120th Advisory Committee on Nuclear Waste (ACNW) Meeting, July 25, 2000
UNITED STATES OF AMERICA
NUCLEAR REGULATORY COMMISSION
***
ADVISORY COMMITTEE ON NUCLEAR WASTE
120TH ACNW MEETING
***
Nuclear Regulatory Commission
Room T2B3
Two White Flint North
11545 Rockville Pike
Rockville, Maryland
Tuesday, July 25, 2000
The Commission met in open session, pursuant to
notice, at 10:47 a.m., THE HONORABLE DR. B. JOHN GARRICK,
Chairman of the Committee, presiding.
MEMBERS PRESENT:
DR. JOHN B. GARRICK, Chairman
DR. GEORGE W. HORNBERGER, Vice Chairman
DR. RAYMOND G. WYMER
MR. MILTON N. LEVENSON
. ALSO PRESENT:
DR. JOHN T. LARKINS, Executive Director, ACRS/ACNW
MR. HOWARD J. LARSON, Acting Associate Director, ACRS/ACNW
MR. RICHARD K. MAJOR, ACNW Staff
MS. LYNN DEERING, ACNW Staff
MR. AMARJIT SINGH, ACNW Staff
DR. ANDREW C. CAMPBELL, ACNW Staff
MR. N. PRASSAD KADAMBI, Office of Nuclear Regulatory
Research, NRC
MR. JACK ROSENTHAL, Office of Nuclear Regulatory
Research, NRC
MR. MEDHAT EL-ZEFTAWY, ACRS Staff
MS. LISA GUE, Public Citizen. C O N T E N T S
ITEM PAGE
OPENING STATEMENT 4
ACNW PLANNING & PROCEDURES 4
REVISED HIGH-LEVEL GUIDEANCE FOR
PERFORMANCE-BASED ACTIVITIES 10 . P R O C E E D I N G S
[10:47 a.m.]
CHAIRMAN GARRICK: Good morning, everybody. Our
meeting will now come to order. This is the first day of
the 120th meeting of the Advisory Committee on Nuclear
Waste. My name is John Garrick, Chairman of the ACNW.
Other members of the committee include George Hornberger,
Ray Wymer and Milt Levenson.
During today's meeting the committee will discuss
committee activities and future agenda items. We will hear
remarks concerned the revised high-level guidance for
performance-based activities, and we will discussed planned
ACNW reports on a number of topics, including risk-informed
approaches to nuclear materials, regulatory application,
comments on the low-level waste NUREG on performance
assessment, highlights of the visit to the U.K. and France
and comments on EDO response to ACNW action plan.
Howard Larson is the designated federal official
for today's initial session. The meeting is being conducted
in accordance with the provisions of the Federal Advisory
Committee Act. We have received no written statements from
members of the public regarding today's session, and should
anyone wish to address the committee, please make your
wishes known to one of the committee's staff.
It is requested that each speaker use one of the
microphones, identify himself or herself and speak clearly
and with sufficient volume that he or she can be heard.
There are some current items of interest I would
like to mention. Number one, announcement of a new
associate director for technical support, Mr. James Edwards
Lyons of the Office of Nuclear Material -- Nuclear Reactor
Regulation has been selected as the associate director. He
has been a top candidate in the SES candidate program and
has served in various positions in the Office of Nuclear
Reactor Regulation. He has a background in Projects where
he served as acting project director.
Over the next few months, Mr. Lyons will have to
complete some training activities, and we anticipate him
joining us during the next ACRS and ACNW meetings. During
the three months, Howard Larson will be acting as the
special assistant to the associate director for technical
support.
Barbara Jordan of the support staff of the
ACRS/ACNW support staff has accepted a position as a travel
voucher examiner and will be departing our staff on August
4th, 2000, and we want to extend our best wishes and good
luck to Barbara in her new assignment.
On Monday, July 10th, the Southeast Compact
Commission for Low-Level Radioactive Waste Management filed
a motion and a bill of complaint in the U.S. Supreme Court
against the state of Carolina. According to the Compact's
press release, the action was taken to enforce $90 million
in sanctions against North Carolina for the state's failure
to comply with provisions of the Southeast Compact law and
to fulfill its obligation as a party state to the Compact.
North Carolina has 60 days from docketing of the
Compact's filings to file its response.
On July 12th, Secretary of Energy Richardson
suspended the release of potentially radioactive
contaminated scrap metals for recycling from DOE's nuclear
facilities. DOE is also undertaking a feasibility study on
the possibility of recycling steel from decommissioned
facilities into items such as waste containers.
An event of some note, a transuranic waste
shipment left Hanford on July 14th bound for WIPP, 1800
miles away. Hanford is the fourth DOE site to ship waste to
the Department's waste isolation pilot plant near Carlsbad,
New Mexico, and will send about 2,500 shipments or 80,000
drums of transuranic waste to WIPP during the next 30 years.
The seven 55 gallon drums of waste that left
Hanford are being transported in the NRC approved shipping
containers. Another 400 drums of Hanford waste is being
reevaluated to ensure that the waste meets the State of New
Mexico's requirements.
All right. The subject for the next little while
is the revised high-level guidance for performance-based
activities. As you all know, for the past several years,
three or four years, the Commission has been moving towards
more risk-informed and performance-based methods of
regulation. There has been considerable progress,
especially in the area of risk-informed initiatives, perhaps
with a little less progress of what we mean by
performance-based regulation. So, hopefully, we are going
to get a bit of an update on that.
The Commission has been very anxious that these
guidelines are developed with input from stakeholders and
the program offices. And I think that as we have reviewed
this material, it has become sort of obvious that there is a
number of problem areas. One of those areas is what
constitutes reasonable performance measures. The thought
here is that we ought to be picking performance measures
that embrace and encompass a lot of what might be more
detailed measures of performance and be thinking more in
terms of the ultimate issues that we are trying to resolve
here, which is some way to measure safety to workers and to
the public. So, performance measures is a key part of this.
There are some issues that the committee is a
little bit concerned about in addition to performance
measures such as the suggestion in the Federal Register that
this is only going to be implemented on new initiatives.
What that really means is not completely clear. The reason
being given that the NRC has limited resources and may not
be able to implement it except on new initiatives.
Given that we have been in the regulatory business
for several decades, what constitutes new initiatives
against that large base of experience is something we would
maybe want to hear more about.
I think there is no doubt that the idea of
risk-informed, performance-based regulation is introducing
some stress into the regulatory process because it is sort a
departure from a prescriptive approach to regulation, or, as
some people might refer to it, a speed limit kind of
approach to regulation whereby, if you meet the prescription
or the speed limit, you are judged to be okay, but if you
exceed it, you are judged not to be okay.
And we know, of course, that radiation safety,
nuclear safety is not that simple and that we ought to be
forced in each issue to be looking at the relevance of the
threat in terms of our safety and risk-informed,
performance-based strategy, if properly implemented, ought
to do a better job of that than an attempt to anticipate and
identify performance measures at a lower level and assign
values to them, thinking that we really understand the
safety margins when maybe we do not.
So, the whole new strategy is one of trying to
turn up the microscope a little bit on what really is
important to safety, what we really mean by risk-informed
and performance-based. On the surface, it would seem to be
a very simple concept that we establish ourselves a
performance requirement such as a radiation dose, and then
we ask ourselves what is the risk that we may not be able to
fulfill that requirement. That perhaps is, in its most
simple from, what one would expect to see or be and
constitute risk-informed, performance-based guidance.
So, we have had quite a number of activities take
place already. There was created a performance-based
regulation working group some several years ago, and I know
I am stealing a little bit of our colleague's presentation
here, but I want to set the stage for this. These
guidelines were developed in draft form and have been
published in the Federal Register. There has been
workshops, there has been written comments on the
guidelines. And the NRC has even responded to the comments
through the Federal Register published in May. And then
there has been an online workshop, and now there are the
ACRS/ACNW briefings.
So, there has been a lot of activity here. There
is a considerable number of issues involved, the transition
to a risk-informed, performance-based thought process is a
major redirection. I think a lot of the problems come about
the NRC works very hard to be responsive to everybody, and
it is a different way of managing safety. And if we are
really serious about doing it, obviously, we are going to
have to do certain things differently, and such concepts as
defense-in-depth, safety margin, subsystem requirements,
fitness for duty, all these concepts will have to be
reexamined in the context of their compatibility with this
approach, and whether or not force fitting those concepts
would compromise effective implementation of a
risk-informed, performance-based process is something I
think we have to be very alert to.
So, we are going to hear from Mr. Kadambi, Prassad
Kadambi from Research, and I hope he has some updates on
some of these questions.
You have the floor.
MR. KADAMBI: Thank you, Mr. Chairman.
I would like to ask if my boss, Jack Rosenthal,
has anything to say on behalf of our branch?
MR. ROSENTHAL: Thank you. Jack Rosenthal,
Regulatory Effectiveness Assessment and Human Factor Branch
in RES. It took me a week to get the full title.
I just want to just put this in a little timing
context, and that is that we have been working on this for a
year or so. As Dr. Garrick said, we have had a number of
public meetings. We owe the Commission a SECY paper in
mid-August. We have met with the ACRS, we are now meeting
with you. The paper which will ultimately be prepared
describes the effort, presents the guidelines, speaks at
length to greater -- to public comment. And when we
published the Federal Register Notices, we didn't have the
luxury of writing at length on public comment, but to be
responsive to the public, I am sure that we will achieve
that goal here, that we have the space, and we will present
some examples of a trial, a proof of principle application
of the guidelines. So that is where it fits into the
broader context.
And with that, why don't we let Prassad get
started.
MR. KADAMBI: Thank you, Jack.
Good morning, I appreciate the opportunity to
address the Advisory Committee on Nuclear Waste. I believe
this is the first time -- well, for me it is the first time
coming to this committee, and I believe it is the first time
you are hearing about the staff's performance-based
regulation effort. I could be wrong in that.
We did make a presentation to the ACRS on June
8th. We have spoken with the ACRS before on this matter.
This briefing is primarily for information. We are not
expecting a letter but we would very much like feedback from
the committee, and, of course, if you would like to write a
letter, you know, that is certainly your prerogative.
The title is "High-Level Guidelines for
Performance-Based Activities." Based on my experience with
the ACRS on this, I would like to just get a couple of
things out of the way right upfront. First is when I speak
of high-level, one of the things that has happened when I
have tried to describe this work for people from the
materials area is their mind immediately goes to high-level
waste. That is not the connotation at all. Here what we
are talking about is the level of conceptualization and
generalization, the generality of the guidelines. Because
they are high-level, we believe that they can be applied to
reactors, materials and waste arenas.
The second term is with performance-based
activities. There was some confusion as to what activities
meant. At this point what we mean is primarily NRC
activities, and it is associated, basically, with developing
or modifying regulatory requirements to see if we can make
them meet the Commission's guidance, the directives that the
Commission has given on what constitutes a performance-based
approach.
This is an outline of my presentation this
morning. I would like to present some historical background
which maybe the Chairman has covered a little bit of that.
I would like to get into a little more detail on what the
Commission has asked the Staff to do.
We have had, as the Chairman mentioned,
interaction with stakeholders and I certainly am gratified
by the attention that some of our stakeholders have given
this topic and I think it is very useful to engage in a
discussion with all concerned.
We will get into the risk information and how it
is tied into performance-based initiatives and we can
discuss the guidelines themselves and I would like to
provide the committee with the latest thinking on what the
Staff expects to do from here on out.
If we do talk about the guidelines themselves, I
have copies of the guidelines in detail, which we can pass
around. I would rather use those than the sort of cryptic
bullets that I have.
As the Chairman mentioned, we are not as far ahead
in the area of performance based regulation as in some other
initiatives that the Staff has undertaken, primarily the
risk inform regulation initiatives, but we are fulfilling
the Commission's directives in this subject area. We are
making steady progress while we do this.
The development of the high level guidelines and
their limited testing represents a significant milestone.
The degree of progress that we have made is I think
commensurate with the resources that we have allocated. It
is a relatively low level effort and we have made
incremental progress towards achieving the goals of
performance based regulation.
By that, one thing that should be made clear is
that it does not mean either less regulation or more
regulation. It is merely what I would suggest is "smart"
regulation and it is an effort that complements the
risk-informed regulatory approaches.
The guidelines have been tested in a limited
manner. I think they need to be tested over a wider range
of issues and to identify some challenges which may limit
their application. Eventually I see that this effort will
be integrated with the mainstream efforts ongoing in other
areas and right now primarily those are the risk-informed
efforts.
Now the Commission has been quite really emphatic
in its commitment to risk-inform performance based
approaches to regulation. They started essentially talking
to us through the direction-setting issue papers that
started in 1996 and they continued to this day with various
issuances and right now they are really an important part of
the Strategic Plan in which performance based approaches are
mentioned in each of the strategic arenas.
The first SRM in this was issued in January, 1997
and as I mentioned we have made steady progress in it, but
when it started it was really focused on -- this is
terminology I am using that was part of an early SRM --
issues not amenable to PRA, and you know, that is what we
tried to focus on early-on but I think we have gone past
that into what I will be talking about as how
risk-information is used in order to help pursue some of
these initiatives.
The most recent paper that the Staff issued was
99-176, and to put it rather bluntly, it was not received
favorably by the Commission and the way I saw it was that
the Commission wanted the Staff to make much more aggressive
progress in this and the plans that we offered did not meet
the Commission's expectations because it lacked specificity
and what we were trying to do was to gather the lessons from
many ongoing efforts within the Staff which are labelled as
performance based efforts and we wanted to learn what that
teaches us before we got very specific about it.
We did make a presentation to ACRS in June of 1999
and the ACRS wrote a letter which was included in the
Commission paper itself and basically the ACRS included this
topic along with some other risk-based performance
indicators efforts.
The SRM really is -- for SECY 99-176 -- the SRM is
quite clear in what the Commission expects of the Staff and
we have tried to fulfill those expectations, as I hope will
come through the presentation today.
The basic direction of the Commission in this SRM
was that we should develop high level guidelines to identify
and assess the viability of candidate performance based
activities. Again this was something that we had
anticipated in the last Commission paper, in SECY 99-176.
We had expected that this would be a downstream activity
which after we had learned the lessons from several ongoing
performance based activities we would be able to then
develop some guidelines, but the Commission essentially
advanced the schedule and said, you know, I think we know
enough to develop guidelines and please go at it.
But the SRM also contains some other elements.
The guidelines should be developed with input from
stakeholders and the program offices. The guidelines should
include a discussion of how risk information might assist in
the development of performance based initiatives. They
should be provided to the Commission for information. In
fact, we have already provided a copy of the first draft of
the guidelines to the Commission -- that was in January of
this year. I take some comfort that we did not hear
anything about it. I assume that if there were any
objections to it we would have heard.
Then the Commission also wants the Staff to
provide its plans and the progress in developing the
performance based initiatives.
We believe that the guidelines that we have
developed will provide the framework for the kind of focused
and integrated activity that the ACRS had suggested we
should do, consistent with the ACRS advice.
As part of working with the program offices, one
of the first things we did was create a working group, the
Performance Based Regulation Working Group. It includes
representatives from NRR, NMSS, the Office of Research and
more recently a regional representative. We have met
essentially as needed in order to share our views and try to
expedite reaching a Staff consensus on the issues at hand.
We have published, as the Chairman mentioned,
several Federal Register notices. The earliest ones, on
January 24th and February 17th, were directed to a workshop
and the workshop was held on March 1st. This was a
facilitated workshop with people representing various
stakeholder interests, a roundtable and free-wheeling
discussion that occurred essentially throughout the whole
day.
We had people from UCS, Public Citizen -- in fact,
the two different groups within Public Citizen, those who
follow the reactor effort and the waste effort -- we had
utilities, we had representatives from radiopharmaceutical
companies, we had NEI and some others also over there, and
it was an interesting discussion.
We had written comments also from a wide range of
external and internal stakeholders.
We published another Federal Register on May 9th.
We responded, provided an initial response to the comments,
and we held an online workshop on June 8th, last month, and
that really elicited a very limited response.
One thing that I want to say about the stakeholder
comments so far is we have noted that at least one
significant stakeholder has expressed a feeling that their
concerns have not been taken seriously. I regret if we
omitted something in the way we responded that would convey
this impression, but we certainly don't want to give the
impression that the views of stakeholders are not
considered.
Perhaps the approach we took was a little cryptic
in that sometimes we take the easy way of telling people
what we do and why we do it by saying that, well, the
Commission asked us to do it and therefore we did it, but we
do hope to offer a better explanation for what we did and
why we did it.
In general though, I would characterize the
stakeholder input as not being necessarily unfavorable to a
performance based approach, although the guidelines
themselves I should say -- you know, they were not
unfavorable to the guidelines themselves but one notes that
those who are favorable to a performance based approach were
more in favor of the guidelines, and those who were opposed
to the performance based approach had various concerns about
the guidelines, and this is my own characterization -- that
they fell into concerns expressed on implementation and
trust issues.
Implementation, what I mean is that there is a
concern that the Staff may not implement the guidelines
objectively. Because they are high level guidelines and the
express concepts, it does require more specificity before
you start applying it in particular issues.
The sort of concern that we heard was that if you
went to a performance based approach you would be waiting to
see broken waste containers, high level casks on the side of
the road before the NRC did anything about an issue. That
is certainly not where a performance based approach would
take us, I believe.
The other, the trust issues, what we heard was a
concern whether the Staff would apply these guidelines in an
even-handed way to increase regulatory requirements if that
is the appropriate thing to do, just as we might decrease
regulatory requirements to reduce unnecessary regulatory
burden.
These are the kinds of things that we would want
to demonstrate, that we will use the guidelines
appropriately, responsibly and equitably.
One of the things in the SRM that the Commission
wanted the staff to address was the use of risk information
relative to performance-based initiatives. Basically, we
have characterized the use of risk information in three
categories.
The first is that risk information can provide a
basis for a level of performance that we would demand of
equipment systems, components, structures level. And this
would arise from answering the question, what is important
to safety?
And when we know what is important to safety from
a risk assessment, then we would try to answer the question,
what is the required level of performance to provide the
kind of risk amelioration that you make seek, that you learn
through the risk analysis.
The third then, depending on how you aggregate the
performance aspects, would be, what are the appropriate
performance parameters and the associated performance
criteria?
Now, when we talk of performance in this manner, I
believe what we are talking about is operational performance
as well as performance under accident conditions. So, we
would be talking about operational risk, as well as accident
risk, as is appropriate. We can also take into account
things like normally operating systems versus standby
systems, versus passive features of the regulated
environment.
The appropriate performance parameters, and I
think this is something the Chairman recognized, is really a
challenge and these are the sorts of things that we want to
work through on specific cases.
The second category is where risk information can
be used for metrics, thresholds and/or regulatory response.
The sort of example that I would offer for this is the
reactor oversight program as it currently works. And if you
look at the metrics and thresholds, essentially, as we go
into the guidelines, you will see that these are reflective
of the aspects of the viability guidelines in a
performance-based approach.
Then there is the category of initiatives
classified as not amenable to PRA. Again, this is
terminology that, just for consistency, I am carrying
forward from an early SRM. But this includes things like
quality assurance or training which are not directly modeled
quite often in PRAs. So, these --
CHAIRMAN GARRICK: But, of course, that is
different than making the observation that it is not
amenable to PRA. I mean there is a big difference between
saying that it is not modeled and it is not amenable,
because I fail to see how any of this stuff is not amenable
to PRA, it is just a matter of scoping.
MR. KADAMBI: I agree, and that is the reason why
I am careful to put that in quotes.
CHAIRMAN GARRICK: Yes.
MR. KADAMBI: I guess what I am trying to say is
that when you are talking about risk information and using
it for performance-based initiatives right now, it is -- in
a sense this category is characterized by a lack of risk of
risk-specific information. So that is -- it is just the
categorization that we are proposing right now in response
to the Commission's question.
Okay. We get into the guidelines themselves. If
there is any discussion on the guidelines themselves, I
would like to use the specific language that we have
developed through the working group efforts because these
are very cryptic headline forms and sometimes may not convey
what we intend. But let me just quickly go through at least
the structure of these guidelines.
The first point I want to make is that these high
level guidelines are not meant to be used as a cookbook of
some sort where you can run through a formulaic process and
come up with any conclusions.
The second point is that there is a high degree of
context specificity that should be expected as you go
through these guidelines.
The first set of guidelines are called viability
guidelines and these are -- they are four in number, and
they are essentially exactly the same was what the
Commission provided in a white paper on risk-informed,
performance-based regulation. The white paper was part of
an SRM to Commission paper 98-144, so that is the way I
referred to the white paper, as the SRM to SECY-98-144.
And under performance-based approach, the
Commission suggested that there be four attributes. One is
that measurable or calculable parameters should be available
or can be developed.
The second is that one can have objective criteria
associated with the parameters.
The third is that once you have set the
performance parameters and the criteria, the licensee should
have the flexibility to design the programs and processes to
meet the criteria. That there should be some provision to
encourage improved outcomes as part of providing this
flexibility.
And the fourth attribute is that even if the
criterion is not met, that there be no immediate safety
concern. And by talking about an immediate safety concern,
the way we are applying it is to say that there is a
sufficient safety margin to alert the licensee and the
staff, if necessary, that a criterion has been exceeded or
missed, that there is time for corrective action to be
taken, and that the licensee is capable to detect and
correct the performance degradation that would be implied by
the criterion not being met.
So, these are what we call the viability criteria,
and I believe that, you know, when these are applied in a
structured and a disciplined way, you know, some of the
concerns that have been expressed about, you know, a
performance-based approach just waiting for things to go
wrong would be alleviated, these concerns, because
especially the fourth attribute really forces the analyst to
look very specifically at the margin of safety in a rather
focused way, depending on what the safety issue is that is
being worked on.
Now, one of the questions that was asked by,
actually, ACRS was, you know, are these all that the staff
came up with? I mean if the Commission offered these four
attributes and we thought that they were not sufficient, I
think we would have proposed, we could have proposed others.
But I feel, especially having worked through a couple of
simple exercises, that these really provide the kind of
leeway for applying high-level guidelines, you know, because
the selection of parameters is, you know, it can be
measurable or it can be calculated, which means that you can
use analyses. The objective criteria can be based on
performance history if one has enough performance data to
fall back on, you know, that would be certainly a basis.
And depending on things like uncertainty, you could use
deterministic analyses or risk insights.
So, anyway, in a nutshell, at least based on the
work done so far, we feel that these viability guidelines
are sufficient to let us know whether a candidate activity
is a viable candidate to make performance-based.
So, the first set of guidelines represented really
answering the question, can it be done? And the second set
of guidelines, the way we see it, is addressing the
question, is it worth doing?
And to begin to answer this question, you know,
what we -- the approach we used was to begin with what the
Commission's performance goals are.
The Commission has indicated that, you know, we
should strive for four performance goals: One is to
maintain safety; two is to increase public confidence; three
is to increase effectiveness, efficiency, and realism; and
the fourth is to reduce unnecessary burden.
So, these four performance goals, we believe can
be addressed in an integrated manner to try to see whether a
given performance-based initiative would be worth pursuing.
And, of course, if the answer is no, then it would
be discarded, and whatever conventional approaches that we
use now would probably be applied.
The next guideline really has to do with, again, a
rather common-sensical question as to what is the net
benefit from doing this? This is, in a sense, a
cost/benefit analysis, but rather than get trapped into to
trying to define costs and detailed aspects of how the costs
are calculated, what we have tried to do is have the
guideline as something of a qualitative assessment on the
merits of pursuing a change, and to examine the benefits
that would accrue to NRC or the licensees, and a simplified
assessment, you know, something like --
As an example, I would say, you know, is there a
lower work exposure to radiation that one can see, you know,
which would provide a net benefit, you know, without getting
into a lot of costs?
Something that simple may be enough to say that
it's worth pursuing.
[Pause.]
Then continuing on the second set of guidelines,
there is a guideline on the regulatory framework. And what
we mean by the regulatory framework is the combination of
the regulatory features that address a given issue, a safety
issue, and that includes the regulation that is the Code of
Federal Regulation, the rule, any regulatory guidance that
we may have developed to provide one way for the rule to be
complied with.
There are occasions when we develop NUREG
documents to provide details on, you know, things that are
in the Regulatory Guide or in the rule itself. There are
times when we develop Standard Review Plans in order to help
the Staff conduct its review against a rule.
There are technical specifications which at the
operational level provide the requirements that licensees
have to meet.
And there is inspection guidance. This is what
inspectors look for and really constitute part of the
regulatory framework.
So what this is saying is that when we apply this
guideline, we would be looking at the framework as a whole,
and not individual components in isolation, and that either
one or more of the components can be made more
performance-based. That would be the question to ask.
And, in fact, you know, it may be that one has to
focus attention on some particular level in this hierarchy
within the regulatory framework in order to gain the
benefits of a performance-based approach. We have found
that to be the case in a couple of exercises that we have
gone through.
The next couple of guidelines are based basically
on the public comments we received, that really should be
the responsibility of whoever proposes the changes. In some
cases it may be the NRC; in some cases it may be an industry
group or somebody else.
The last of these guidelines is that, you know,
inspection and enforcement considerations should be
addressed early in the process, rather than as has happened
in the past, often it's an afterthought that comes much
later in the process.
The last of the guidelines under Section 2 is that
new technology should be accommodated in a performance-based
approach. This is if one has a set of parameters
sufficiently well defined, and you have acceptance criteria
for them, then it should really be amenable to applying the
latest changes in technology in such a way that, you know,
the safety issues are properly addressed.
The last set of guidelines have to do with the
consistency with regulatory principles. This is just sort
of a catch-all to make sure that we have not gone off track
in some significant way while we go through this process.
And it includes things like the principles of good
regulation, the PRA policy statement, Reg Guide 1.174, which
has to do with licensing basis requirements in reactors,
primarily. But the principles are applicable much more
widely.
And the strategic plan, of course, philosophical
issues like defense-in-depth and treatment of uncertainties
would also be definitely part of it.
[Pause.]
Now, what do we plan to do from here on out? Just
briefly, I'd like to say that we plan to build on the
progress we have made. The progress includes a couple of
simplified exercises where we have looked at initiatives
that, one of which is underway as part of the risk-informed
initiative. Option 3, it's called in our jargon.
It's a regulation having to do with combustible
gas control that's being considered for change. We selected
one part of it, and just went through an exercise of
applying the viability guidelines.
And although we are trying to prepare the
documentation on it to support a Commission paper which is
due very soon -- it's due to the EDO on the 15th of August
and we intend to meet that date -- we are trying to put this
altogether.
But basically the point I'm making is that it
looks like the guidelines work in terms of taking you
through a thought process that leads to the right kind of
questions to ask, to modify the regulatory framework where
it makes the most sense.
So, there's another exercise which we are planning
to conduct very soon, in a matter of days, and this has to
do with Subpart H to 10 CFR Part 20, Respiratory Protection.
This rule was changed just last year, in October
of last year, and basically the change was made to remove
some of the very prescriptive aspects of the earlier rule,
and to provide flexibility and some reduced regulatory
burden, while making sure that there was no decrease in
worker protection.
So, these were the sorts of things that, you know,
a performance-based approach would aspire to, so we are
trying to see how the guidelines might be useful in
assessing the kind of improvement that was instituted there.
But as we try to apply these guidelines and make
more progress in it, we hope that we can institutionalize
the use of the guidelines, and have it become part of the
planning, budgeting, and performance measurement process,
and eventually it would become part of what the Staff does
during the normal course of operating plans and budget
resources, et cetera.
Right now, we are suggesting a one-year trial
period in our Commission paper, and we would apply it to
suitable rulemaking and regulatory changes. Eventually we
would have to have procedures for the Staff to use it within
each of the Offices, and in a way that would be helpful for
the particular issues that each Office deals with, or even
portions of the Office that they address.
And going further out, I would think that if the
guidelines are found to be useful, they could be, you know,
made available to Agreement States and to industry
organizations, including standards committees, because there
are a lot of people who would like the standards-developing
process, the standards-developing organizations, to become
more of what they call performance-based. So it would be
good to have a consistent set of guidelines to inform that
effort.
And we would report to the Commission on the
results of this trial period about a year from this paper.
In conclusion, I believe that the Staff has
responded to the elements of the SRM, as directed by the
Commission. We will reflect whatever Advisory Committee
inputs we receive in the paper that we will send up to the
Commission by August 21st.
By the way, copies of the basic information that
was provided to you was also provided to ACMUI, the Medical
Advisory Committee.
Generally, I believe the guidelines themselves
were favorably received. Maybe I don't characterize all of
the views, but, generally, I think, you know, people didn't
point to any major problems with specific guidelines.
And the Advisory Committees will be kept informed.
We will provide whatever we come up to all the Committees.
That concludes my presentation. Thank you, Mr. Chairman.
CHAIRMAN GARRICK: Thank you. I suspect that the
members have a few questions. Milt, we'll start with you.
Do you have any questions? You better pull your microphone
down.
MR. LEVENSON: Well, I have a question which
borders a little on the philosophic, but in most fields of
engineering, we do our best effort to calculate something,
and then premeditatedly add a safety margin, whether it's a
factor of two, a factor of five, or a factor of ten.
When you use deterministic things, as we've been
doing in regulation, you really have no idea what the safety
margin is. We have the potential in performance-based
activities to, in fact, separate safety margin and apply
specific safety margins.
Has any thought been given to what is required in
order to do that, in a way? Is it the guidelines for doing
the performance assessment that have to be consistent and
uniform? You can't arbitrarily assign bounding calculations
one place ten times what's normal and in other place, 20
times what's normal?
That there is something inherent in
performance-based is really to do the best analysis you know
how, and then get concurrence on how big a safety margin you
add. Has this potential aspect of performance-based been
considered at all in what you're doing?
MR. KADAMBI: I believe the answer to your
question is yes, because it's really embedded in the
guidelines themselves. I think it occurs in a couple of
different places.
It occurs within the viability guidelines
themselves, and also it is part of the guidelines that deal
with effectiveness and deficiency, where, you know, you have
to make sure that there is a sufficient safety margin, but
if it is a safety margin that is, you know, not based on
robust information, if it uses deterministic analysis, you
know, which may not reflect the actual circumstances that
the performance demands, then I think the guidelines force
you to make those kinds of inquiries. That's my own view.
MR. LEVENSON: Well, from what little we've seen
or heard here, it isn't obvious to me, where they are. I
would expect there to be a separate guideline as to how one
determines what is an appropriate safety margin to use in
any case.
The size of the safety margin should be a function
of how bad are the consequences. And if it's embedded,
generically, I don't know how you do a good job of
identifying that.
MR. KADAMBI: Well, I guess I'm not sure that I
can give you an answer that would fit all cases, because I
think the guidelines are meant to be sort of things that you
apply in specific areas.
I mean, for example, if you're dealing with
transportation, it would be one set of approaches in terms
of what kind of safety margins you deal with, as opposed to,
you know, waste disposal facility or even in reactors,
whether you're talking about the reactor coolant system or
the containment system.
So, I mean, I'm reluctant to offer much by way of
a definitive answer, but I believe the language of the
guidelines is that a sufficient safety margin exists.
That's what it says in the Attribute D of the Viability
Guidelines.
And I think what we mean by that is that it should
be definitely enough to make sure that even if the
performance parameter is exceeded or is not -- the criterion
is not met, that you don't face an immediate safety concern.
MR. LEVENSON: Yes, but that's a different issue.
In fact, you're not setting specific safety margins as you
normally do in engineering. I don't expect an answer. This
is a work-in-progress, and you were asking for feedback.
MR. KADAMBI: Yes.
MR. LEVENSON: And I'm just saying that from my
personal standpoint, one of the potential advantages of
going this way is that you can separate out and say we've
done the best analysis with the best data we have, and now
we're going to add another safety factor of ten or two or
five, depending on consequences.
And that doesn't seem to me to be inherent in the
guidance you have so far. And maybe it is; maybe it isn't.
That's just personal feedback.
MR. KADAMBI: Okay, well, thank you for the
feedback.
CHAIRMAN GARRICK: Ray?
DR. WYMER: I don't have a question. I have a
comment. Like so often when I hear these general high level
type criteria, goals, things that are not specific, I have a
hard time understanding them. I have a hard time wrapping
my mind around that. At best, I really only understand them
when I see them applied to specific fairly large, complex
cases, because then I have a lot better grasp of what they
really mean and what they are trying to do.
That's sort of what Milt was getting at a little
bit, I think, too, and so I guess what I would say is I look
forward to and would expect to, sometime in the not terribly
distant future, to hear a discussion of the application of
these things to a real live, large, and fairly complex
situation so that all the facets of these guidelines are
brought out, so that at least I can understand them.
That is all I have to say.
MR. KADAMBI: We will try to do that.
CHAIRMAN GARRICK: George?
DR. HORNBERGER: I basically had a similar
reaction to Ray's. I was going to make essentially the same
point.
I am curious along those lines. To a certain
extent your high level guidelines obviously make sense. You
know, you find a parameter directly related to safety that
you can measure easily and monitor in real time and et
cetera, et cetera.
My question is you must have given some thought as
to how they might apply to a somewhat more complex case.
You mentioned you were doing a simple case, or a relatively
simple case with combustible gases, and I don't know much
about this, and I am not looking for a tutorial on
combustible gases, but I could at least postulate a
situation where you might be looking at a gaseous emissions
parameter and basically say fine, you want the activity
level to be less than "x" and you are going to regulate that
way, and then you let it up to the licensee to determine
whether it is satisfied by precipitation or filters or
whatever. That sort of makes sense in the performance based
language.
On the other hand, then I think about it a little
more deeply. That radioactivity doesn't just go away. It
goes perhaps into the fly ash, and then the question comes
in an overall systems framework have you improved things by
having rewarding the licensee for putting the radioactivity
into the fly ash, which then has to be disposed, and is
there some other performance measure that should somehow get
weighed in there.
It is a hypothetical, but have you gone through
the thinking as to how you would deal with fairly
complicated situations?
MR. KADAMBI: I have done some thinking on my own
on this. It is not -- at this stage of development, you
know, in the Staff's work we are not really at a stage where
we can try to answer some of these questions, but I think
there is no doubt that there are different levels of
aggregation within a hierarchy that you can put these things
together.
You can bring systems together into providing a
functionality and then you can start putting things together
in different ways, sort of, you know, the opposite of the
reductionist approach, and you can apply the same principles
though at whatever level one chooses to apply it, as long as
you define the boundary conditions and say that whatever
conclusions are drawn apply within those boundary
conditions, so I believe that the benefit of the guidelines
is that it sort of forces you into answering some of these
questions as to draw the boundary lines around the issue and
also points to the limits of whatever conclusions that you
can draw from it.
CHAIRMAN GARRICK: I just want to leave a couple
of comments with you.
One is on this issue of performance measures.
When I hear people talk about components and safety trains
and systems and release and what have you as candidates for
performance measures, I get a little nervous. If we are
really genuine about risk informed performance based
regulation, and if we are really genuine about relieving
licensee and licensing burden, I think in the guidance we
want to be very careful to be offering guidance on measures
that integrate the effect of what all of these measurements
of all of these subunits might constitute, and this is why
the NRC is in a bit of a dilemma in their reactor side right
now is because the principal measure is core damage
frequency and in a sense you can argue that that is not
really a major risk in the spirit of the Atomic Energy Act,
which is to protect the health and safety of the public.
Now to be sure, it is a precursor and a pretty
darned important one, but the point is if you are really
genuine about performance based issues the performance of
core damage in one plant has a different impact on safety
risk than that same core damage frequency in another plant,
so I think that one of the things you would be very
well-advised to do is to be thinking in terms of performance
measures that carry out as much of the integration, of the
impact of other measures as possible.
The ultimate integrator, of course, is health and
safety to workers and the public.
The other thing I want to mention is that in your
guidance I think you also need to be very careful about
putting constraints on what is meant by risk assessment
itself. To do a risk assessment and not be sensitive to
crew performance for example, or operator performance for
example, or fitness for duty for example, is not doing a
risk assessment, and yet there is the indication in the
guidance that these are outside or the implication that this
is outside the scope of PRA.
As I said earlier, it may be outside the scope of
a specific PRA because the analyst chose to do it that way,
but the thought process is not constrained in any way, and
we all know that the evolution of risk assessments of
nuclear power plants has been extensive with respect to
increasing the amount of time we spend in addressing
operator performance, for example, which with that kind of
thinking one might say is outside of a PRA.
So we should not confuse what has been done or
maybe what the practice is as to what could be done, and we
all know that there is some very interesting work going on
in the human factors arena, human factors research, in
organizational performance research in terms of being able
to embrace those kinds of impacts into our models, so that
is just a couple thoughts that I would like to leave you
with.
Okay. Are there any questions or comments from
the Staff?
DR. HORNBERGER: Could I ask you a question?
CHAIRMAN GARRICK: Yes.
DR. HORNBERGER: I don't think that I quite
understand your statement that one could go directly to
something like human health as a performance measure.
It seems to me that that really would go to the
problem that some people have pointed out, that if you ever
measured human health effect you would say it is too late
and it would violate I forget which of the -- IV in the
guidelines.
It seems to me that you have to have some
surrogate performance measure well before human health.
Did I miss something?
CHAIRMAN GARRICK: Well, partly and partly you
didn't. My philosophy on this is that clearly you do
calculate what these precursor events are in terms of their
likelihood of occurrence.
As a matter of fact, if you do the right kind of
job of calculating something like a health effect you should
have embodied in that analysis a very transparent indication
of what these precursor contributors are to that health
effect risk, but what I am saying is that I can envision
scenarios that would improve the performance of a nuclear
power plant, for example, in terms of core damage frequency
that would actually increase the risk to the health and
safety to the public.
That is the thing that we have got to be very much
on the alert for is that ultimately we are committed here to
protect the people and the environment, and it just seems
that if we are talking about high level guidance and we are
not starting with the people and the environment, and we are
worrying about components and trips and what have you, we
are on the wrong track because that is plant-specific, it is
facility-specific, and general guidance on that in most
cases won't have much meaning in a specific application.
So, no, I don't think that you get into a
situation where you don't understand very clearly what the
role is of a subsystem or what have you with respect to an
overall and overarching performance measure, but I am
suggesting that we ought to be addressing measures that
really address explicitly what it is we are concerned about
and what it is we are concerned about and obligated to is
protect the health and safety of the public, and if we don't
focus on that then we are missing the boat.
I think that we have got quite a bit of work to do
in some of these areas in that arena.
Okay, yes, Milt?
MR. LEVENSON: I have a question. I am a little
confused. I made the mistake of going back and reading your
viewgraphs, and one of them says that the Commission
directed Staff to develop high level guidelines to identify
and assess the viability of candidate performance based
activities. I guess it isn't very clear to me how the
guidelines that you have discussed could be used to identify
and assess viability of performance based activities. I am
not sure what those words mean.
MR. KADAMBI: Well, the way I have interpreted the
words is to say that the guidelines become attributes which
would characterize a performance based activity. In other
words, if there is a candidate activity and you want to
determine is it performance based right now, you would use
the guidelines as attributes to compare its existing
attributes with what are the attributes of a performance
based activity.
If there is a difference between the two, you
would say, well, it may be worthwhile making this activity
performance based by changing the attributes in some defined
way.
MR. LEVENSON: Okay. John, just as a follow-up to
your comment about core damage, you know we need to remember
there have been 12 reactors that have had core meltdowns and
Chernobyl is the only one with any significant health safety
impact, so core damage per se doesn't automatically mean
that.
CHAIRMAN GARRICK: That is part of the point.
Okay. We have a request for somebody to make a
comment from the Public Citizen group. I think we ought to
accommodate that right now, and would you announce your name
and affiliation, et cetera?
MS. GUE: Yes, thank you. My name is Lisa Gue. I
am a Policy Analyst with Public Citizen's Critical Mass
Energy & Environment Program.
Public Citizen has previously given comment on
this issue both through the workshop process and also in a
statement to ACRS committee as well, but I do appreciate the
opportunity to briefly just outline our main concerns with
this process to your committee as well.
I think Mr. Kadambi has accurately categorized our
concerns into those relating both to confidence and then
also implementation, and in terms of confidence I guess
fundamentally we are concerned with a process that orients
itself with the objective to reduce regulatory burden to
licensees.
The regulatory structure should emphasize safety
at its core, as you mentioned, Mr. Chairman, and if this is
too much of a burden on industry, then the industry itself
is not viable.
Secondly, relating to our confidence concerns, I
guess, is that we are skeptical of the benefits of the
flexibility, of providing flexibility to licensees in terms
of how they meet these performance objectives.
We find it hard to believe that this is actually
going to result in industry clamoring to be more creative,
to be more stringent in how they protect public safety.
Certainly from a public interest perspective, historically
we don't really have any reason to be confident that this
would be the result and certainly from an economics
perspective it would seem that the more likely scenario
would be for a race to spend as little money as possible,
resulting in cost-cutting to meet the lowest possible
standards.
Relating more to the implementation concerns,
clearly the risk information to determine, using risk
information to determine performance indicators would be key
to the process, as the presentation indicated, and yet
thinking of the Yucca Mountain proposal right now for a
repository and also an unprecedented transportation program
of high level waste and spent fuel, I guess our concern
would be that our ability to accurately assess the risk is
really limited by a relatively short history of experience
in this area, and I mean that relative both to the scale of
the transportation program, of the specific transportation
program that is being considered for the Yucca Mountain
repository and also in terms of the license period for that
repository.
Secondly, that same ability to accurately assess
the risk is limited by simply the limits of our own human
imagination to conceive of the possible risks, given the
shortcomings of experience in that regard.
I guess related to that then is this underlying
concern that has sort of been alluded to already, that a
performance based structure really can only respond to
failure and a viable regulatory structure needs rather to be
conservative enough to ensure public safety.
I think that some of viability criteria that seek
to integrate some of the comments that have been raised
previously actually do address some of these concerns, and
yet I am left wondering if they could actually then be
usefully applied to any waste scenario, particularly with
respect to high level waste and spent fuel, and if not, if
all of the waste scenarios would basically be ruled out,
according to these viability standards then I am wondering
in what sense and why are these guidelines being proposed as
high level.
Thank you.
CHAIRMAN GARRICK: Thank you. Are there any other
comments?
[No response.]
CHAIRMAN GARRICK: All right. Well, thank you
very much.
MR. KADAMBI: Thank you, Mr. Chairman.
DR. HORNBERGER: Amazing time management by the
Chairman.
CHAIRMAN GARRICK: I think this brings us to a
point in our agenda where we adjourn for lunch.
[Whereupon, at 12:16 p.m., the hearing was
recessed, to reconvene at 8:30 a.m., Wednesday, July 26,
2000.]
Page Last Reviewed/Updated Monday, October 02, 2017