 |
Search Options | |||
| Index | Site Map | FAQ | Facility Info | Reading Rm | New | Help | Glossary | Contact Us |  |
|||
Home > Electronic Reading Room > Document Collections > Commission Documents > Commission Meetings, Slides, Transcripts, Meeting SRMs, and Full Written Explanation for Closed Meetings > 1997
>
02/13/1997
1
UNITED STATES OF AMERICA
NUCLEAR REGULATORY COMMISSION
- - -
BRIEFING ON OPERATING REACTOR OVERSIGHT PROGRAM AND
STATUS OF IMPROVEMENTS IN NRC INSPECTION PROGRAM
- - -
PUBLIC MEETING
Nuclear Regulatory Commission
One White Flint North
Rockville, Maryland
Thursday, February 13, 1997
The Commission met in open session, pursuant to
notice, at 2:05 p.m., Shirley A. Jackson, Chairman,
presiding.
COMMISSIONERS PRESENT:
SHIRLEY A. JACKSON, Chairman of the Commission
KENNETH C. ROGERS, Commissioner
NILS J. DIAZ, Commissioner
EDWARD McGAFFIGAN, JR., Commissioner
. 2
STAFF PRESENT AND PRESENTERS SEATED AT THE COMMISSION TABLE:
JOHN C. HOYLE, Secretary of the Commission
KAREN D. CYR, General Counsel
HUGH THOMPSON, JR., Acting EDO
FRANK MIRAGLIA, Acting Director, NRR
FRANK GILLESPIE, Director, Inspection & Support
Program, NRR
ROY ZIMMERMAN, Associate Director for Projects,
NRR
. 3
P R O C E E D I N G S
[2:05 p.m.]
CHAIRMAN JACKSON: Good afternoon, ladies and
gentlemen. We are pleased to have the NRC staff here to
brief the Commission on the operating reactor oversight
program. This is the first of three Commission meetings
addressing different aspects of the operating reactor
oversight program.
The subject of this meeting is the Status of
Improvements in the NRC Inspection Program. The subjects of
the other two Commission meetings are Analysis of
Quantifying Plant Watch List Indicators, at which time the
staff will discuss the Arthur Anderson review of the senior
management meeting process, and the Millstone and Maine
Yankee lessons learned as they relate to policy issues
affecting the regulatory process.
These meetings are scheduled for February 18 and
19, respectively.
I would just ask you, the staff, as you discuss
areas of improvements, please provide for the Commission the
results which have been achieved and/or examples which
demonstrate improvement or are aimed at demonstrating
improvement, so that if activities are in progress, you can
provide a time table. Please provide a time table for the
completion of the activities.
. 4
I understand that copies of the slides are
available at the entrance to the room. Unless fellow
Commissioners have any comments, please begin, Mr. Thompson.
MR. THOMPSON: Thank you, Chairman Jackson and
Commissioners. Before we go into the presentation, I should
remark for you and members of the public. In our efforts to
upgrade the TV coverage of the Commission dialogue, we have
downgraded. The motherboard didn't fit up and we burned it
out. So we now have a camera in the corner over here. We
will have a general view of the proceedings, but slides will
still be presented. Hopefully the public will be able to
follow along without much trouble.
With me at the table this afternoon are Mr. Frank
Miraglia, who is the Acting Director of NRR; Mr. Frank
Gillespie, Director of the Division of Inspection and
Support Programs, and Mr. Roy Zimmerman, Associate Director
for the Reactor Projects.
We are here to brief the Commission on the
operating reactor oversight program, as you said. The
reactor oversight program is continuously evolving.
Self-assessments, both external and internal, have
identified problem areas which are being addressed by a
number of actions.
These actions include improvement in inspector
training and guidance as well as our performance assessment
. 5
methodology and criteria. These actions, combined with
initiatives that were underway prior to the more recent
review, will enhance the staff's ability to identify safety
issues and assess licensee performance.
In addition, the recent reorganization and
appointment of the Deputy EDO for Regulatory Effectiveness,
Mr. Ed Jordan, will play a prominent role in future
oversight of the reactor program.
This completes my opening remarks. I would like
to turn the briefing over to Frank Gillespie, who will do
the major portion of the briefing dealing with inspection
efforts, and Mr. Zimmerman will address those portions
dealing with the project management.
CHAIRMAN JACKSON: Thank you.
MR. GILLESPIE: Since Hugh has already covered the
objectives, if I could go right to slide three, which is
just to outline the organizational structure of the overall
program.
The focus today is on the inspection program. The
other presentations next week are really going to touch upon
licensing aspects and other aspects.
[Slide.]
MR. GILLESPIE: NRR as the headquarters function
develops policies and procedures, but we are by our own
procedures very closely integrated comment-wise at every
. 6
step with the regional offices because they are the
implementers of the program and they do the chief
implementation and the real inspection.
The regions do implement the majority of the
program.
Other organizations provide independent review.
The new Deputy Executive Director position is
expected not only to continue doing what AEOD did with the
lessons learned from DETs and lessons learned from case
studies, but that will also be very much, we would expect,
enhanced in our interface with them.
CHAIRMAN JACKSON: Let me ask you a question.
What are some of the feedback processes to which you refer,
or feedback mechanisms in the second bullet?
MR. GILLESPIE: The regions are not hesitant to
tell us that we have made an error. We get feedback on the
ability to carry out a procedure where we actually in
coordination with the regions have published an inspection
procedure and we are finding it's not getting an objective
we need. We get feedback from the senior residents and
residents. We periodically go out and talk to them
individually and ask how are these aspects of the program
going.
A most recent example. I did a follow-up visit to
Maine Yankee on the testing question: What did it take to
. 7
find that one broken wire by that contact? What would it
really take? How would that impact the program? Could we
look in that much detail at every other facility?
So we are going out trying to get direct feedback
from the people carrying out the program, which will tend to
lead us into maybe a more in-depth audit of what we are
doing and how we are doing, and potentially then a rewrite
of a procedure or re-promulgation of a manual directive.
MR. ZIMMERMAN: If I can add a thought. When
senior managers from the region and from headquarters also
go out and visit the sites it is typical for them to have
dialogues with licensee management as well and receive
feedback at that time on their view on the inspection
program as well.
CHAIRMAN JACKSON: But you don't have formalized
feedback processes as such.
MR. MIRAGLIA: We have both, Madam Chairman. I
think what Frank and Roy were talking about are somewhat
informal. As part of the agency response to the reg impact
survey, there are a number of mechanisms for getting
feedback: requiring the inspectors' supervision to get out
to the facility and explore with licensing managers their
perspective of how the program is going, strengths and
weaknesses, and there is a feedback mechanism.
Is it an annual report now, Frank, to the
. 8
Commission on some of those aspects as well?
MR. GILLESPIE: Yes.
MR. MIRAGLIA: So we have formal processes as
well.
CHAIRMAN JACKSON: But then you say other
organizations provide independent review. What does this
mean? Is this headquarters organizations reviewing the
regions?
MR. MIRAGLIA: I think Frank's intent and the
intent on that slide is to say that we have other
independent audits from, say, the GAO, our own IG, that also
provide oversight, and we look at that, and that's a
feedback mechanism.
CHAIRMAN JACKSON: We do have this new Deputy
Executive Director for Regulatory Effectiveness, et cetera,
and you said that this organization will play a prominent
role. Do you intend to have some kind of internal
assessment? There is self-assessment by the organizations
themselves, but do you intend for this organization to
provide --
MR. THOMPSON: Joe Callan and I have discussed
some of the ways we would do our own assessments. As you
know, in the materials area we do kind of impact program
review where we go out and review against a set of criteria
the regional offices. We are looking at potentially using
. 9
that model as an approach to go by and look at the regional
offices with respect to the reactor program. We have not
formalized anything specifically on that yet. I am looking
forward to Joe coming up here so I can be looking at things
like that. I have not had an opportunity recently to look
at those type activities.
MR. MIRAGLIA: But we do have our own
self-assessments. I think internally in the program we
would go out and look at our program and self-assessments.
We have done that on a periodic basis. That would include a
visit to a facility or to facilities within a region,
discussions with the residents, discussions with the
licensees, and looking at headquarters.
So we do have those kinds of assessments as well,
and I would think as a program office that we would continue
to have self-assessments based upon routine
self-assessments, and if there are indications from other
places that we have weaknesses, that we would try to go out
and look at them.
CHAIRMAN JACKSON: It strikes me -- and I think
Commissioner Diaz has a comment or a question -- that what
you have essentially described is a mechanism where the
program offices assess what goes on in the regions, but the
program offices themselves do those self-assessments. The
question is, is there any regularized way of having a
. 10
quasi-independent review of the program offices' role in
reactor oversight?
MR. THOMPSON: I believe the answer to that is
yes. In the materials program we do look at some of the
activities in headquarters as part of the overall
programmatic review, kind of doing self-assessments with
those same criteria we use to evaluate the regions and the
agreement states.
CHAIRMAN JACKSON: All I am really talking about
is consistency of how assessments are done. There is
self-assessment and then there is assessment by others, peer
assessments. So you have described a food chain where
everyone self-assesses and everybody is assessed except the
program offices in terms of someone else assessing them.
MR. THOMPSON: Obviously the Deputy EDO for
Regulatory Effectiveness is a --
CHAIRMAN JACKSON: That's the intent here.
MR. THOMPSON: That would not be the only thing.
We would obviously be doing some of our own assessment. I
don't want to speak for Ed. He can obviously speak well for
himself, but that is the intent. That organization would
also look specifically at how the program offices are
carrying out their functions, and that program is being
developed now by Ed and his staff.
CHAIRMAN JACKSON: Commissioner Diaz.
. 11
COMMISSIONER DIAZ: Just a comment on
Mr. Gillespie's statement that the regions are very candid
in providing criticism of the office. I hope that is not an
open loop, that you close the loop by being equally critical
of the regions. In this way you actually close that loop in
the feedback, but it was not implied in your statement.
They actually do come back to the regions; is that correct?
MR. GILLESPIE: Yes, it does.
CHAIRMAN JACKSON: Commissioner Rogers.
COMMISSIONER ROGERS: On that word "oversight" in
the Deputy Executive Director of Regulatory Effectiveness,
do you really mean assessment there? Is that what you mean
by oversight?
MR. THOMPSON: That is my sense of it. It is not
any kind of day-to-day oversight activities.
COMMISSIONER ROGERS: You really mean assessment
when you say "future oversight"?
MR. THOMPSON: Right. I think that's just the
title.
COMMISSIONER ROGERS: It's just a question of what
the relationships are between the organizations.
CHAIRMAN JACKSON: You mean assessment.
MR. THOMPSON: Yes.
[Slide.]
MR. GILLESPIE: If we go to slide four, which we
. 12
have basically covered, we do have a formal process of
assessing the regions with a team of people going in each
year with predetermined criteria. We in fact generate a
written report which we supply to the region for comment.
We do this every spring.
We also are assessing not only the region but
ourselves when we are looking at it, because inevitably you
will find two types of problems when you are looking at
someone implementing your program.
One is they have implemented the program and it is
not the program you thought you wrote, and therefore you
have program problems where clarifying guidance is needed or
making some adjustments. On the other part you might just
have problems with implementation.
We do visit, as Frank said, several facilities,
interview people in the regions, the division directors, and
we are getting two kinds of feedback. One is programmatic
feedback and one is on the direct implementation in that
region itself.
And we compare across regions to look at
consistency in implementation to see ourselves if we are
getting the depth and scope we thought we should get. That
is done once a year with every region, and we do share the
results. We are very open with them on that.
We do special reviews also where we see particular
. 13
problems. An example would be we saw particular problems in
how we as an agency were dealing in inspection space with
safety assessment quality verification. So we did a look at
safety assessment quality verification, how we are carrying
it out, is it meeting the objectives, and we are now taking
some corrective actions on that, and in fact recently issued
some changes to inspection procedures where basically we
felt that the program was being carried out more on a
piecemeal basis. And the program allowed that.
The procedure allowed different pieces to be just
added up to say we meet the whole, but from recent
inspection results we are finding that when you had a group
of people go out and do the procedure, all at once you had a
much better picture of safety assessment quality
verification of a licensee, and that came through in some
inspections we did focusing on this at both Fermi and
Crystal River.
So not only do we do the annual audit, but we do
special audits where we see a problem starting to evolve or
have some insights from some other source that we need to
look at it. Then we also do follow-up on that to see that
we are achieving the end we thought we got with it. That is
done on a continuing basis.
We do do direct observations with SALP, PPRs.
This all comes from the same basic branch and organizational
. 14
unit, and it is all integrated together. From those
observations you tend to get the special reviews.
CHAIRMAN JACKSON: Who actually performs the
observations?
MR. GILLESPIE: They are all coming out of the
inspection program branch. That is our primary focus. We
also have all the managers. Basically division directors
and above in NRR periodically go out and sit in on SALP
meetings as observers to see how they are going on. We get
feedback also from that process and we get feedback because
anyone who is a SALP board chairman in any region is
supposed to observe a SALP board in a different region. So
we also get feedback from them. We have got it covered from
different layers. When the information coming back from
these people starts to correlate together, then we are off
taking action.
CHAIRMAN JACKSON: How do you solicit feedback
from industry and the public?
MR. GILLESPIE: As Frank said, we have a formal
process in a form. It is typically dominated by the
regional managers who are going out to see what the
residents are doing or visiting the site which asks them to
sit down one on one with a licensee representative and
basically try to ask him, does he have a problem?
CHAIRMAN JACKSON: Is this form used on a
. 15
consistent basis?
MR. GILLESPIE: I would not want to say it is used
on a consistent basis, but it is used with regulatarity.
I'm not trying to evade your question, but it is spotty how
we get it back. We annually put a Commission paper together
and report the results of what comes out of those reported
forms.
At a higher level is the EDO's process, and each
region has a process for very senior managers at nuclear
facilities to call in if they have complaints about actions
taken by residents or inappropriate actions or any problems
at all with how we are carrying out our program.
CHAIRMAN JACKSON: Is our solicitation of feedback
from the public as formalized?
MR. GILLESPIE: No, it's not.
CHAIRMAN JACKSON: Are you thinking about that?
MR. GILLESPIE: I hadn't been. It is not.
MR. MIRAGLIA: It has been spotty.
MR. GILLESPIE: It has been spotty.
MR. MIRAGLIA: There have been broad procedures
such as the maintenance inspection procedures. There were
workshops. On some occasions we have had broad workshops on
those kinds of activities.
MR. GILLESPIE: Out of fairness, the feedback from
the public we get here is --
. 16
CHAIRMAN JACKSON: And newspaper articles.
MR. GILLESPIE: No. Actually, it's a little more
structured than that. It is things like the UCS write in
and we get appraisals from them. They are happy to appraise
us. There is a person at the University of Syracuse who
quite frequently gets computer runs.
CHAIRMAN JACKSON: The real answer is that at this
point in time it's not formal.
MR. GILLESPIE: It's not formal.
MR. MIRAGLIA: It's not formal.
COMMISSIONER DIAZ: Would you say that the
feedback from the licensees, the industry is consistent in
the sense that if there is a problem they do come and say "I
have a problem with this"? Is that done on a consistent
basis? Is that dialogue established and that feedback
mechanism clear and unimpeded?
MR. GILLESPIE: I personally believe it's very
unimpeded and it's very clear. What I would have difficulty
answering is, does any particular utility have an inhibition
for using it?
I think on our side we hope we have destroyed any
blocks to someone calling up. In fact, only last week we
had an immediate call where an inspector said something on
site which was probably out of context and out of line. The
call came in immediately, and we have taken care of the
. 17
situation.
For the majority of licensees, I think it's both
clear and unimpeded.
MR. MIRAGLIA: As a matter of context and history,
the form and the feedback mechanism of managers going out
was a result of the regulatory impact survey that we
conducted back in 1989 and 1990.
One of the corrective actions of that is there was
a concern -- industry voiced that concern -- about the
retribution issues, and there is reluctance to complain
because of their input to SALP and other things that affect
licensee performance. So there is a reluctance. That issue
is out there.
As a result of that survey, we took the corrective
actions of having the survey form, the periodic report to
the Commission. We had training sessions for our
inspectors. We have upgraded the fundamental of
inspections. There were a number of corrective actions.
Notwithstanding those activities, the Tower
report came out in 1992 and raised the issue again of
concern and reluctance to bring forth those kinds of issues
to the agency. As a result of that, another level was put
in, which is the management implications group where senior
managers from the utilities were encouraged to call to the
EDO and the EDO staff to raise issues at that level. That
. 18
has been used and that has been in place for about a year or
two years and has been used in a limited kind of way.
MR. THOMPSON: We don't get lots of calls in that
context. I think each utility executive as they have their
communications links established with the regional staff and
with headquarters staff probably make their own judgment as
to whether they raise issues and select what issues they
raise. I think if there is a truly serious issue, they
don't hesitate to raise it. It is those that are kind of
borderline, would be my guess, that some may have reluctance
to raise.
COMMISSIONER DIAZ: Are you satisfied that we have
provided guidance at every level that it is important and
necessary for the licensee to communicate reservations that
they have and that is all we can do?
MR. MIRAGLIA: The answer is yes. We say there
are many vehicles. We encourage the licensees to use those
vehicles, and for these other cases here's another mechanism
to be used.
COMMISSIONER DIAZ: Thank you.
MR. GILLESPIE: Going to slide five.
[Slide.]
MR. GILLESPIE: There is almost a continuous,
ongoing relationship between the Inspector General and the
program, because there is almost always some aspect where 50
. 19
percent of the agency is being reviewed. So there is
actually a very good relationship in the suggestions made by
the Inspector General on the program as they look at
different aspects and to try to understand the
interrelationships between one recommendation that they make
on one more limited scope investigation or audit as it might
be applied across the board.
The General Accounting Office, that was only
yesterday. We always seem to have a GAO audit of some kind
going on. In fact, one of the people working on the one
that is going on right now was also on the South Texas one.
So we have periodic feedback almost of a constant assessment
nature coming from GAO also.
Industry and public feedback. Predominantly
industry feedback. Public feedback is very indirect. It
tends to be different groups that come in. It is not
systematically collected.
[Slide.]
MR. GILLESPIE: Going to slide six, this list in
bullet form are the issues I was going to try to cover today
to give some sense of what we are doing to improve the
program in both a problem statement in each of these areas
and some idea of short- and long-term action, in some cases
a short-term action having already been achieved.
[Slide.]
. 20
MR. GILLESPIE: Slide seven, the identification of
design issues. We have broken down the problem into several
pieces here to somewhat try to align with the actions we are
taking.
Insufficient design considerations in the
inspection program and lack of design of expertise in the
staff. I will split that as I discuss it into both a
long-term and the front part of that problem that we are
trying to deal with a little more in the short term. The
last part of that sentence is a longer term aspect.
Lack of systematic incorporation of design
inspections into the core or our regular inspection program
so that we have a repetitive nature of going back with some
periodicity and looking at design.
CHAIRMAN JACKSON: Is there an inspection module
that is in fact focused on design inspection?
MR. GILLESPIE: Yes, there is.
MR. MIRAGLIA: There is one in development.
MR. GILLESPIE: We have one in development. We do
have an inspection module, several in fact, that are
specifically focused on design, but they are not part of the
mandated core program. Now we are looking at how to make
that digestible and actually get it into mandatory program
so that we come up with a periodic and systematic look at
it. It is right now a procedure that is used more in a
. 21
reactive nature when we have some insights that there is a
problem.
The next one is insufficient information from
licensees on the status of design basis reconstitution. It
goes down to the 50.54(f) letter which we have recently put
out, and we are starting to get responses back.
And lack of availability of design information,
which is something we see when we do visit the sites, the
ease of even the licensee coming up with his design
information for us to review.
CHAIRMAN JACKSON: Who performs engineering
inspections in the regions?
MR. GILLESPIE: We start from the top down. There
are two divisions in each region. There is a Projects
Division and then there is the Reactor Support Division,
which is basically the engineering division. So engineering
inspections are really focused on regional specialists.
Our residents are more generalists. We are
focusing and are continuing to keep them focused on day-to-
day operations where the idea that avoiding a transient, the
idea of stable operations is a very important aspect of what
we do and is a significant contributor to the improved
industry performance as we look at the performance
indicators that have come down so much in the last ten
years. So we are really dealing with the engineering
. 22
inspections being done out of the regions with regional
specialists in this area.
That will get me to my last bullet on long-term
actions, the basic tools that we have been able to give
these people.
The other piece is we tend to supplement these
people with contractors, which I will discuss as I go
through the short-term actions here on the first bullet.
CHAIRMAN JACKSON: We have decided that is the
best way to proceed?
MR. GILLESPIE: I think as a short-term decision
we have decided we definitely -- lessons learned from
everything we have seen is, if we don't look at something,
then there is a chance that the industry is going to back
off and not look at it. So initially, and not wanting to
lose the gains we have made in operational safety, we have
maintained that and we are trying to maintain that resident
core of generalists focused on operational safety.
In addition, the types of problems that have come
out of Haddam Neck, that have come out of Millstone and
Maine Yankee and Crystal River really were identified by the
narrow specialists taking the time to really dig into
something. So the nature of the problem that we are finding
in the design area calls for a different kind of person than
the resident also by background, training and by current
. 23
capabilities.
COMMISSIONER DIAZ: Using the language I will use
with the licensee when we assess an area, we find strengths
and weaknesses. Would you quickly tell us what is the
strength that we have in our identification of the sites?
What are the main strengths where we say we can do this? I
know you have identified the problems, but what are our
strengths? What are our capabilities?
MR. GILLESPIE: I think our strength is definitely
operational safety.
CHAIRMAN JACKSON: No. He's asking what are the
strengths in the design area.
COMMISSIONER DIAZ: In this design issue.
MR. GILLESPIE: Let me jump to some of the
results. I believe the EDO recently sent a short note up on
the result of the architect/engineer inspections. Actually
the results are very consistent with results of past safety
system inspections, which are very design-oriented, that we
have done. They are not atypical type results.
The strengths we are finding --
CHAIRMAN JACKSON: Identify whether the strengths
you are finding are licensee strengths or our strengths.
He's interested in our strengths.
COMMISSIONER DIAZ: You said we have these problem
areas. What are our strengths?
. 24
MR. GILLESPIE: Our strengths right now as we are
carrying the program out, I think is using the, quite
honestly, limited number of people with design background
that we have as team leaders to organize how we leverage
ourselves, which is my first bullet here on the
architect/engineer inspections. We have leveraged ourselves
tremendously with contractors on these inspections.
MR. MIRAGLIA: I think maybe your question is more
our strengths and our focus has shifted to operational
safety and that is where our strengths are. The residents
have that operational focus. Their training is in the
operational safety focus. At one point in time, when we
were doing lots of licensing, lots of construction, we had
more design specialists within the context of the agency,
and we are drawing on those now.
CHAIRMAN JACKSON: I think the simple way to
answer the question is, do we have any strengths in that
area?
MR. MIRAGLIA: In design?
CHAIRMAN JACKSON: Yes.
MR. MIRAGLIA: Yes.
CHAIRMAN JACKSON: What are they?
MR. MIRAGLIA: I think we have a large number of
qualified engineers who have worked at AEs, but they are
smaller in numbers than they were perhaps years ago. The
. 25
question is, how can we use that smaller amount and how
could we augment those?
COMMISSIONER DIAZ: Are we getting them somehow so
they become a strength?
MR. MIRAGLIA: Yes. And the longer term issue is,
if we need to get more, how do we -- the training. You will
hear we have to get there from here.
CHAIRMAN JACKSON: Maybe a strength is that we
have a few good men or women.
MR. MIRAGLIA: Yes.
CHAIRMAN JACKSON: Is that what you are telling
us? I don't want to put words in your mouth. That you have
learned how to leverage them through these
architect/engineer enhanced inspection teams? Am I putting
words in your mouth?
MR. MIRAGLIA: That's absolutely correct.
CHAIRMAN JACKSON: I don't know if that answers
the question.
COMMISSIONER DIAZ: That certainly answers my
question.
CHAIRMAN JACKSON: He wanted it from you.
MR. MIRAGLIA: Yes.
COMMISSIONER DIAZ: The identification of problems
must come from a strength. I think it is very important
that we assess our strength even before we assess our
. 26
weaknesses.
MR. MIRAGLIA: Yes. As an indication of that,
Commissioner Diaz, some of the special teams that we sent
had those strengths on it to look at the design areas. In
terms of the special teams that we sent to Millstone and
Haddam Neck, special teams that we sent to Dresden, and the
special teams that we sent to Maine Yankee, they had NRC
specialists in that area and we augmented those as well with
contractors. So, yes, we have. We are not totally absent
of that strength within the context of the staff.
MR. ZIMMERMAN: I would add that in addition we
guide those contractors and take their findings that may be
raw and put them into regulatory context and determinations
on whether enforcement is appropriate or not. We deal a lot
with the raw findings that the contractors provide.
COMMISSIONER DIAZ: I see that as a strength. It
should be clearly spelled out as a definite strength.
COMMISSIONER ROGERS: Before we leave that, you
say we put these into regulatory context. Does that involve
a relative risk consideration?
MR. ZIMMERMAN: I would say yes, and we are
getting better with the training that we are providing to
our staff, our ability to use risk insights and also know
their limitations are improving. So we try to always look
at things from a risk perspective, take the findings and
. 27
marry the two, but there is clearly room for us to continue
to grow in that area.
MR. MIRAGLIA: And I think risk is a factor in a
number of areas, Commissioner Rogers. In terms of deciding
what systems to inspect, we try to use the risk informed
insights to look at those issues.
In terms of regulatory space, one has to try to
say safety significance, risk significance. That is a part
of the enforcement policy, and we are looking at ways of
enhancing the enforcement policy with respect to risk
significance.
So risk is a factor and an element at the various
stages of the regulatory process.
CHAIRMAN JACKSON: I think Commissioner McGaffigan
has a question.
COMMISSIONER McGAFFIGAN: One of the things
Mr. Gillespie just said, and I took it down, is if we don't
look at something, industry will tend to back off and not
look at it. That's a fairly profound insight and it says
something about risk. Maybe you could say something about
the compounding of things. If we only look at the most
safety-significant, the most risk-significant systems and
they were working perfectly, can things fall apart by
looking at non-risk-ignificant things that compound and turn
out to be more safety-significant?
. 28
MR. MIRAGLIA: That's a challenge. The challenge
is another balance. I think we cannot focus strictly on
risk-significant things.
As an example, spent fuel pool cooling, which was
an issue that got pulled at Maine Yankee. From a risk-
significant point of view, we perhaps should have been
paying more attention to that area, and we were focused on
other areas. That's a lesson learned.
There has to be a balance not only between
operations and safety, but you have to have enough in your
program that is sampling enough areas, both risk-significant
and perhaps less risk-significant, but providing insights to
the robustness of licensees' programs and their ability to
comply with the regulations. We have to look at both, and
that's another balance that is going to have to be looked at
in terms of the inspection program. It can't be absent and
just risk dominant only. You have to look at a little bit
of everything.
MR. GILLESPIE: We've got a bullet that isn't in
my viewgraphs. I will try by way of example to illustrate.
One of the things that we are looking at is the basic
inspection procedure that most of the resident activities
are covered under in operations. We require the residents
to do some things, and we will hear anecdotally back from
the residents: Why am I looking at this? I never find
. 29
anything wrong.
We have tried to maintain this balance that Frank
talked about in the program of touching all the necessary
activities and relate that same philosophy to the
engineering of the facility and just take a risk analysis
and recognize that some of the major weaknesses in our risk
analysis are the cumulative assumptions that went into it.
Every resident and every senior resident basically
some place in their office will have the IPE chart that has
been generated on risk-significant system, least to most to
most to least risk-significant system. Yet when you look at
that, if you spent all your time on the top two systems, you
might have a highly risk-significant problem because the
support systems that are assumed to work are going to be
flawed.
It is clearly not, to me, a simplistic argument.
In achieving this balance people sometimes take us to the
extreme. Well, that is not a risk-significant system. Yes,
but HVAC happens to be a very important support system, if
you have an accident, to maintain the environment in the
control room.
Other things that we might be looking at, for
example, the maintenance rule, is looking at the reliability
of a particular component. Well, there are some assumptions
in the risk analysis that reliability of that component is
. 30
X. If the reliability isn't X, then it trickles through the
whole thing. It may in fact be almost as important to look
at the underlying assumptions in the risk analysis and make
sure the assumptions are valid as it is to look at the
specific most risk-important system it identifies.
There is a real balance there. So we have got to
be real careful that we approach this with the same type of
balance of what is the information telling us, what are the
assumptions. I'll use the example of the sleeping operators
at Peach Bottom. If we weren't going into the control room
basically each day with a resident or a senior resident as
part of their tour observing staff turnover, we might not
have seen that.
You will find very few items of noncompliance or
statistics generated from control room observations, but
it's a very important element of what we think an inspector
does each da, particularly the resident.
That same principle needs to be carried over into
engineering. Right now we are basically looking at the most
risk-significant systems, and quite honestly, we try to pick
the two or three systems that are risk-significant to look
at and analyze. I recognize that we may have a flaw in our
system, because we are not necessarily looking at as much
detail in those systems that might get ignored.
COMMISSIONER DIAZ: What you are really saying is
. 31
that the entire process falls in the area of systematics.
In other words, it is this system of risk you are going to
be working with rather than the risk of any one component or
area.
MR. GILLESPIE: Yes. It's a compounding just
because we have a complex machine and everything is
interrelated. So it's not a simple solution. You might say
that the entire QA process in a risk analysis isn't modeled.
It is all included in this little beta factor that the risk
people like to put in front because we don't understand it
real well. Yet the failure of a complete QA program would
be very risk-significant in our opinion.
The limitations of the mathematics give
limitations to the applicability, and I think we have got to
keep that insight in the back of our mind as we approach
inspection.
CHAIRMAN JACKSON: It strikes me that all of this
is tied together. Commissioner McGaffigan spoke of where
you could have cumulative effects. You spoke of looking at
what you think may be the most risk-significant things but
if you ignore some others, you may be ignoring the effect or
the implication of that.
MR. GILLESPIE: Exactly.
CHAIRMAN JACKSON: Commissioner Diaz spoke of
having systematics. So the whole thing requires an
. 32
integrated and systematic approach that has risk insights
built into it, because that helps with the systematics. I
think that is all they are really saying. Sometimes we fall
into traps where we think we are talking about something
here or something here, and it just doesn't work that way,
because in the end the plant is an integrated entity itself.
MR. GILLESPIE: Jumping to the short-term actions,
quit honestly the short-term actions we have significant
leveraging of the staff. We have three AE teams with five
contractors on a team; two team leaders assigned from the
NRC so that we can keep the paperwork going when the team
goes out again and we can maximize our use of the
contractors. I think we are leveraged to the maximum extent
possible.
We also have a significant program where we supply
engineering specialists to the regions when they have a
particular problem to do a safety system functional
inspection, which is our traditional reactive procedure for
doing a vertical slice or really digging into a specific
system, which we are also supplying. This did not push off
the more routine efforts. This was clearly an addition to
what we were already doing.
Regional inspections of engineering. We are
trying to address now getting this incorporated rather than
in a reactive way, which it was before. I am talking about
. 33
the engineering aspects of design versus when we were
looking at engineering more traditionally. We were looking
at engineering in support of operations: were they being
responsive when the operator said he needed assistance?
It was the responsiveness to operations, not
maintaining the integrity of the basic design itself and the
basis for that design. So it's a different aspect of
engineering that we are looking at now. We are not throwing
out looking at the other piece, but we are looking at this
right now in the short time. It's all an addition.
COMMISSIONER DIAZ: Where are the senior
specialists? We place two in each region and two in here?
MR. GILLESPIE: That's our senior risk analysts.
I do have a whole separate slide on that.
COMMISSIONER DIAZ: All right.
CHAIRMAN JACKSON: Let me raise one other thing,
which is basically a comment on your comment. I recall, if
I can paraphrase him, Commissioner Rogers always emphasizing
the point that it's not just looking at design per se, but
you want to look at how design and design changes get
propagated into operations and plant changes and into
procedures. I assume that when you speak about that that in
fact is what you are looking at.
MR. GILLESPIE: Exactly. That's why I said I
don't want to throw away what we have been looking at
. 34
before, but what we are now saying is we are not taking for
granted the engineering solution or responsiveness to the
operational problems. Now we are saying, well, when you
propose that solution, where did you get your numbers?
Where did you get the pressure requirements? Where did you
get these other things?
More goes into the robustness of the solution. So
we are going kind of backwards down the stream flow of
information.
Longer term actions --
CHAIRMAN JACKSON: Before you go, I can't let you
go by the 50.54(f) letter follow-up. You have already
worked out who is going to do that and that the resources
are available? Is it going to be in the headquarters where
the reviews are?
MR. MIRAGLIA: We are having a conference call.
Hugh, did you get it set for tomorrow?
MR. THOMPSON: It is tomorrow afternoon. We will
reach a final position with respect to how we will integrate
the review process between the NRR and the regions in order
to assure that the right people look at this in a timely
fashion.
CHAIRMAN JACKSON: Have you worked out what any
follow-on regulatory actions would be?
MR. MIRAGLIA: We are looking at those elements as
. 35
well. We just started to get some of the input from the
licensees. I think the due date was the 9th of February,
and the letters are coming in. But yes, one would have to
say, given those letters, what appropriate action should
there be for follow-up inspection questions or enforcement,
and that all has to be part of our evaluation.
CHAIRMAN JACKSON: And you have developed criteria
for review, that is, what it is you are looking for?
MR. MIRAGLIA: Yes.
COMMISSIONER ROGERS: Before we leave this, if we
could back up one half bullet or so to the new inspection
procedure, what are the dollars and FTE resources that are
going to be required in addition to what we have devoted to
our inspection activities to carry out this new inspection
procedure?
MR. GILLESPIE: The reason this is short term is
we are actually trying to achieve that within some limits
that we placed upon ourselves. Our first question is those
limits, and the limits that we have placed upon ourselves
really are this would be done with existing staff.
So I would have a diversion, and at a budget level
dollar support-wise for contractor support at about the same
that we have right now. Right now, for this element of it
we are spending at the rate of about $2 million a year for
individual contractor support.
. 36
If you would like, I could outline what we are
looking at.
COMMISSIONER ROGERS: I don't need it right now,
but if you could supply us with that, it would be
interesting to look at it.
MR. GILLESPIE: Yes.
CHAIRMAN JACKSON: Are you comfortable with the
trade-offs?
MR. GILLESPIE: For the short term I'm comfortable
with the trade-offs. I'm speaking from a prejudiced
position, because I'm seeing immediate gratification from
the results of the architect/engineer inspections, what you
might call the safety severity or safety significance of the
findings.
I feel comfortable right now with the trade-offs
in that I am not giving everything away and I am not
diverting from some of the things that we are doing that are
well. Current engineering, which we had just beefed up last
year relative to the resources and the core inspection
program in engineering, we had raised the number of hours
and adjusted it. We are trying to work within that volume.
It would be the kind of approach that says if you
haven't had anyone look at a specific system in detail, in
the vertical slice kind of sense, in the last two years you
haven't had one of the architect/engineer inspections, then
. 37
you should consider them for this procedure. If you have,
then you should carry out the old core procedure.
What we would do is end up like on an every second
SALP cycle first looking at engineering one way and then
looking at engineering from a more design detail way.
We are exploring that. That is conceptually how
we are going. Will everyone be satisfied with the pace that
that is? I'm not sure. I can only say this is the kind of
thing we are exploring, because you have to have a baseline
for people to comment on.
MR. MIRAGLIA: This is the first attempt. In fact
it's out for regional comment. The goal would be to utilize
the resources from the region, augment from headquarters if
necessary, or use contractors. When you use the regional
resources, perhaps they are not doing something else that
they would have been doing. So those are the trade-offs.
And to really assess what that impact is. Or if we use
headquarters people. These are the trade-offs that are yet
to be determined. We are going to have to try the
procedure, apply the procedure.
CHAIRMAN JACKSON: Within the overall head count
numbers, you had people who were doing design reviews
relative to design certification. Is there any fungibility
of people?
MR. MIRAGLIA: Yes. What has also happened in
. 38
that area is we have been ramping down in terms of those
resources over the last few years as design certifications
for the evolutionary designs have come to conclusion. There
is only one design under active review. So those resources
have been coming down and we have been applying those
resources that were there, as well as our staffing levels
have come down.
CHAIRMAN JACKSON: You have been shifting them?
MR. MIRAGLIA: Yes, and that's how we have been
able to keep up with lots of the initiatives that we have
had. The increment that is left from that to roll over to
that is much smaller than it was two years ago.
CHAIRMAN JACKSON: I understand.
Okay, Frank.
MR. GILLESPIE: Long term actions. We are going
to have a presentation next week on Millstone lessons
learned. The major policy question that comes up here is
the licensing basis, what it's called, is it tabulated.
In the simplest sense it might be. If an
inspector was going to go out and do an inspection on the
auxiliary feedwater system at a plant, does he have a list
of all of the commitments and requirements for that system
easily available to him? Right now the list is not easily
available to either us or the licensee. That is going to be
a major focus of our discussion next week. So I am going to
. 39
try to not do it today.
The other one is long term, the balance between
operations and design inspections as it applies to our own
staff. Quite honestly, I think it's very well known.
I just mentioned we are putting about $2 million a
year into contractor support for engineering type support
for the regions on a reactive basis. The A and E
inspections are budgeted at about $4.5 million a year to
keep three teams going. That's $6.5 million.
That's a big resource, and at some point we have
to step back and say can we continue to pay for that for
contractors or do we have to bring our own staff talent to
bear on it and recruit people with maybe different skills
than we had before, and how do we do that and how do we
factor them in.
That's why it's long term; it's not something we
would do overnight, and requires some planning out, and a
different resource base, a different recruiting base, quite
honestly. Our focus for recruiting has very much been the
resident type person operations, former operators.
[Slide.]
MR. GILLESPIE: The next topic is inspector
qualification and training. This came up. The basic
question the IG was asked: Are we following our own rules
on how we are documenting our training?
. 40
There were weaknesses found in our training
documentation.
Management expectations for performance-based
inspections were not clearly communicated. Let me go
through the list, if I could, and I would like to address
the difference between performance-based inspection and
inspecting performance-based rules. I find that our own
inspectors get confused, and I would like to take a shot at
at least trying to distinguish between those two.
A combination of NRC organizational changes. We
eliminated 40 section chiefs in the regions. What those
section chiefs provided was an integration function, in a
sense, of some of the results coming out of their individual
reactors. That integration function in some cases got
pushed down on the resident inspectors. You will hear
resident inspectors say, I'm doing administrative tasks.
What we have got is inspectors, seniors
particularly, who are fulfilling two roles. There is
information collection and then there is diagnostics. The
traditional inspector really enjoys information collection
and he may not enjoy diagnostics, and that is preparing for
PPR, preparing SALP packages, preparing for senior
management meetings. So we are really going into that right
now and trying to understand it, which gets us to one of the
other actions down here.
. 41
Short term actions. We have republished our
manual chapter. You will hear in the staff the acronym
1245, like everyone in the world is supposed to understand
it. That is our Inspection Manual Chapter which covers
training. It reflects our current policy on training,
meaning it added some new training courses in.
In addition, a significant change which is now
going around the halls, because I held a training session
for 250 people in NRR yesterday on this, is that it makes
inspector classifications generic. What we found in the
regions was that the regions very much comply generally with
our program; they have inspector qual journals that are
signed off by different people; it is very structured, very
disciplined.
In NRR we turned around and we have evolved to a
program office with many, many people supporting the regions
in doing inspections. In fact, we at least equate to one if
not two division equivalents for the engineering divisions
in the regions in supporting inspections in pieces here and
there and everywhere when you add them up: fire protection,
maintenance inspections, PMs filling in for seniors that go
on leave or take vacations. My people do special team
inspections, the architect/engineer inspections.
What we have done is taken the word "region" out
of inspector and now we are looking at making sure that
. 42
people who carry out functions have the same qualifications
independent of their geographic location.
We are creeping into this right now. We are
working with our partnership and we are working with NTU.
You will see our new postings going up. Yesterday two
postings went up in NRR which now have a new page on a
position description that says you are expected to be
qualified to one of these categories in 1245, and you may be
called upon to have as much as 10 percent of your duties as
supporting inspection activities. So we are trying to be
open and honest with people, and we are getting into that
now.
This is a very big change. It's a cultural change
within the organization in integrating the two functions of
licensing and inspection together.
A new inspector training course was developed.
And now let me address performance-based inspection before I
try to address inspection of performance-based rules.
Performance-based inspection and the way we were
viewing it -- and we started this back in 1988 when we
published our first NUREG on this -- is basically an
inspector looking at the system, understanding first what
the system's function is supposed to be and first looking at
the system as actually a hardware system installed and
functioning and saying, is this system carrying out or
. 43
capable of carrying out its basic function, and what does it
have to do that?
That would lead you hopefully into things like if
you have a pump and the tank and the suction of the pump is
on the same level. Do you have a problem with that positive
suction head, yes or no? Then it leads you into possibly
looking at the paper and tests.
In fact, I'll use a success. The senior resident
at Fermi, who has recently moved to Zion, asking this type
of question, found a tank that should have been three levels
higher for net positive suction head. He approached it this
way, and in fact they had to nitrogen load the tank to make
sure that this surge tank would work.
That is a techniques course, because we hire
people in with the technical expertise of being a good
mechanical or electrical engineer, but we have not
necessarily hired someone with the conceptual idea of how to
approach a system.
So while we had a performance-based inspection
course already there, it wasn't good enough, and this is our
next step, to try to provide input insights and conceptual
insights on how to approach inspecting a system and trying
to get at the root cause of the problem.
It's a first step. As you can see, it's scheduled
for March 10, 1997. But it builds on the way the
. 44
contractors we hire actually approach a system: Will it
remove the heat?
Well, the first question on a heater exchanger is,
look at the nameplate data. I think this was something that
came out of Millstone. They looked at the nameplate data
and found out the Btu rating was significantly lower than
the expectation. That is a performance-based inspection
independent of what the rule says.
Inspecting a performance-based rule is more like
what we are doing now on the maintenance inspections. The
first go-around on the maintenance inspections is more
programmatic, quite honestly, to make sure the system is in
place that can deal with this: How are they using PRA? How
are they classifying system?
After the first go-around, which is really looking
at the paper, then we will hopefully use our
performance-based approach to inspection techniques on the
systems when we are inspecting the rule.
I am trying to make sure that it is clear.
Performance-based inspection is not superimposing anything
different on a licensee; it's how the inspector actually
approaches his day-to-day work.
We are going to continue to emphasize this. We
are putting a new training course in place, and it's really
a thought process rather than a cookbook. You can apply
. 45
this conceptually to an electrical system, to a mechanical
system, to an instrumentation and control system. Now we
are trying to work to that point. It is not something I can
write an inspection procedure for and use a checklist. It
really is a mental process to try to get into the program.
COMMISSIONER ROGERS: I like what I am hearing you
say here. It sounds like a very sensible way to proceed.
How much of this is all written down someplace that somebody
can take a look at? What you are saying seems to make a lot
of sense, but I haven't seen it written down anywhere in
anything that has come before my eyes.
CHAIRMAN JACKSON: Do you have it in terms of any
of your inspection modules? Is it part of some training
course?
MR. GILLESPIE: As it happens, I had that same
feeling, because I've been working in this. My mind has
been working this way since I was an inspector. It is not
written down in a real visible way, but I went back and got
our performance-based inspections report, which was a NUREG
that was completed in March of 1988. The words are still
valid today. Have we pulled it together the way I
articulated here in a policy document? No.
COMMISSIONER ROGERS: I think that would be very
useful to do.
MR. GILLESPIE: We have put the training courses
. 46
in place. What I have just articulated is the inspection
begins -- I am reading from the abstract -- "the inspection
begins with a performance-based observation and then the
inspectors let discrepancies or uncertainties lead to the
inspection of other areas such as quality verification,
organizational . . ." And you start looking at the system,
and then you go on.
CHAIRMAN JACKSON: I would put it in the realm of
more explicit guidance for your own people and ensuring that
there is consistency between what is in the training courses
and what is in that guidance.
MR. MIRAGLIA: As Frank said, back in 1988-89 we
started to talk with the inspectors, that you need to
inspect performance as opposed to programs. We were
articulating that in the broad context in the training
program and in broad context within fundamental of
inspection. Perhaps not everyone had the same mental
picture of what we were talking about.
CHAIRMAN JACKSON: The way Frank just explained it
is very well articulated, and I think that articulation
perhaps needs to be written down somewhere.
COMMISSIONER ROGERS: I wouldn't want to just see
this buried down in inspection manuals. I think this is a
very important point of view that ought to be very clear as
to how we view our activities here. Your connection between
. 47
inspecting for performance versus inspecting against a
performance-based rule and drawing that distinction in the
context, I think that is very important, because we have to
be clear on what it is we are talking about. I thought what
you said was very good. It sounded excellent to me.
CHAIRMAN JACKSON: We may want to see it surface
as this is the NRC approach.
COMMISSIONER ROGERS: This is our concept.
CHAIRMAN JACKSON: Right. This is the concept. I
think he is saying to elevate it, and we can promulgate it.
MR. MIRAGLIA: We will certainly do that.
COMMISSIONER DIAZ: I do agree that it sounds very
good. I just want to make one observation of your points.
We are talking about training and you said that inspectors
are more prone to do information collection than to do
diagnostics. I take that as a value judgment on your part.
That might very well be true, but in training, to be able to
streamline the information process you have to introduce
diagnostics at that very first level.
The fact is it is diagnostics that allow him to go
from one component to a system. So it is critically
important in the training that, even if they like more to do
information collection, diagnostics is an indispensable
element, and maybe we should put an emphasis on that in our
training programs.
. 48
MR. GILLESPIE: And we do. I was drawing a more
distinct diagnostic line, and I will go through that when I
have a diagram here. You often hear residents when you
visit the sites saying, I spend too much time at my
computer. I'm drawing the line of diagnostics at preparing
for senior management meetings, preparing for plant. That
is performance-based inspection as you just described.
Yes, is the answer. I was drawing a different
line just from what we are finding in interviewing people
and the job task analysis is also finding that we are going
around and doing, which is coming close to an end.
CHAIRMAN JACKSON: Let me ask you this question.
It's kind of a judgment question. Given what you have
outlined in terms of this increased focus on definition of
the distinction between doing performance-based inspection
versus inspecting performance-based rules, have you been
able to do a survey or detect how much of a problem it is in
terms of a lack of understanding of that by our own
inspectors?
MR. MIRAGLIA: I think Frank referred to an IG
audit that looked at our training. The maintenance rule
audit brought the performance-based issue not being well
understood into focus.
CHAIRMAN JACKSON: How comprehensive was that
audit?
. 49
MR. MIRAGLIA: I'd have to go back and look at
that report, Madam Chairman, but they talked to lots of
inspectors and regional people and headquarters people.
CHAIRMAN JACKSON: Does that square now with your
own judgment?
MR. GILLESPIE: Yes, it does. The people were
focusing on the phrase, and the phrase "performance-based"
had "rule" after it when it was used once and "inspection"
after it when it was used again. Then you had the question
of, well, why are we looking at the inspector question? Why
are we looking at paper with the maintenance rule on the
first go-around?
Well, the first go-around you have to make sure
that the processes and systems are in place by which you can
judge performance. If an inspector sees a component
failing, he then says, okay, that component is failing. Are
you treating it correctly? That way the inspector is not
trying to second-guess the system and the paper. He doesn't
have to reinvent the look at the paper.
We had to tell the inspectors, yes, this is still
performance based, because you have to make sure the
performance process is in place so that when you do
performance-based inspection you don't have to go back and
do that all over again.
MR. MIRAGLIA: With some of these concepts I think
. 50
it would be preferable for us to be cartoon type characters,
so when you use a phrase you have the little balloon over
everyone's head and everyone can say, what is the mental
picture?
I think what we will find on some of these
concepts is we have talked past one another; we are not
fully communicating because the definitions are a little bit
different. I think that is an element that is here.
Even if we have this broad statement, Commissioner
Rogers, that is the start. It has to permeate the training
programs; it has to permeate the communications.
COMMISSIONER ROGERS: It shouldn't be so broad
that it leaves out what you are seeing.
MR. MIRAGLIA: I understand. That's the starting
point, and then it has to permeate.
CHAIRMAN JACKSON: You are talking about different
tiers of the same thing.
MR. MIRAGLIA: Yes.
CHAIRMAN JACKSON: You can have the overarching
thing that is some broad-based kind of point of view, but in
the end, if it isn't inculcated into everything that helps
our inspectors do what they need to do, then we haven't
accomplished anything.
MR. GILLESPIE: I felt very good when I was at
Fermi and met with the senior and he showed me this, and I
. 51
said, aha, I at least now know we have reached one.
Long term we have a job task analysis which will
be completed this March by the contractor. One of our
questions was this question of diagnostics in the sense that
I had used it earlier. We have made some changes to the
organization. We shifted a lot of things around.
What has been the impact? Are we really getting
what we think we are getting? If someone is doing
something, do they have the necessary training to do what
they are doing? The split of work: what is done at a site
and what is done at a region.
That is all we are looking at in this job task
analysis. It is not one of specific positions. It gets to
that, but it is one of the division of reactor projects in
the regional organizations, and we have looked at all four
regions, because all four regions carry out the same program
slightly differently and functions get allocated differently
even within a region between branches.
We are looking for the most efficient, effective
process through the whole thing to deal with it: How should
we split functions? Where should they be? Who should be in
charge of it? Signature authority.
CHAIRMAN JACKSON: What kind of time line have you
laid out for yourself with milestones or for getting the
whole thing done?
. 52
MR. GILLESPIE: The contractor effort will be
completed in March. I've got some informal insights back
from them. Just like any anxious person, we call them up
and say, what are you generally finding?
We are generally finding some of the things I have
actually already said. The level of discussion about time
at the computer, probably there is more discussion to it
than there is the actual time. We went out and put expert
panels together and said, how much time do you really spend
that goes with this anecdote?
What we were trying to do is take all this
anecdotal information that everyone had gotten and put it
into a total perspective so that we are not underinflating
or overinflating what a particular problem might be in the
whole system. So weighting is what we are hoping to get out
of this report to pull it all together.
All the regions participated; headquarters
participated; there were many, many meetings. So it's going
to be, I think, a very beneficial rebaselining.
MR. MIRAGLIA: The output from this is going to be
useful in many ways. As Frank said, it's a function, an
assessment of the function. We have many changes in the
program or contemplate in the program improvements. We'll
have to also say, given those changes, where are we today
and what does that mean to assessment of functions and
. 53
training?
CHAIRMAN JACKSON: I guess it still is important
that we get some sense of when you are hoping to really see
some of this reflected in the inspection program.
MR. GILLESPIE: It will be. I have no problem
coming back once I get the report in hand and we can deal
with it and get people's opinions again.
CHAIRMAN JACKSON: All right. Is a similar thing
being done for the project management organization?
MR. MIRAGLIA: That brings us to the next slide.
The Chairman is moving us along.
MR. GILLESPIE: Roy is going to cover this slide
and I get to have a rest.
[Slide.]
MR. ZIMMERMAN: There are areas for noted
improvement with regard to project manager training and
qualifications. The Inspector General issued a number of
reports over the past year indicating areas where we needed
to improve our activities as well as our own
self-assessments.
Although there were some findings where individual
staff performance could be improved, mostly the areas where
we needed improvement were for management to provide clear
expectations to the project managers on what we wanted them
to do and to provide the necessary training and tools to
. 54
accomplish that job.
To that end, in the short-term actions we issued
expectations on the interactions with the regional office.
It became very clear from lessons learned from
Millstone and from Maine that there needs to be a very close
linkage between the project manager, the resident inspector
and the region-based inspectors. They need to talk often
about activities that they are both working on.
The inspectors in the field benefit from the
insights that the project manager has from the licensing
matters that they have been involved with both of a
plant-specific nature and also the benefit of being here in
headquarters and having a generic knowledge.
CHAIRMAN JACKSON: How in fact are you
implementing this shared safety responsibility? How are you
ensuring or enforcing this integration?
MR. ZIMMERMAN: This is one of the first key
messages that we have articulated a number of times. I
personally have done it as well as my direct reports. We
have had discussions in workshops in the auditorium to go
over the issue that our job is early identification of
problems at facilities and declining performance, trying to
find those problems early. That is not just the region's
responsibility; that's all of our responsibility.
CHAIRMAN JACKSON: How do you ensure that this
. 55
kind of interaction and interaction with the desired outcome
occurs in fact?
MR. ZIMMERMAN: We have put a feedback mechanism
in place. We have indicated that these phone calls should
take place with the region no less often than three times a
week. It is documented in our PM handbook.
We have in our process improvement plan a feedback
mechanism for the project directors to be able to come back
and say that they have monitored these phone calls, they
have reason to believe and to support their reasons that in
fact we are having success in this area. It is not enough
to put out the expectation; there is a need for feedback to
test it, to see if it is actually occurring, and we have
built that into our process.
CHAIRMAN JACKSON: Have you also built that into
your accountability and your expectations in terms of
people's actual performance?
MR. ZIMMERMAN: To a degree. We are moving in
that direction. We are working on our elements and
standards. We see there are some areas that need to be
improved so that it is in line with the expectations that
I'm going over. We need to do the necessary activities in a
coordinated way with our partnership group. That is where
we are headed. We are headed to element and standard
changes to reflect the points that currently aren't in the
. 56
standards.
COMMISSIONER ROGERS: How many times per year
would a project manager actually visit the project?
MR. ZIMMERMAN: It varies, depending on the
particular site and geography to an extent. Typically it's
about four times a year. If it's a plan that is having more
difficulties with us and there are more management meetings,
SALP meetings, PPR meetings, I would expect the project
manager and his supervisor to be making more visits to the
site and to the regional office.
COMMISSIONER ROGERS: Okay.
MR. MIRAGLIA: There is broad guidance out on that
in terms of the numbers that Roy indicated. There are
reasons why that may change because of specifics on the
project.
COMMISSIONER ROGERS: I understand.
MR. MIRAGLIA: There are those kinds of guidelines
out there.
MR. ZIMMERMAN: One of the lessons that we clearly
learned out of Millstone and the partial core offload is
that we have blind spots, and we have to do our best to
avoid those blind spots via the communications between the
project manager, the resident and the region-based asking
for input from the region that we are closing a licensing
matter; we are ready to approve this; and talking to the
. 57
region and to the inspector in the field about that.
They've had a chance to think about it. They can provide us
thoughts that maybe we haven't thought of. We can provide
insights to them on where they can inform inspections, that
we have been looking at some activities here that make us
think that this particular system might be one that warrants
sampling during your next inspection.
That is the coupling that has been there in the
past, but we are trying to make it front and center heavy
emphasis that that needs to be done. That was a clear
lesson over the past year.
Another area has to do with the fact that we have
also issued expectations on maintaining the FSAR current.
We have put out an internal requirement that the project
managers are to update their version of the FSAR with the
latest revision within 30 days after they receive that
revision, and that we expect that that FSAR will be used as
they carry out their licensing matters and other readily
available portions of the current licensing basis.
Again, not all portions of the CLB are readily
available, but there should be a reasoned attempt to add to
the FSAR as they go through looking at licensing actions and
licensing activities.
Many of the lessons that we have been learning
over the past year are process related. We have talked
. 58
about a few of those. We also had some challenges over the
past year in the area of handling of allegations, errors
that we have had in identification of allegers' identities
inadvertently.
Sensitivity to the way we treat and deal with
allegers is an area that we have also had significant
emphasis in our training to not repeat some of the problems
that we have had in the past.
Other issues that have come up -- and we track
them actively in our process improvement plan -- get into
areas of how do we handle E-mail that we get in. In this
day and age we get a lot more E-mail than we used to get
from various stakeholders. We have put out guidance about
if it's a professional record that needs to be maintained,
how it needs to be maintained; does it need to go into the
PDR? Answering questions about, is it like a phone call or
is it like a letter? We have come a long way in providing
guidance in areas like that.
Another lesson learned that we had was if a
licensing manager were to call up a project manager and
indicate that there is a concern that they are working on
and "we just want to let you know about it." What is the
expectation of the project manager?
We have gone over the fact that we would expect
that project manager would let his or her supervisor know,
. 59
get the word to the region so they're aware of it, and mark
it down, that that is an item that maybe is owned right now
by the licensee, but we need to follow up on it in a
reasonable amount of time. That's to stay on our radar
screen.
This was really separate from the Maine Yankee and
Millstone lessons learned, but we recognized the benefit of
rotating project managers on about a five-year period
similar to what we do with resident inspectors. A fresh set
of eyes is always good to have over a certain period of
time.
We have seen benefits clearly in the regions when
this has been done. We have been doing it in my
organization for probably the last six to eight months, and
I will expect the same dividends. Take your knowledge and
expertise to another facility. It helps consistency as
well.
CHAIRMAN JACKSON: When was this implemented?
MR. ZIMMERMAN: I believe it was about six to
eight months ago. In our process improvement plan we have
the closure date. I have that with me and I would be glad
to provide it to you.
CHAIRMAN JACKSON: Do you assign more experienced
PMs to more challenging facilities? Do you factor that in?
MR. ZIMMERMAN: Yes. We are trying to put our
. 60
stronger, deeper individuals on the plants that we find most
challenging. I think we have made headway in that regard,
but there is still more that we can do.
The process improvement plan, just to mention it
again. I believe a few months ago we sent a draft of that
document to the Commission. It's a very active document.
We are going to continue to add issues to that for us to
track, big and small. It seems to be working well.
We have a project manager, project director
advisory board. As we look at making changes we provide
those changes to this group, almost like an internal
partnering group, to give them an opportunity before we
update procedures: is there something from their vantage
point, from their perspective that maybe we haven't thought
of that we need to consider before we make the change?
Because changes can be a challenge for people to adapt to.
So we have tried to have a grass roots movement that I think
has been very successful.
In the long-term actions, we spoke a moment ago
about the job task analysis out in the field with the
residents. We are getting ready to undertake a similar
activity here with the project managers. We are working
currently on a statement of work that should be issued
shortly. It's approximately a six-month effort, and we will
likewise be monitoring it very closely.
. 61
We are aware that AEOD is working on a knowledge,
skills and abilities developmental activity, and we have had
dialogue with AEOD. We want to coordinate closely. We see
how this could be a good fit between what we are planning on
doing and the larger agency action, and we will incorporate
those.
CHAIRMAN JACKSON: It seems to me one has to be a
part of the other.
MR. ZIMMERMAN: Right. There is a natural feed
that occurs.
MR. MIRAGLIA: We had some activities underway
that were related. So there is a clear understanding to
make sure that they do fit and there is no unnecessary
duplication.
MR. ZIMMERMAN: We try to minimize duplication and
build off it.
I guess my closing message on this would be that
we are not waiting for the job task analysis, we are not
waiting for the knowledge, skills and ability review. We
will work in parallel as items come up. We are going to be
looking for ways to improve our performance.
CHAIRMAN JACKSON: The jobs task analysis for the
PMs has already begun?
MR. ZIMMERMAN: No, ma'am. The statement of work
should go out within the next week or two. Right now we are
. 62
hopeful that it will begin in March. It's the same
contractor.
MR. MIRAGLIA: It's the same contractor. So when
they are finished with the residents.
CHAIRMAN JACKSON: How long do you think again?
MR. ZIMMERMAN: Six months.
CHAIRMAN JACKSON: Thank you.
[Slide.]
MR. GILLESPIE: Going to the next slide, PRA
applications in the inspection program. Here I want to both
discuss where we are and the limitations of where we are
based on our own regulatory structure and how we are
approaching it and how it fits into the overall PRA
implementation plan.
We do recognize that there are areas for increased
improvement in using PRA in the inspection program. The
basic use over the most recent past has been in the
prioritization of what we look at, particularly in design
inspections, or when you select a sample.
If you select a sample within what safety system
am I going to look at for maintenance, what system am I
going to look at for this observation or test, then that is
where that one chart that seems to come out of all the IPEs
generally will get used in the resident offices and by our
team inspections. That could be said to be a simplistic
. 63
use, but that dominates how we do use it for right now.
The evolving use gets me into the short-term
actions.
CHAIRMAN JACKSON: Can you give us some examples
or at least an example of where you see an opportunity for
increased use of PRA in the inspection program?
MR. MIRAGLIA: I think we are going to get into
it. Frank is going to talk about the senior reactor analyst
and trying to get someone with a little deeper knowledge in
risk assessment, having two individuals within each region.
Those individuals now are assisting and participating in the
maintenance team.
CHAIRMAN JACKSON: That tells me people. You gave
an example of how we are using it today in terms of
selection of safety systems to review. What else would you
be using such people for?
MR. GILLESPIE: I think what we are going to need
to do in the future is be significantly more articulate and,
to use a term that we have used here before, transparent on
how we use risk relative to the balance between support
systems and primary systems; how do we allocate things.
The primary source of our information for
enforcement is the inspector. As risk gets more inculcated
into our enforcement policy as a measure of severity or
weighting factor, it's actually the inspector who is the
. 64
primary person writing down those initial words on what his
observation was. So it is going to be important in the
future as we get risk into things like the enforcement
policy past what we look at, get involved in how much we
look at it.
But the inspector is the source. He is the guy
that pulls it together. It all comes together at the
bottom, quite honestly, rather than coming together at the
top.
MR. ZIMMERMAN: That's happening. We had a
meeting just yesterday on a particular plant that we are
considering escalated action on. We had the region on the
phone, Office of Enforcement involved, and we were talking
about what we do know from the IPE in this particular case,
what insights does that give to us. So it is something that
is being actively used now.
MR. GILLESPIE: A specific example is, as we
change our own rules, it's going to be the maintenance rule.
In fact, the SRAs were actively participating in what I will
call the risk element of the maintenance rule, which has
become kind of a critical element: which systems are
classified in what class, how are they characterized,
because that leads to what actions will be taken later if
the reliability values are found to be deficient.
The other thing is now the inspector, once that
. 65
rule is out there in place, process approved, is going to
have make judgments as to, is this component failing
consistent or inconsistent with the guidelines in place?
That's one rule, and in fact the inspector is now going to
be inspecting against that.
Our approach here. What you see in the slide is
two years ago when we proposed the senior reactor analyst
program we had also proposed in the risk implementation plan
the three level diagram on training. The SRA is the middle
level diagram. It's a practitioner who is expected to apply
something but not expected to necessarily have the skills to
go out and completely do a PRA himself, to get that analytic
view in the regions.
He coincidentally hit, if you will, or got in
place at the same time the maintenance rule was happening.
It came together very nicely and they participated on them.
In addition, to make sure that they did have the
analytic background, 29 of the 39 presentations on the IPE
results were given by the SRA when they were doing their
rotation through Research, when they were giving it to both
the PMs and the regional staff, to make sure they were
familiar with the facilities that were going to be in their
regions. So we really through the training program have
tried to elevate them to that second level.
I will go to the last bullet, which is the PRA
. 66
development course for inspectors on long term. Basically
what we have got is a course we are putting together with
TTC which is going to treat PRA as a basic technology that
should be known by each inspector and reviewer.
Much the same way we have a GE or Westinghouse or
B&W basic technology course we expect inspectors to have
gone through, we are looking at a two- to three-week course
which takes the individual courses and lessons learned from
the SRAs going through it and comments and puts them
together in a two- to three-week comprehensive course that
is intended to be the bottom level of what was on the risk
implementation plan.
That's a person who knows enough about it and how
it's applied so as he sees it applied in a 50.59 review he
might be doing, as he sees it used in a continuing
categorization of a failed component, or he sees it as part
of the justification for a change to the plant, he has the
wherewithal to either make a judgment that he can say, yes,
that looks okay, or "I need to call the SRA who is my backup
back at the region who is at the next level in the thought
process and the reference period."
So what we are trying to do is have the people in
place with the talents to keep pace with the change to what
they are inspecting to, because, quite honestly, we still
have to inspect to the rules that are in place today and the
. 67
license that is in place today. But as that becomes more
risk oriented or risk has an application of compliance or
inspectability, we have to really have the people there with
the tools.
That is where our focus is now, getting them
prepared. We were right there at the right time with the
SRAs, which was a very nice coincidence to have occurred. I
would like to say we planned it that way two years ago, but
it worked out very, very well.
CHAIRMAN JACKSON: There are some pilots going on
on risk informed tech specs, ISI, IST. Do we expect those
to have --
MR. MIRAGLIA: As those programs are developed,
then we will have to say if the rules are changing in that
area, then that has impact on how we have to train and how
we have to prepare the people.
MR. GILLESPIE: Absolutely. The engineering
inspector in the region who traditionally inspects ISI now
has to have this other tool in his background. Once you say
the program is okay, the implementation of the program is
left to that engineering inspector in the region. He is now
inspecting a different technology, a different set of
records, a different approach, and making judgments as to
reduced frequencies because of lack of flaws found or
increased frequencies: are they proportional?
. 68
It's as much reliability engineering as PRA in the
aspects that the inspector is looking at. So we think we
can give the inspector that level of tools, although we are
not making him a full-fledged analyst.
That is where we are going with that, and we are
trying to keep pace.
The next slides really come together totally. We
call it performance assessment process.
[Slide.]
MR. GILLESPIE: The short-term actions as we are
looking at improving our performance assessment process.
This has been a particular interest to the Commission for a
time now. It has also been an interest to us. We would
like to improve ourselves.
We have issued some management directives which
very much try to document what we are doing today. There is
a diagram on one of these management directives which has
lines going all over the place. My simplified view of that
is actually the last picture on this whole package, and it's
listed as backup slide.
[Slide.]
MR. GILLESPIE: It becomes very important only
because what we have done is we have set up our system with
the plant issues matrix to mesh the plant issues matrix
report. Now what we are doing is identifying where do the
. 69
facts of the situation of our process come from.
This diagram tries to show that the inspection
report is where everything really starts to come together;
every LER is followed up on; AITs have results, and those
are followed up on. Inspection reports become not the only
source but a significant source of assessment information.
After that -- and this gets to the root of what I
was calling diagnostics before -- you are basically dealing
with the same database but you are refining it into
different levels. The PPR uses the same information; the
senior management meeting uses the same information. Every
three PPRs you do a SALP. It uses basically the same
roll-up of information.
This will introduce me to my second to last slide,
which is, how could we use technology now to maybe help us
in this as the agency comes out of an era of the early 1980s
of technology or information management, which was really
document management, into the technology of the late 1990s
where we are going to try to use information management to
help us do this and to get more diagnostic, more
transparent, if you will, into the whole process?
So the short-term action really here in these
management directives was to document what we are doing
today in a way that people could understand it.
We also then have the Arthur Andersen report that
. 70
needs to be considered and how does that factor in. I think
what you can see is that leads to this idea that we have to
be able to digest more information more rapidly to do it
efficiently and not get inundated with the weight of our own
paper in our inspection reports. We need a rapid indexing
to the information.
The next diagram, which is a pictorial, while it
says future assessment model, this actually lays out the
information sources that we currently have in the current
system.
The problem with our information in the current
system is that they are basically disconnected. If I want
to know what were all the items in our compliance found at a
Westinghouse four loop plant, I've got one computer that
tells me what are all the Westinghouse four loop plants, but
I've got another computer that tells me what all the items
on our compliance are, and while the docket number should be
able to relate this information together, the two computers
don't talk to each other. So you get a manual printout from
here and a manual printout from here, and immediately you
are in a system that is so awkward we don't do it.
CHAIRMAN JACKSON: Is IRM in the room?
MR. MIRAGLIA: No.
MR. GILLESPIE: What this is trying to do is be a
pictorial. What we are looking at in the inspection report
. 71
half on the plant issues matrix that was developed and has
evolved is as an index to the details in the reports. This
will both be technology and program.
We are going to be coordinating changes to our
manual chapters and process to describe how we do things
with the introduction of the technology because it is going
to change what an inspection report looks like from today.
For us this is a relatively high risk operation
because we have never done this before. It is pulling an
awful lot of information together from a lot of dislocated
places right now to try to put it into the same useful form.
Which gets me to the next slide, because the key is being
able to access and sort information rapidly.
[Slide.]
MR. GILLESPIE: Reporting and information
technology is, I think, going to be key to our success here.
If I could describe this in a way that Commissioner Diaz
explained it to me, he saw the diagram, and he said, I'm
glad you got rid of the triangle on the backup diagram.
Once you identify where is your basic factual base
coming from, then you can get a sense that everything after
that is a different sort on the same information. Now the
question is, how do we want to sort the information?
If I can relate the facility's characteristics to
the docket number, which gets me a relationship then to the
. 72
list, then I can do peer plants much the way AEOD does their
performance indicators right now, and I can look for
correlations; I can do a lot of diagnostics that I can't do
today. Not because we don't have the information, but
because we have so much information.
The ability to diagnose, sort and filter the
information is extremely tedious and manpower intensive.
That is one of the complaints from the residents. They are
doing what they are doing today manually. They are
rehashing it for the PPR; they are rehashing the same stuff
for the SALP. If we can come up with a way of doing the
cutting and pasting in the electronic form rather than the
manual form -- in fact, Region II even graphs these on graph
paper with pencil and ruler manually.
CHAIRMAN JACKSON: You've got to be kidding.
MR. GILLESPIE: I have to automate it. We have to
come out of docket management into information management,
and it's a 20-year leap in technology for us.
CHAIRMAN JACKSON: And you are going to do it in
two years?
MR. GILLESPIE: We are hoping to do it within two
years. That is what our commitment is. We are hoping to
prototype this in a full region within 12 months. I'm going
to give it one heck of a try.
CHAIRMAN JACKSON: Twelve months. I'm writing
. 73
this down.
MR. GILLESPIE: We have two what I will call small
prototypes that we have actually tried in our own LAN
environment already with regional participants. With time I
am gaining more confidence that it's fairly doable.
We found some glitches in the whole system, things
you would think smart people -- we're all engineers --
should know. In the last two years we have created a number
of databases and we were smart enough to use the same kind
of off-the-shelf commercial software, but we weren't smart
enough to label everything the same. So docket numbers in
one system are zero five zero dash something; in another
system it's five zero dash something; and they don't even
talk to each other.
CHAIRMAN JACKSON: This is where the CIO
organization should be helping you out, because your jobs
are one thing. The best way to handle information is
another thing.
MR. GILLESPIE: If I could jump into the
conclusion on 14.
[Slide.]
MR. MIRAGLIA: I think, Frank, we ought to say we
have been coordinating with IRM.
MR. GILLESPIE: We have joint project managers
with IRM and my organization and people working full time in
. 74
both organizations to try to make this work. Actually we
have no conflicts. The conflicts are technological ones.
CHAIRMAN JACKSON: All right.
MR. GILLESPIE: The conclusions. We have an
evolving program, particularly as it applies to inspection.
One of the conclusions and one of the strengths, I would
say, of the inspection program which we have been talking
about is the discipline of having an inspection manual in
some detail and generally having a baseline to change. That
includes detail in training, detail in how we do things. If
we have an unusable procedure, at least we can point to it
and know what we have to change. That is a strength, the
discipline we have approached it with.
Both self-assessments and external reviews have
led to program improvements. I would not want to claim
either one is stronger than the other. They tend to fit
together and actually be self-supporting.
Future direction will be closely coordinated and
integrated. This is this tying of the diagnostic and the
inspection and a recognition of how it's the same group of
people doing it. So we have to give them the tools to do it
as best we can.
The balance between operations and design is going
to be with us long term as we make decisions: Is it staff?
Is it contractor? And out of the whole how much is it?
. 75
What is the safety significance? How does it queue up?
Performance assessment I've touched upon.
Our real challenge is how do we use technology to
try to get us this giant leap forward in the technology
area.
With that, I thank you.
CHAIRMAN JACKSON: Thank you.
Commissioner Rogers.
COMMISSIONER ROGERS: I think this is very
interesting. A lot of questions I had when I walked in here
got answered, but let me touch on a couple of points.
On your performance assessment backup slide, one
of the problems that I have with this is that it looks like
it's a totally event driven process.
[Slide.]
COMMISSIONER ROGERS: It starts with an occurrence
observation or allegation. It looks like there is something
external to what we do that starts a process. In other
words, it looks like it's event driven and not initiated in
a certain sense by us.
MR. GILLESPIE: That's my drawing problem. The
observation piece is our routine program. We have got 434
people out there who are for the most part working the
observation aspect of it.
What I was trying to do is get a recognition of
. 76
even when we have an event and have an AIT, even the AIT
report is basically an inspection report while we call it
something else because it doesn't generally come through to
compliance. But any unresolved issues or any compliance
items or problems that come out of that event-driven end up
right back in our system of observations, because it's the
inspector where the rubber meets road who has to follow up
on it.
This was only intended to show that really we do
have a way. If all the information comes together in enough
of a bottleneck, we may actually be able to deal with this
if we can deal with how we articulate our findings. We
added in the PIM about a year and a half ago, and if I wrote
an inspection report so that the list item was in the
executive summary and every list item has detail in the
background of the report and it's a relationship, and I
index that item to SALP functional area, cause code -- in
fact in the second part of the Arthur Andersen report there
is one page on it which had, here's what we did to come up
with what we have; we think you should consider these other
things. But we didn't do it.
If you look at those other things, they look very
much like the LER cause codes; it's procedural compliance
personnel over design.
So those are actually things we already feed our
. 77
computers but we can't get it out because it's not connected
to anything, and that is called the IV system, which is
something we had hoped to fix.
So observation should probably be in much bigger
letters than occurrence and allegation follow-up.
Observation is what we do day to day.
CHAIRMAN JACKSON: You should capitalize it.
MR. GILLESPIE: Yes.
COMMISSIONER ROGERS: We don't want to be
controlled simply by events.
MR. GILLESPIE: No. Absolutely not.
COMMISSIONER ROGERS: Otherwise we will be chasing
those all the time and not taking another kind of look.
CHAIRMAN JACKSON: I just think it's presentation.
COMMISSIONER ROGERS: The other point on
performance assessment is that I was just curious as to what
your view of performance assessment is compared to the one
that the materials people have, particularly in looking at
waste sites. They have a rather large performance
assessment, or had a large performance assessment effort
there. I wonder if the terminology here is really basically
the same or different, and whether you had looked at their
approach to performance assessment.
MR. GILLESPIE: I have only got a sketchy
background on high level waste. Performance assessment in a
. 78
waste sense is the assessment of the repository?
COMMISSIONER ROGERS: No. It's not just the high
level.
MR. THOMPSON: It's a modeling process, which ends
up with a release rate into the environment. The words are
the same, but they really are focusing on something
different.
COMMISSIONER ROGERS: It's the process that I am
talking about. I'm not saying they should be the same. I
am just asking whether you have looked at and understand any
comparison between the two.
MR. GILLESPIE: We are not trying right now to
create a mathematical model. So in that sense it's
different. In principle, my view would be that the PIM
should actually be part of a report, because the licensee
then gets to see it. If the licensee agrees with the fact
that the inspector saw, now maybe I have a fact, and now I
can index and classify the fact.
MR. MIRAGLIA: I think the key is we are assessing
a different type of performance. I am not that familiar
with the performance assessment models used.
COMMISSIONER ROGERS: I don't want to get into it
too deeply here, because it might take us a little bit far
afield. I just want to raise the point that we are using
terminology that has kind of a superficial look about it
. 79
that is trying to achieve something close to the same thing
in one area versus another. I am just asking you to make
sure that you understand anything that might be beneficial
that comes from that.
MR. THOMPSON: The more we go to performance-based
regulations in the reactor side the better overlay there
will be, that is, where you can actually do some information
input into a model and you do calculations that come out to
drive some aspect of your evaluation to something that
matches your regulatory criteria. In those cases they will
actually come closer as we get that way.
COMMISSIONER ROGERS: As you use this approach to
performance assessment that you sketched out, Mr. Gillespie,
it does seem to me that there may be a little problem here
in that we are moving back into the use of judgment.
I'm not averse to that, but I think we ought to
recognize that in looking at the total ability of a system
to perform. You cited some examples which I think were very
good, the tank problem. That came about because somebody
exercised some engineering judgment in looking at it. That
was a darn good thing.
So it's not a bad thing to use engineering
judgment in this performance assessment, but I think we
ought to recognize that that is probably going to have to be
a part of it. At any rate, I would ask you to think about
. 80
that and address it as you proceed.
The other point is that I am a little concerned
about the resident inspectors. I wonder if you are not
perhaps overloading them. They have got a heavy burden
here, and I think particular care should be taken to see
that as you add into their activities these new
responsibilities that they get the kind of personal support
that they need to be able to get through this initial
period.
MR. MIRAGLIA: I think the job task analysis will
certainly give us a basis to evaluate that in a very
conscientious kind of way.
MR. GILLESPIE: Let me give you an example of the
feedback from the job task analysis. There is real
aggravation with the residents about preparing PPR packages,
SALP packages, senior management meeting packages, and
cutting and pasting. In fact, what we are trying to do with
the technology is, if we could standardize that, I can
relieve them of that burden of right now literally manual
cutting and pasting and then Xeroxing it to make it look
like an original piece of paper.
One of the objectives is to actually reduce that
burden through some sense of standardization and who does
what function. In fact, to try to reduce the cutting and
pasting or administrative burden of the whole thing all
. 81
across the board.
CHAIRMAN JACKSON: Frankly, I think one of the
greatest achievements in addition to actually saving in
terms of people spending their time on administrative things
is that in fact it can help to ensure actual consistency of
approach as you move up the line from PIMs to PPRs to SALPs
to whatever.
MR. MIRAGLIA: I was going to say something very
similar. In terms of not only efficiency and technology, it
will have us look at the functions: What functions are we
performing? Are they clearly the expectations out there and
are they understood? I think that is another thing that is
going to be an outcome of that. It is going to be an
iterative type process. We are going to learn as we go
along.
CHAIRMAN JACKSON: Commissioner McGaffigan.
COMMISSIONER McGAFFIGAN: I don't want to get into
things that we are going to do next week, but this last
backup slide I think is one of the best things I've seen
since I've been here in trying to get a lot of information
down on one piece of paper.
The plant issues matrix, when you get through this
computerization that you are going to do, is it going to be
the central document? It seems to be on this slide. If you
have a good plant issues matrix, then everything else flows
. 82
from it. So that's the heart of it?
MR. GILLESPIE: Yes.
COMMISSIONER McGAFFIGAN: We talked in my office
the other day and you just mentioned some day the goal is to
be able to -- at the moment, as I understand it, licensees
every six months get an update to their inspection plant,
and it is sort of like reading tea leaves or the Kremlin
wall: Where is the NRC today? They are worried about X, Y
and Z; there is a change there, therefore I better do
something. But we don't convey any more information than
that. If we could convey the plant issues matrix, would we
be conveying a lot more information?
MR. GILLESPIE: Yes. This is something that has
kind of come up. It is a public document. If someone
submitted an FOIA request, we would release it; we wouldn't
fight it. So then the question came up among the staff,
well, what is the best place to promulgate it since it's a
critical element in going into the PPR relative to the
summary document?
What we have toyed with within my group is going
to be a recommendation to the regional administrators that
we just attach it to the letter that already goes out with
the inspection schedule.
MR. MIRAGLIA: That hasn't been run up the chain
yet. Let's talk very candidly and frankly. It's not a
. 83
secret. Everything in the PIM is going to be public.
That's what our goal is.
We just instituted this process last year. Each
region approached it differently, and there are differences
in consistency. I think our goal is we don't want to have
one PIM in Region I looking very different from Region II.
So we are evolving. We are going to get there. At a point
in time I think we can share this information. That is our
goal and objective.
CHAIRMAN JACKSON: We may do something to hurry
you along in the process as a motivator for driving toward
the consistency faster.
MR. MIRAGLIA: That's the goal. The PIM was a
tool that we tried to use and we are learning. There are
differences; they are getting closer together; and as we get
to that point, as Frank says, I think their recommendation
will probably be accepted. It's just to make sure that
everybody is doing it in a consistent kind of manner.
COMMISSIONER McGAFFIGAN: I would agree entirely
on the consistency. From my perspective, this all leads
into the senior management meeting and the watch list and
all that sort of stuff. It would be nice to be conveying to
all 108 operating plants where we think they stand every six
months.
I think somebody said the PPR drives resources
. 84
around here, and since it drives so many resources, the
outcome shouldn't perhaps just be the X number of plants
that are on the watch list; it should be information
conveyed to everyone.
MR. MIRAGLIA: It's a tool to the whole process,
and we have to make decisions on how to apply the resources
to all 110 plants. This is the mechanism for doing that.
Another issue with respect to the PIM besides
consistency is we need to have a common understanding among
all of the regions and ourselves as to what is the threshold
for putting something on there.
COMMISSIONER McGAFFIGAN: Right. When is it an
issue?
MR. MIRAGLIA: Yes, and that is another thing
where there are differences and we are trying to sort out.
I think we are striving to that same goal and objective, and
the question is how soon can we get there.
MR. GILLESPIE: Because we are in a public
environment, we have a programmatic requirement that each
PIM item actually have its reference. Nothing is really
allowed to be on the plant issues matrix that doesn't have a
reference. The reference will be either an inspection
report or an inspection report which then gets you to like a
licensee report, because what we are interested in is, is
the licensee system working? Is he finding and correcting
. 85
things also?
So it's important to note that we give credit on
that. The basic question is, if you find something wrong at
a facility, that thing not showing up again is some evidence
that the root cause has been fixed. If that thing shows up
again even though it was fixed once, then you have to
question whether it has been fixed. So there is a trending
nature to why you would want put even things that we give
them credit for in one report on it if they are safety
significant.
We are thinking out how to use the PIM. These are
the thoughts that we had in place when we put it in place.
It becomes an index then to what is in the reports, because
it does have that link to a public document. The details
are in there.
COMMISSIONER McGAFFIGAN: Thank you.
CHAIRMAN JACKSON: Your backup slide is a big hit,
because what I want to tell you is that Commissioner Diaz
asked me to, and I will just read it, commend the staff for
transforming the pyramid into a functional line process that
is conducive to implementation of pass/no pass filters and
actual weighted feedback to decrease or prioritize
information for decision-making. N. Diaz.
So I would like to thank you for a very
informative briefing. Obviously the Commission is very
. 86
interested in this and obviously is closely monitoring it.
I think you have begun to address it as a guide to a thought
process. Your backup slide suggests this.
As you continue to implement the improvements you
talked about and others, what we are particularly interested
in is how our NRC staff activities provide the inputs into
the assessments that we make and the regulatory actions that
follow from those assessments, and then how they lead
ultimately to our attaining and providing to the public and
the licensees a clear and coherent picture of their
performance, the performance at operating reactors in this
particular case.
We are looking forward to the additional
Commission meetings. Again, thank you very much.
I would like to remind the Commissioners that we
do have an affirmation session.
[Whereupon, at 3:55 p.m., the briefing was
adjourned.]
|
Privacy Policy |
Site Disclaimer |