?
6 MR. DINSMORE:
No, sir.
7 DR. APOSTOLAKIS:
So -- well, but in your SER you 8
also have some comments that I have to implement.
9 MR. ALI:
They have already given us the words 10 that they are going to put in the --
11 DR. APOSTOLAKIS:
But where is this document --
12 MR. ALI:
This is the September 30 letter from 13 Westinghouse.
14 DR. APOSTOLAKIS:
The September 30 letter.
Since L(gJ-15 these questions were raised later, though, obviously they 16 have not been --
17 MR. DINSMORE:
Well, no, but then there's a second 18 letter.
19 MR. ALI:
But for example --
20 DR. APOSTOLAKIS:
On the 3rd -- of the 3rd.
21 Well --
22 MR. BAGCHI:
Let me emphasize the regulatory 23 process.
It has to be robust.
We just cannot do it on a 24 guesswork of what they might submit.
After we discussed 25 these things, we agreed on the technical resolution.
They (Q_,A ANN RILEY 6. ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
'l 313 1-proposed what they are going to put in on the docket.
)
That's the September 30 letter.
That is the basis on which 2
3 we are saying that this should be acceptable if those l
'4 changes are made.
r.
j l
5 MR. WEAKLAND:
George?
6 MR. BAGCHI:
They have already committed to making 7-those changes.
8 MR. WEAKLAND:
The Westinghouse Owners' Group has L
L 9-committed to the staff that through the SER process and the l-10' comments that we get back that those comments will be 11 reflected in the revised WCAP.
That is a commitment on our l
12' part.
We will do that.
And we will incorporate those 13 necessary changes so-that things are clear.
- L4
,DR. APOSTOLAKIS:
By when?
When do you plan to do 15-this?
16-MR. WEAKLAND:
When the SER is issued.
17 DR. APOSTOLAKIS:
When is --
18 MR. WEAKLAND:
The SER makes it approved.
Those
'19 comments will be incorporated into the WCAP, and then that's 120-
.the: working document.
21 DR. APOSTOLAKIS:
If some of these comments that l
were made today are in the ACRS letter, then you will l
22 23-respond to those as well at the same time, or is that
-24
-creating a new round of reviews and --
25 MR. WEAKLAND:
Our commitment is to put in what is
[3-ANN RILEY &' ASSOCIATES, LTD.
,Ls,/
Court Reporters.
m l
1025' Connecticut Avenue, NW, Suite 1014
[
Washington, D.C.
20036 (202) 842-0034
.2
1 1
314 l
1 in the SER, and that's what we will do.
i 1
([.
2 MR. MARKLEY:
George, the only thing that would 3
be -- could possibly change is what they might have in the 4
SER if the staff decided to modify the SER in some way 5
before they issued it to Westinghouse.
l 15 DR. APOSTOLAKIS:
Okay.
.7 MR. BAGCHI:
.I would really urge a strong l-8 deliberation within the full ~ Committee because, you know, t
L
-9 it's not just one specialized area.
It's a process.
It's l-
.10 an engineering process which has continued for a long time, 11 and we're not proposing to change it altogether, we're just-
.12 making the selection process more systematized based on the j
l 13 proposed methodology.
We think that it produces a very 14 desirable outcome, both in terms of risk as well as in terms 1
b a
ag i
17 DR. SHACK:
And I think,. George, you would find 18 the selection'is actually quite robust in practice, because 19 I think, you know, your panel will -- there will be a 20-reasonable technical consensus as to whether there is L
21 erosion corrosion in this segment or there is not, and you p
'22 will find that if they assign.5 or
.3, it will still, you l
- 23 know, you're going to end up in the 100-percent bin or not.
L 24 The process is going to be very robust to all those inputs.
I L
25 In fact my guess is it's almost independent of these, l
V F
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 i
315 1
because, you know, at least under the present scheme where
(~
'2' they are selecting 100 percent of the susceptible locations,
(
3
-once the expert panel decides it's susceptible to that, it 4
doesn't matter what the code calculates, you know, so the 5
statistical selection processes,.which is the only place 6
where the numerical values computed by the code really enter 7
.in, so you select just how many of those you're going to 8
' choose in the statistical selection process to meet your 9
target leak rate.
10 MR. ALI:
And those are not subject to any 11 degradation process.
That's why they're there.
12 In the SER, I think what you pointed out that, you 13 know, one of the solutions is that the expert panelists 14 should be given some sort of a criteria that these results
'()
15 need to be, you know, looked at with the light that there 16
'are some uncertainties.
I thought we addressed that.
We 17
.have a: separate section in the SCR section 3.4.
It's page 18 19.
-Integrated decision making.
In which we talk about,
- 19
'you know, the importance of the expert panel, the 20
' qualifications they should have.
We also talked at several 21-
. places that these failure probabilities are really relative 22
. values.
They should be taken as relative failure potentials 23
'rather than absolute values.
And that the expert panel
.24 takes those results and uses them.
-25 You know, you asked before when Jack Strosnider I
- O)
ANN RILEY & ASCOCIATES, LTD.
k-Court Reporters
's 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 i-
316 1
was.here what are some of the qualifications of the expert
/~N 2
panel.
And on page 20 in the middle of the page you see U
3 those bullets.as to the expertise that the --
4 DR. APOSTOLAKIS:
I'can't find your pages.
5 MR. ALI:
Well, section 3.4.
Can you find section 6.
3.4?
7 DR. SHACK:
The page. number would be different, 8
though?
9-MR. ALI:
Well, yes, page number I'm giving you --
10 DR. APOSTOLAKIS:
Integrated decision making.
11 EMR. ALI:
Yes.
12-DR. APOSTOLAKIS:
That's on.page 16.
13 MR. ALI:
Yes.
Yes.
14 DR. APOSTOLAKIS:
Well, as I said earlier, maybe
()
15 some stronger advice there on the model uncertainty issue 16 and'what'the panel should take into account is probably the 17.
.only way out of this.
18 MR. ALI:
You can=further, you know, strengthen
~19 that to add that the -- we are at the end of a process.
We 20
-are the end run.
This is a safety evaluation report.
We 21-
.had produced a draft.
We had identified the open areas.
We
~ 22:
met with Westinghouse.
23
- We resolved many, many issues there.
And as a 24 result of that resolution, they had to make commitments.
. 25 These commitments appear in the September 30 letter to us.
' [] '
ANN RILEY & ASSOCIATES, LTD.
(mg/
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034 s
,....:...r.
317 1'
And then. clearly if it is significant, if it is a consensus
N 2;
with a recommendation from the ACRS, we are going to be able (O
3 to accommodate that.
But please, weigh your options 4
carefully and look at the version that we gave you this 5
morning, a redline strikeout version of the safety 6
evaluation report.
Please take a close look at that, and if 7
you need to augment this area, I think that would be a 8
suggestion we can accommodate.
9 CHAIRMAN SEALE:
I think we'll try to be as 11 0 reasonable as we can.
I think we also have to point out 11 that this is one of the penalties.you possibly face when you 12 have a document that's only available at the lith hour, and 13 so -- but we'll try to do the best we can on it.
14 MR. ALI:
I appreciate that very much.
r I
15 DR. APOSTOLAKIS:
I think we have covered --
16 unless the staff feels that there is something they really 17 want to say, I propose we stop.
18 CHAIRMAN SEALE:
Okay.
You want to wind it up.
l 19 DR. APOSTOLAKIS:
I mean, I looked at other 20 viewgraphs, and it seems to me we all understand.
So back 21 to you, Mr. Chairman.
22 CHAIRMAN SEALE:
Okay.
23 None of the members of the committee have any 24 other questions to ask these gentlemen before we release 25 them to their lunch hour or whatever?
B
()N ANN RILEY & ASSOCIATES, LTD.
A,;
Court Reporters s
~
1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C.
20036 l.
(202) 842-0034 l
I; l
l
318 1
[No response.)
/~'
2 CHAIRMAN SEALE:
Okay.
This is the end of the 3
presentations for today that are going to involve the 4
necessity for a record, so I think we can allow the reporter 5
to take the afternoon off, or whatever he plans to do.
And 6
we will try to get back here at 1:15, and we will talk about 7'
what we are going to talk to the Commissioners about next 8
month.
9
[Whereupon, at 12:20 p.m.,
the recorded portion of 10 the meeting was concluded.]
11 12 13 14 f%
(
i 15
%J 16 17 18 19 20 21 22 23 24 25
! [}
ANN RILEY & ASSOCIATES, LTD.
's _ j Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C.
20036 (202) 842-0034
REPORTER'S CERTIFICATE t
i h This is to certify'that the attached proceedings J
before the United States Nuclear' Regulatory Commission in the matter of:
NAME OF PROCEEDING:
MEETING:
457TH ADVISORY COMMITTEE ON REACTOR SAFEGUARDS (ACRS) i i
l DOCKET NUMBER:
1 PLACE OF PROCEEDING:
Rockviller MD
/^
L Q}
'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 l
_ accurate record of the foregoing proceedings, O. E C
Mike Paulus l
Official. Reporter Ann Riley & Associates, Ltd.
l L
it
. _ _ _ _ _. _. _ _ _ _ _. _. _ _ ~. _ _ _ _ _ ~ _ _ _ _. _ _ _. _ _ _. _
i Changes in the h
Alternative Source Term Proposed Rule Package Due to Office Concurrence Comments PERB received comments from some of the offices during the office concurrence process that occurred in parallel with the review by the ACRS. The majority of these comments were editorial in nature and will not change the technical content or intent of the documents. The following are comments that were deemed to be significant.
Statements of Consideration (SOC)
Based on discussions with OGC and others, the phrase " accident dose guideline"has been replaced with " accident dose criteria"in the SOC, except in those instances where the reference is to the guidelines in $100.11, Whlie Part 100 definitions of EAB and LPZ may have been siting guidelines, in design space, the numeric dose values are considered to be criteria.
This change would clarify intent.
OClO requested changes in the language of the Paperwork Reduction Act statement. No specific OMB clearance will be requested as stated previously. Proposed wording will indicate that the agency determined that the information collection burden was insignificant.
Proposed Rule Language RES suggested revising proposed footnote 2 to $50.67 to remove material that is addressed in the SOC:
$~!"~j_",5t?5 #"i"M'f"f'5RN55 O'Mt'N'f.'"f'33 ??!?j"Hi? #"i
"_""?i'.'"'"'_"""""'",,"","~"",~"."""L't""_"_'~,",'~'"~J:""~.":*f_~~n"_~_
i""~~*:""' ' "Y"';' "':," ' ";" "": " """:"~ ~. _"""" "" ""L"' """"' ' L~ :X"; " "",~~1 "_"" "' "
,"'"X'" L """ L" ".C '2 " ' ".' "L'? " _ 'T '~"~ ~;:' ". ","" "_~ :: ~' " "' ~"' """::; :- " _""" _ _
L
".".""...",".'.".".'.I ?.?." ',r # ' '" "!._~.." ~..."~'.~,~,,.."..".? ".. ".".".... '. '.".M.T._" ~.."._ 'E T~ M..."..T".".'i.. 1'_' "_"'
y la 7e O.,.......e.ea'e C.'.:j Oee .'.e,.
- . e.e,The use of 0.25 Sv (25 rem) TEDE-inihese eee;Ga; Cee i,.;Gl.,.ee is not intended to imply that this value constitutes an acceptable limit for emergency doses to the public under accident conditions. Rather, this 0.25 Sv (25 rem) TEDE value has been stated in these guides as a reference value, which can be used in the evaluation of,
proposed design basis changes with respect to potential reactor accidents of exceedingly low probability of occurrence, and low risk of public exposure to radiation.
RES suggested revising the definition of source term in 950.2:
Source term refers to the magnitude and mix composition of radionuclides released from the reactor core, their physical and chemical form, and the timing of their release.
Environmental Assessment l
OGC requested changes to the section entitled " Conclusion" to remove the appearance that this section is a draft determination under the provisions of NEPA. The draft determination is to appear only in the SOC. The SOC already contains the draft detelmination.
i n
' O
-4 O
O O
k l
g8REGg i
47 0,
O 4/
A o
E o
a>
c E
i os G
4* * *V+
o t
NRC l
NON-ESCALATED
~
ENFORCEMENT PROGRAM I
l l
ACRS.WPD
[
o o
o i
1 i
PURPOSE OF ENFORCEMENT PROGRAM l
SUPPORT NRC'S SAFETY MISSION I
IN PROTECTING THE'PUBLIC & ENVIRONMENT l
F e
ENFORCEMENT ACTION SHOULD BE USED:
i f
V AS DETERRENT TO EMPHASIZE THE IMPORTANCE OF COMPLIANCE WITH REQUIREMENTS, AND l
V TO ENCOURAGE PROMPTIDENTIFICATION AND PROMPT, COMPREHENSIVE CORRECTION OF VIOLATIONS e
2 i
i
r-9 9
9 i
o a
ENFORCEMENT ACTION FORMAL ACTIONS V
V civil PENALTY V
ORDER S
IMPOSITION t
I 9
MODIFICATION i
e SUSPENSION i
i O
REVOCATION 4
e WRONGDOERS (NON-LICENSEE) 3 ii
O O
O l
i i
SAFETY SIGNIFICANCE l
i i
i SAFETY SIGNIFICANCE l
t i
l l
t l
ACTUAL POTENTIAL SAFETY SAFETY CONSEQUENCE CONSEQUENCE t
l l
l t
i REGULATORY SIGNIFICANCE 4
j
~ _. _ _. _ _. _.. - _. _.. _ _... _ _ _ _. _ _ _.. _ _ _. _. _ _ _ _. _ _. _. _ _ _
O O
O i
r i
SEVERITY LEVEL i
i MOST SIGNIFICANT h
I VERY SIGNIFICANT l
II SIGNIFICANT CONCERN i
Ill IV MORE THAN MINOR MINOR VIOLATIONS t
I EIGHT FUNCTIONAL AREAS l
t REACTOR OPERATIONS i
SUPPLEMENTI SUPPLEMENT 11 FACILITY CONSTRUCTION l
SUPPLEMENT 111 SAFEGUARDS HEALTH PHYSICS SUPPLEMENT IV TRANSPORTATION SUPPLEMENT V SUPPLEMENT VI FUEL CYCLE & MATERIALS OPERATIONS SUPPLEMENT Vil MISCELLANEOUS SUPPLEMENT Vill EMERGENCY PREPAREDNESS 5
j i
I.-
O O
O NOTICE OF VIOLATION &-
NON-CITED VIOLATIONS NOTICE OF VIOLATION 4
FORMAL CITATION SETTING FORTH REQUIREMENTS & FACTS e
NORMALLY REQUIRES LICENSEE RESPONSE:
Basis for denial or reasons For violations Corrective actions to prevent recurrence Date when corrective action will be taken NON-CITED VIOLATION (NCV) e NO LICENSEE RESPONSE e
SEVERITY LEVEL IV VIOLATIONS:
Licensee-identified Corrected l
Non-recurring l
Normally non-willful i
l i
I
O O
O CIVIL PENALTY PROCESS
[
i CRITERIA CONSIDERED FOR SEVERITY LEVEL 111 VIOLATIONS NORMALLY ISSUED FOR SEVERITY LEVEL i AND 11 VIOLATIONS i
f ASSESSEMENT FACTORS I
PAST PERFORMANCE f
i HAS THERE BEEN ESCALATED ACTION IN THE PAST 2 YEARS OR INPSECTIONS?
i CORRECTIVE ACTION l
WAS THE LICENSEE'S CORRECTIVE ACTION REASONABLY PROMPT AND COMPREHENSIVE 7 i
IDENTIFICATION SHOULD THE LICENSEE GET CREDIT FOR IDENTIFICATION?
DISCRETION
[
l IN VIEW OF ALL THE CIRCUMSTANCES, DOES THE MATTER IN QUESTION REQUIRE l
THE EXERCISE OF DISCRETION?
t 7
i l
i
-m
..m.
O NRC ENFORCgENT PROCESS O
~
MINOR Not Subject i
VIOLATIONS to Formal m
^
Enforcement
' Action I
1
/
NON-ESCALATED
/
PROCESS Licensee YES Non-YES YES YES Non-Identified Repetitive Corrected m
Willful Non-Cited issue
?
/
9 Violation Violations
?
9 D
i NO y (NO No NO g g, Violation 1
D
~
i t
Notice of YES Violation
[
m r
j YES Credit for O Civil 4
Corrective Penalty m
) k Action i
YES
?
f NO Notice of Violation ESCALATED PROCESS Non-Severity Level I, il Willful NO>
Credit for identification Base and til Violations SL in Civil
?
YES Penalty
[
y l
NO Credit for Corrective Notice of Action Violation
?
r NO 2 X Base Civil Penalty
[
{
O O
O 1
t i
t i
I ISSUES RAISED CONCERNING NONESCALATED ACTIONS l
Number of violations appear inconsistent with performance i
Cost to licensees to respond to enforcement actions 1
NRC findings are relatively small percentage of licensees' findings and not always the most important I
NOV's require compliance efforts that may be inconsistent with other station
{
priorities I
Not all requirements have a clear nexus between safety and compliance j
i i
Need to integrate inspection, enforcement, and assessment i
i l
I t
i I
9 f
t
~
~
O O
O i
l ACTIONS ADDRESSING NONESCALATED'ENF i
Formalized appeal process EGM 98-006 issued to provide interim guidance to address level IV violations within j
existing Enforcement Policy
[
I
- 1. Emphasized NCV guidance of the Enforcement Policy
[
I
- 2. Responses to cited Severity Level IV violations not required if needed information i
is on the docket.
i
- 3. Licensee corrective action program records should not be " mined" solely for the l
purpose of identifying violations.
i
- 4. Multiple examples of the same violation should be included in a single citation if i
similar root causes such that corrective action is similar
- 5. In considering whether corrective action is prompt and comprehensive, standard is reasonableness, i.e., licensees' actions should be commensurate with safety j
10 l
l i
. -.. ~ - ~
4 w
O O
O i
PHILOSOPHY OF NEW APPROACH i
Recognizes i
1) level IV noncompliances are not in themselves of significant safety concern and 2) power reactor licensees generally have broad and effective corrective action programs j
such that issuance of formal enforcement action with licensee response may be unnecessary resource burden on licensees and staff and may interfere with safety
[
priorities of licensees i
Maintains focus on compliance with requirements but with increased regulatory l
attention on corrective action program thereby addressing precursors before significant events occur Integrated with inspection and assessment; does not prejudice future outcome of new inspection and assessment approaches
^
Need an ability to trend and track noncomphances to assist in determining i
performance trends 11 l
r i
i
~
O O
O PROPOSED NEW APPROACH Utilize three types of noncompliances Violations - Subject to normal enforcement actions Non-cited violations (NCV)- Written up in inspection reports Minor violations - Not normally written up in inspection reports Level IV noncompliances normally considered NCVs, described in inspection report but not cited (NOV) if licensee takes appropriate action commensurate with safety to restore compliance and addresses action to prevent recurrence in corrective action program Level IV noncompliance normally closed out when entered into corrective action program; not individually closed out based on completion of action; use smart sample in inspecting licensee's corrective action program Level IV noncompliance only considered a violation and cited if:
licensee does not take appropriate action commensurate with safety to restore compliance licensee does not enter the noncompliance into corrective action program, or NRC identifies recurring noncompliance demonstrating ineffective corrective action willful 12
O O
O i
PROPOSED NEW APPROACH (Cont'd)
Minor violations continue - focus inspections away from minor noncompliances so l
that violations have a better nexus between safety and comphance i
i By implementation time need to modify NRC inspection guidance to focus more on licensees' corrective action program; adjust inspection frequency and scope of corrective action program beyond Part 50, Appendix B Recognize that noncompliances may recur because corrective actions to prevent l
recurrence have not completed as a result of licensee's schedule commensurate with safety significance l
Continue current documentation of findings in inspection reports j
i Continue to use PIM to trend and track noncompliances to assist in determining performance trends j
Provide for an appeal process I
initially would apply only to reactors Continue efforts on thresholds between level IV and minor violations Driven by assessment and inspection process Reconsider citations based on procedures i
4 13
-o o.
o
?
SUMMARY
i i
New Approach will i
Reduce unnecessary licensee burden Make the regulatory process more risk informed by permitting licensees to i
address violations consistent with safety and risk significance i
i Continue to permit NRC to carry out its oversight repsonsibilities 1
l t
f P
I i
i i
P 14 j
O
,s* " %,,
l UNITED STATES NUCLEAR REGULATORY COMMISSION RISK-INFORMED INSERVICE INSPECTION PROGRAMS FOR PIPING PRESENTED TO:
O THE ACRS MAIN COMMITTEE NOVEMBER 5,1998 Goutam Bagchi, DE, NRR (415-2733)
Richard Barrett, DSSA, NRR (415-3183)
Steve Dinsmore, DSSA, NRR (415-8482)
Jack Guttmann, DST, RES (415-7732)
O 1
i i!O i
j
,p** "%,
l
(
)
UNITED STATES NUCLEAR REGULATORY COMMISSION i
i CONFORMANCE WITH REGULATORY i
GUIDANCE i
l Compliance with Applicable Regulations I
i Defense-in-depth Evaluation O
Safety Margins Risk Impact implementation and Performance Monitoring
O
- f. -..s,
(
)
UNITED STATES NUCLEAR REGULATORY COMMISSION FUTURE ACTIVITIES Schedule of Pilot Plant Application SERs Surry Unit 1 12/31/98 Vermont Yankee 11/30/98 ANO Unit 2 12/31/98 O
ANO Unit 1 07/31/99 Schedule of Topical Reports 4
WOG WCAP-14572 01/31/99 EPRI TR-106706 TBD O
3 4
$~
i;-
i iO f* **%e
(
j UNITED STATES NUCLEAR REGULATORY COMMISSION l
Information Notice i
i j
For licensees that intend to implement RI-ISI j
for piping and do not have a pilot plant j
application currently under staff review, the j
staff will consider authorizing a delay in the l
implementation of the next ten-year ISI
!O program for piping in order for the licensee to develop and obtain approval for the RI-ISI program for piping.
i O
a k
- O f- ~s,,
i UNITED STATES NUCLEAR REGULATORY COMMISSION 1
I Endorse RI-ISI Code Cass_s l
Endorse Code Cases N-560, N-577, 09/2000
{
and N-578 in RG 1.147 O
O s
i
I i
1
!O
)
UNITED STATES NUCLEAR REGULATORY COMMISSION i
i
SUMMARY
l 1.
Scope includes Class 1,2, and 3 piping l
2.
Owner Required to Submit Updated Plan for 10 Year Inspection Interval l
3.
Each 10 Year inspection interval Subdivided into 3
'O Peri ds f,4,and Years
- 4. -
Examination of Sample during Each Interval First 3 years:
16% to 34%
First 7 years:
50% to 67%
First 10 years:
100 %
5.
Sample Selection Class 1: 25% Surface & Volumetric +PressureNisual each refueling outage Class 2: 7.5% Surface & Volumetric +PressureNisual each inspection period Class 3: PressureNisual each inspection period 6.
Locations Selected Based on Stress Level, Cumulative Usage Factor, etc.
O 6
a O
(
)
UNITED STATES NUCLEAR REGULATORY COMMISSION
SUMMARY
OF RI-ISI PROCESS 1.
Scope and Segment Definition 2a. Consequence Evaluation 2b. Failure Probability Assessment O
3.
Risk Evaluation (RRW and RAW) 4.
Expert Panel Safety Significance Categorization 5.
Sample Selection 6.
Implement Program 7.
A reduction in radiation exposure will likely result from RI-ISI O
7
l O
(
j)
UNITED STATES NUCLEAR REGULATORY COMMISSION l
SELECTION OF EXAMINATION LOCATIONS l
l l
f j
(A) Susceptible Locations (100%)
High Failure Owner Defined lO Importance Program Statistical l
Selection l
(3)
Process (1)
Only System l
Low Failure Pressure Test &
Statistical importance Visual Selection Examination g)
Process (2)
Low Safety High Safety l
Significant Significant lo 8
l!O j
,s*"%
(
)
UNITED STATES NUCLEAR REGULATORY COMMISSION i.
i i
UNCERTAINTIES IN CALCULATED FAILURE i
PROBABILITIES l
e The staff performed detailed uncertainty analyses related to calculated estimated failure l
probabilities (NUREG-1661).
!lQ e
The upper bound curve was based on the largest of the 100 failure probabilities calculated from the 100 pc-PRAISE runs for each given cyclic stress q
l level.
The largest uncertainties are for those cases that e
have very low values of calculated failure probabilities. The uncertainties decrease with increasing failure probabilities.
e The categorization of piping segments as high-and low-safety-significant is a function of the degradation mechanism and consequences.
" inactive" versus " active" degradation O
9
~
l
~
!;O i
,p ~m,,
I UNITED STATES NUCLEAR REGULATORY COMMISSION 1
i mechanisms result in significant variation in
{
failure probabilities. This variation renders the impact of the large uncertainties for components j
with low failure probabilities as having have a relatively small impact on the categorization. The j
effects of uncertainties on component categorization can be accounted for through 1
numencal evaluations, such as Monte Carlo lO anaiyses.
i e
j The calculations for components with very low e
j failure probabilities are particularly sensitive to l
the tails of the distributions assumed for input i
parameters such as flaw depths and crack growth rates. The large uncertainties in the calculated l
failure probabilities are a direct results of the fact
]
that the tails of these input distributions are j
based on extrapolations from actual data.
i e
Failure rates for components with high calculated failure probabilities can be assessed for consistency with plant operating experience and 10
O f....,,
(
ja UNITED STATES NUCLEAR REGULATORY COMMISSION l
with industry data bases on reported field failures.
The ability to make such comparisons helps to l
minimize the uncertainties in the calculated probabilities.
l llO t
1 i
a
- O 11 e
i.
l O
g ~.g,s i
j UNITED STATES NUCLEAR REGULATORY COMMISSION
\\,...../
r 1E+00 1 E-01
- -- - #/-
-e'#s 1 E 02 -
- - - - - - -Upper Bound-
/ -
/
k 1E43
- - Failure Probability ~
,~#
)
- ' ~
- ~
j 1 E 04 -
f
- /
-/
o 1 E 05 -
-p n
/
A 1 E - - -##-
/
d-e Lower Bound _
l
]
I:2 1.E - Mediart -
/
\\j
.1 E Os.. Failure Probabibty
/ [ fafure frobabHity j
/
i E
1.E 09 -
-/
/-
y
- - -j i
1 E 10
/-
,c
~
,/
~
" Fatigue of Stainless Steel Piping y
1 E 12
/ -
6-inch Schedule 120 (2 75 inch OD r 0.562 inch Wall) l
- Cumulative Failure Probabilty Over 40 Years 3
1 E-13
-/
/
1 E 14 -
1 E 15 1 E 11 1.E 10 1 E 09 1 E-08 1.E-0 7 1 E-06 1 E 05 1 E-04 1 E 03 1 E 02 1 E-01 1.E
- 00 Best Estimate Failure Probability i
l i
1 ei)
V 12 i
O i
j UNITED STATES NUCLEAR REGULATORY COMMISSION
..../
STAFF REVIEW OF PRA QUALITY How much staff review is required Weight of Evidence (that the quantitative results are sufficiently accurate) l needs to be commensurate with
- O-surden of proof (as measured by the sensitivity of the decision on the quantitative results) l I
l i
6 i
13 i
I y
,f, ~,,,
O- (g; uNrTED STATES NUCLEAR REGUMTORY COMMISSION ISI Burden of Proof-How sensitive are the decisions to the quantitative results PRA is primarily used in the ranking PRA results combined with pipe failure likelihood to place segments into broad safety significant categories Intermediate safety significance category provides increased qualitative engineering attention to segments not clearly Low Safety Significant PRA is not the sole decision criteria orjustification Recognition that " intelligent" element selection is an improvement Inspection strategy and location determination based on safety significance and presence of degradation mechanisms Documented deliberations and decisions by plant O
expert panel 14
O
(
)
UNITED STATES NUCLEAR REGULATORY COMMISSION Checks and Balances using the quantitative results a net calculated risk decrease, inspections to be added to dominate risk systems if dominant segments are not being inspected inspections to be added to all systems with relatively large increases in CDF/LERF between Gection XI and RI-ISI O
inspections to be added to ensure that HSS segments receive enough inspection to provide confidence that failure likelihood will not increase O
15
(
I UNITED STATES NUCLEAR REGULATORY COMMISSION ISI Weight of evidence - that the quantitative results are sufficiently accurate Substantial licensee involvement in PRA developed in support of IPE NRC IPE review identified methodological shortcomings or unusual modeling techniques Limited review of correspondence between current plant O
and PRA models during MR inspections External peer review or Certification process would be very useful PRA update process provides periodic re-evaluation of models and results by PRA and plant staff
,O l
l-16
\\
!:i
!I l:ii>
- it!l!
r_
3 S.
B J
.l 5
A
- E.
m, 43
/
b O 4
O i
N.
e h
o.
d 4
m a
a m
i E4 e
i.
h.
4 a
1 s
e W
s w
ds.
i s
k v
h S.
4 e
h.
ta e
h W
a c
b h
En R
o h
M M
h u
E M.
b u
r p
G.
W h
4 a
B P
S.
M r
t o
O f
ne k
m ro ec wr e
o mfn a
C r
I F
O
o o
o o
l I
l Enforcement Should:
Reflect the move to risk-informed, performance-based regulation t
Recognize (credit) industry's improved safety performance and established corrective action programs Incorporate understandable and objective criteria
/
N,E I i
~
10 0
O i
Enforcement Should Not:
Drive the overall assessment process
}
r Expend XRC and licensee resources on non safety-significant violations e
Be based on new interpretations or expanding views of compliance Be driven by subjective terms NE l l
1
9 0 o
o Areas for Change:
Redefine enforcement categories to SL I, II and III to reflect risk insights i
Process non-safety significant violations through corrective action programs i
I Make enforcement consistent with thresholds l
established for performance assessment process i
i I
/
NEI I
l
\\
l
'O O
O Areas for Change (cont'd):
l l
Reserve escalated enforcement for:
Actual safety consequences Potential safety consequences Certain well-defined significant violations Eliminate subjective bases for escalated i
enforcement:
Regulatory significance" Regulatory concern" Concerns about " management" ggi i
j i
I
I O O O
O l
r Existence of Violation:
1 1
1 Based on clear requirements l
E Not:
I
- vague criteria or evolving knowledge
- backfits t
- random equipment failures j
i l
- minor procedure violations 1
i l
l
- -. -~
O O
O PROPOSED ENFORCEMENT PROCESS
,/
N
's CP
/
Actual
'N,
,/
Consequences:
d TABLE A
/N&-
N' (Adjust)
[
~~"'~~^~-
l 4
Assign Severity Adjust CP y
Potential Level McrRed including Consequences:
"co(nsideration
- II Enforcement Action TABLEB r
of 4
~ ~ - ~ ~
(Adju p
,,\\
quantitative risk data)
/ NOV, \\
t.__
_ f N
No Press /
I
/ No CP s
No Other defined s
N 8,eleas,a '
j l
categories, e.g.,
's/
50.5, 50.7. 50.9, i
safeguards, FFD, l
TABLE C Non-Safety Violation of reg ion or other Yes->
[
Cir u sta ce Infra io N St) 7 l
I Yes No i
h NRC inspector No Enforcement Action documents entry into
' No Enforcement.
licensee CAP t
. Action l
l l
t
,r Licensee enters into CAP l
l
o o
o L
Table A - Actual Consequences This table applies to violations of NRC requirements having actual radiological safety consequences.
i CONDITION SEVERITY LEVEL Offsite Release Due to I, II, III, or NSSI (criteria to Reactor Operating Event be developed)
I, II, III, or NSSI (criteria to
)
Offsite Release of Contaminated Material be developed) i Onsite Releases or Exposures I, II, III, or NSSI (criteria to be developed) i
(
l l
4 l
1
TABLE B - POTENTIAL CONSEQUENCES I
i V
\\
This table applies to reactor operations, maintenance, and design activities (including programs and procedures that apply to those activities) that are found to be in violation of NRC requirements. As referred to in this table,
" intended safety function" refers to the function (s) that caused the system te be classed as a " risk significant" or "high risk significant" system. Inability of all trains to perform safety function refers to cases in which each of those trains could not perform intended safety function pertinent to the event or circumstances at issue for a period beyond the allowed outage time (AOT) that would apply to each train if other trains were operable. For single train systems with redundancy provided by diverse trains or systems, the single train system and the diverse redundant train or system should be treated as redundant trains of a single system.
CONDITION HIGH RISK RISK NON-RISK SIGNIFICANT SSC SIGNIFICANT SSC SIGNIFICANT SSC All trains of system fail to I*
II*
NSSI perform intended safety function when actually j
called upon.
All trains of system cannot II*
III*
NSSI perform intended safety functions.
All trains of system cannot III*
NSSI NSSI
~'N (V
perform intended safety functions under some circumstances.
One train ofmultitrain II*
III*
NSSI system fails to perform intended safety function when actually called upon.
Once train of multitrain III*
NSSI NSSI system cannot perfonn intended safety function.
One train of multitrain NSSI NSSI NSSI system cannot perform intended safety function under some circumstances, or system or systems degraded but capable of performing intended siTety.
- If reasonable compensatory capability was available, or compensatory action was taken, lower O
one severity level.
G
- O I
t
D-R-A-F-T
{')
October 13,1998 TABLE C l
DEFINED VIOLATIONS WITH DETERRENCE IMPLICATIONS l
1 Table C would address certain violations where enforcement action is important for deterrence reasons, apart from any remedial purpose. These include certain l
v'.olations of 10 C.F.R. Section 50.5 (deliberate misconduct), Section 50.7 (employee
[
(.iscrimination), and Section 50.9 (complete and accurate information).
l l
Table C may also address certain violations related to licensee programs that are not susceptible to treatment on Table B. However, use of Table C should not undermine the philosophy that the enforcement program should not assume the performance assessment function.
With respect to programs and program implementation, Table C would focus on issues with realistic potential consequences because there is a failure to achieve a required program element, not on isolated implementation issues.
i rS The following objectives and guid<. lines would be employed in applying V
Table C:
l Table C addresses enforcement actions issued to power reactor licensees.
i i
Severity Levels of actions issued directly to individuals (including l
licensed operators and licensee employees) would be assessed separately in accordance with the guidelines of Section VIII of the Enforcement Policy.
Table C reflects the philosophy that licensees are held accountable for the acts of their employees, but should not be held accountable for matters outside their control such as random equipment failures or certain examples of personnel wrongdoing.
As with all violations, the presumption is made for Table C that the licensee has initiated corrective actions through the Corrective Action Program or other initiative.
l n
y/
l..
I t
e I -
d Condition Guideline U
Licensed Operator Attentivenessi Multiple instances ofinattentiveness by licensed operators in control room SLII Inattentive licensed operator in control room SL III e
1 e
Section 50.59 Failure to apply for required NRC approval SL III due to flawed safety evaluation (change /USQ later determined unacceptable)
Change made without Section 50.59 safety Assess based on evaluation Table B or C Failure to apply for required NRC approval due to flawed Section 50.59 safety evaluation (change /USQ otherwise acceptable)
NSSI h>
Deliberate / Willful Violations 1 Willful violation by licensee management Assess based on e
individual (licensed operator, or manager-Table A, B, or C and i
level and above), with actual or potential safety raise 1 Severity consequences Level Willful violation by licensee management SL III e
individual (licensed operator, manager-level and above), with minimal potential safety consequences Willful violation by low ranking individual, Assess based on with actual or potential safety consequences Table A, B, or C Willful violation by low ranking individual, NSSI with minimal potential safety consequences i
l l
Severity Levels in this category may be adjusted iflicensee management has o
taken reasonable corrective actions, including disciplinary action.
b 2
l '
p _
Employee Discrimination
- (7 Licensee official culpable?
V Remediated with individual (i.e., settlement)?
Chil' ling effects addressed?
Incomplete / Inaccurate Information
. Incomplete or inaccurate infonnation on SL II a matter that would have warranted an immediate NRC order required by public health and safety if complete / accurate Incomplete or inaccurate information on SLIII a matter that would have resulted in other regulatory action (enforcement, withholding an approval)if complete / accurate Deliberately incomplete or iriaccurate information Raise Severity based with the knowledge of senior licensee management on above examples by 1 Other examples of deliberately inaccurate or.
See" Deliberate /
incomplete information WillfulViolations" above; consider individual actions taken by licensee, NRC Other isolated example of incomplete NSSI or inaccurate information which would not have resulted in regulatory action if complete or accurate -
Fitness For Duty Licensed operator at controls makes errors See Table A/B/C; that cause or exacerbate an alert or higher consider raising one i
level emergency, and later tests positive -
SL, but also consider with actual or potential safety consequences actions by licensee and NRC against individual qV 3
t
~ Licensed operator at controls makes errors, SL III, but consider
]b and later tests positive - with minimal safety actions by licensee
- consequences and NRC against individual Failure to implement a significant element of SL III e
the FFD program as required by NRC regulations Isolated FFD program implementation NSSI deficiencies Security Actual act of sabotage in which security SLI system did not function as required and, as a result, there was a significant event such as accidental criticality, theft of formula quantity of SNM, or unauthorized production of SNM Security system did not function as required SLII e
and, as a result, there was an unauthorized
~
entry into vital area by an individual who s
was a threat, or there was a theft of SNM of moderate strategic significance Failure, inability, degradation, or other SL III e
deficiency in the established security systems or procedures such that there was a reasonable potential for unauthorized personnel who represented a threat to gain access to a vital area or to circumvent systems to introduce contraband, in such a way that there were significant potential consequences Isolated safeguards failures, including both NSSI system design and procedure implementation deficiencies O
t 4-
~
6 m,
a
~ ~-
,na
I M
(]
Significant failure to effectively respond to a general SLI U
emergency Significant failure to effectively respond to a site e
emergency SLII Significant failure to effectively respond to an alert SLIII e
Significant failure to maintain equipment and/or systems SLIII e
such that a planning standard involving assessment or notification capability is compromised Failure to meet or implement an emergency NSSI planning standard not directly related to assessment or notification capability
. Isolated failures to implement emergency NSSI response procedures Fire Protection (Annendix R)
Actual fire event resulting in damage that Varies depending p
e V
a Fire Protection program element should upon consequences; have prevented assess based on Table A or B ifpossible Failure to implement more than one element Varies depending e
of the Fire Protection program as required by upon risk; consider NRC regulations more tf an 1 SL III Failure to implemmt one element of the Fire SLIII e
Protection prograrn m required by NRC regulations Isolated failures in the implementation of the NSSI Fire Protection program These violations are separate from, and in addition to, any other violations that
,V.
relate to causing an event at a nuclear plant.
i 5
9 l
O o
o Risk Consideration:
Incorporated into Table B
- AEOD Equipment Categories A~dditional consideration of risk data where available
- CDF, containment failure probability, as appropriate l
- b)
. o o
o CP Adjustment:
i Discretion to adjust / eliminate civil penalty:
- licensee-identified during extended shutdown
- licensee-identified as part of voluntary initiative
- mitigating circumstances in Section 50.7 cases
- Enforcement free for two years r
- In Performance Assessment " green zone" l
N,k I 7
i
?
-