1
1 UNITED STATES OF AMERICA
2 NUCLEAR REGULATORY COMMISSION
3 ***
4 BRIEFING ON PRA IMPLEMENTATION PLAN
5 ***
6 PUBLIC MEETING
7 ***
8
9 Nuclear Regulatory Commission
10 Commission Hearing Room
11 11555 Rockville Pike
12 Rockville, Maryland
13
14 Wednesday, October 15, 1997
15
16 The Commission met in open session, pursuant to
17 notice, at 10:05 a.m., the Honorable SHIRLEY A. JACKSON,
18 Chairman of the Commission, presiding.
19
20 COMMISSIONERS PRESENT:
21 SHIRLEY A. JACKSON, Chairman of the Commission
22 GRETA J. DICUS, Member of the Commission
23 EDWARD McGAFFIGAN, JR., Member of the Commission
24 NILS J. DIAZ, Member of the Commission
25
. 2
1 STAFF AND PRESENTERS SEATED AT COMMISSION TABLE:
2 KAREN D. CYR, General Counsel
3 JOHN C. HOYLE, Secretary
4 MARGARET V. FEDERLINE, NMSS
5 SCOTT F. NEWBERRY, NRR
6 L. JOSEPH CALLAN, EDO
7 ASHOK C. THADANI, OEDO
8 THOMAS L. KING, RESEARCH
9 PATRICK W. BARANOWSKI, AEOD
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
. 3
1 P R O C E E D I N G S
2 CHAIRMAN JACKSON: Good morning. I'm pleased to
3 welcome members of the staff to brief the Commission on the
4 status of the PRA Implementation Plan.
5 The PRA Implementation Plan was first issued in
6 August 1994. The Plan is intended to be a management tool
7 that will help to ensure the timely and integrated
8 agency-wide use of PRA methods and technology in the
9 agency's regulatory activities.
10 The last written update on the status of
11 activities in the PRA Implementation Plan was received
12 recently by the Commission. The Commission was last briefed
13 on the Plan in May of this year.
14 During today's briefing, the staff will cover its
15 recent accomplishments -- status of key activities,
16 responses to SRM's, and future activities. These new
17 activities include the development of standards for PRA and
18 the evaluation of the need to revise the Commission's safety
19 goal policy statement.
20 The staff's recent accomplishments -- and I'm
21 taking Joe's thunder, probably -- include the issuance of
22 draft risk-informed regulatory guidance for inservice
23 inspection for comment. A public workshop to discuss public
24 comment on these documents is being planned for later this
25 year. I am my fellow commissioners are looking forward to
. 4
1 your briefing today, and I understand that copies of the
2 viewgraphs are available at the entrances to the meeting.
3 So, if none of my colleagues have any comments they wish to
4 make, Mr. Callan, please proceed.
5 MR. CALLAN: Thank you, Chairman. Ashok Thadani,
6 to my right, who is the deputy EDO for regulatory
7 effectiveness, will lead the staff's discussion this
8 morning, but before I turn the meeting over to him, let me
9 introduce the other members at the table.
10 We have a diverse group of executives at the table
11 representing all the large program offices -- to my far
12 left, Margaret Federline, representing the Office of Nuclear
13 Material Safety and Safeguards; Scott Newberry, representing
14 the Office of Nuclear Reactor Regulation. I've already
15 introduced Ashok. To his right, Tom King, representing the
16 Office of Nuclear Regulatory Research; and then, Pat
17 Baranowski, representing the Office of AEOD.
18 With that, Ashok.
19 MR. THADANI: Thank you, Joe. Good morning.
20 CHAIRMAN JACKSON: Good morning.
21 MR. THADANI: Well, Chairman, as you know, the
22 most prominent activity underway in the Implementation Plan
23 continues to be the development of regulatory guidance
24 documents -- that is the Reg Guides and Standard Review
25 Plans.
. 5
1 In August, we held a workshop. I attended that
2 workshop -- part of the workshop. It was a three-day
3 workshop. It was very well attended. It was very lively;
4 there was a great deal of give-and-take. We received 30
5 sets of written comments on the guidance documents. Some of
6 the comments are quite significant, some significant in
7 terms of technical issues, as well as policy matters.
8 An example of a policy issue that has been raised
9 is concerned with having so-called tight limit of using a
10 core damage frequency of 10-4 per reactor year in terms of
11 considering any further small increases in risk.
12 There were questions along the lines of
13 clarification as amendments come in with varying impact on
14 core damage frequency, what was meant by different level of
15 analyses, as well as different level of management
16 attention, as we discussed in the past.
17 There were also substantial questions in the area
18 of uncertainties -- to what extent the detailed uncertainty
19 analyses need to be conducted for very minor or very small
20 changes in, let's say, core damage frequency.
21 We're analyzing these comments, and we're in fact
22 planning to discuss our initial thoughts on these issues
23 with the Advisory Committee next week. We would expect --
24 as Tom is going to summarize some of these issues, we would
25 expect to come back to the Commission on the policy issues
. 6
1 for guidance.
2 CHAIRMAN JACKSON: Let me ask you a question that
3 occurs to me immediately. You mention this issue of this
4 tight upper limit vis-a-vis the core damage frequency -- you
5 know, 10-4. At the same time -- and I'm only looking aqt
6 what I read in the trade press -- there's discusion about
7 the industry having a perspective of wanting to use PRA's as
8 they are.
9 The third piece is, when I've pressed the staff in
10 meetings that have been going on since I've been here, about
11 the PRA's -- what you can say about them, the quality of
12 them, et cetera, et cetera, et cetera, et cetera -- there
13 has been some kind of squishiness and indeterminacy, and
14 there was some issue about PRA's that were graded, you know,
15 one, two, three, or something.
16 And so, the real question that I have is, frankly,
17 this -- that I think, in terms of your bringing a policy
18 issue to the Commission that I think has to be addressed, if
19 there's some variability in the PRA's, either in terms of
20 the methodologies or assumptions, et cetera, or fundamental
21 quality of them, that tracks with this issue of what kind of
22 limits or how much flexibility there can be in limits, et
23 cetera, et cetera, et cetera, that if you don't get at that
24 -- okay? -- and somebody says, "Well, my core damage
25 frequency is 10-5," and if you did the calculation another
. 7
1 way, maybe it would be 10-3, and if you did it another way,
2 maybe it would be 10-6.
3 What are you doing about that? I have a question
4 about that. How do you handle that kind of issue in getting
5 at this? Because there's an assumption underlying all of
6 this that the numbers, as presented, have meaning.
7 MR. THADANI: Yes.
8 CHAIRMAN JACKSON: And that's a question.
9 MR. THADANI: Yes, clearly. And, in fact, that
10 was also one of the issues that was debated, which is the
11 guidance that we have in a NUREG document in terms of
12 quality for risk assessment.
13 CHAIRMAN JACKSON: The statement has been made
14 that you're looking for a gold-plated PRA.
15 MR. THADANI: Yes. There is concern that the risk
16 assessments don't need to be of that quality, and the
17 comment was made that perhaps the use of PRA or the quality
18 of the PRA should be driven by the application.
19 We have some thoughts on those issues in terms of
20 if the change has a truly negligible estimate impact on core
21 damage frequency -- we could define negligible - and, for
22 that change, does one need to go through a detailed
23 uncertainty analysis, for example.
24 It may be that that's not necessary, hut that
25 these are the issues that the industry has raised. We're
. 8
1 looking at them. We haven't really come to any conclusions
2 on that.
3 CHAIRMAN JACKSON: Right, but I think you have to
4 be thinking about these things. I mean the issue is, if you
5 have variability in the PRA's, in the quality of them or how
6 they're done, and getting straight at this issue of
7 uncertainty, because I've raised the issue in the past --
8 MR. THADANI: Yes.
9 CHAIRMAN JACKSON: -- the question is how much
10 variability and what degree of uncertainty, how much can be
11 tolerated for which regulatory use? Because I think it
12 tracks into some of the legal questions that arise.
13 MR. THADANI: Yes, it does.
14 CHAIRMAN JACKSON: And what's the difference
15 between the use of PRA in a risk-informed framework as
16 opposed to what I think some of the legal analysis has
17 focused on, which is risk-based?
18 MR. THADANI: That's right.
19 CHAIRMAN JACKSON: We're talking a risk-informed
20 framework, and how do these questions play into that?
21 MR. THADANI: Yes, indeed.
22 CHAIRMAN JACKSON: And I'm saying that, if you
23 don't address those, then don't send the paper, because I
24 think you're going to have to address these.
25 MR. THADANI: We would intend to address them.
. 9
1 Those are clearly the central issues. And I think, as you
2 have noted, it is significant to make sure that we are
3 talking about risk-informed and not risk-based, because
4 risk-based -- as you have noted, and it's pretty clear -- is
5 truly relying on numerical analyses almost as a central
6 basis. Our guidance documents don't really do that. I
7 think, in any case --
8 CHAIRMAN JACKSON: Sorry. I didn't mean to
9 preempt anything you were going to talk about.
10 MR. THADANI: I think, instead of my taking any
11 more time, it's better to just jump right into the issues.
12 CHAIRMAN JACKSON: Well, I think this commissioner
13 wants to jump first.
14 COMMISSIONER McGAFFIGAN: I just wanted a
15 clarification question. On the 10-4 --
16 MR. THADANI: Yes.
17 COMMISSIONER McGAFFIGAN: -- core damage
18 frequency, the upper limit, is that an IPE or an IPEEE
19 number or the sum of the two? Because the IPEEE numbers,
20 some of them, were quite large.
21 MR. THADANI: The intention was not only it's the
22 sume of IPE plus IPEEE, but it also should include shutdown.
23 That is, it is the overall core damage frequency, a mean
24 value. And that's what we had said, it would be a mean
25 value, which, by the use of the term, "mean," we imply some
. 10
1 knowledge of uncertainties in that. So the core damage
2 frequency of 10-4 include internal events, external events,
3 and low power and shutdown.
4 COMMISSIONER McGAFFIGAN: My recollection from
5 yesterday's briefing is that, on IPEEE's, we've reviewed
6 some, but we're still waiting for in the teens of
7 submittals, if I'm correct.
8 MR. THADANI: That's correct. That's correct.
9 COMMISSIONER McGAFFIGAN: So there's a large body
10 of folks who aren't at square one in this.
11 MR. THADANI: That's correct. There's a little
12 history to this. In June of 1990, there was an SRM which
13 gave some guidance. At that point, the Commission did not
14 want us to subdivide the so-called 10-4 in sub-allocating to
15 different contributors, so to speak.
16 Recognizing that there are significant questions
17 on methodology of shutdown risk, that that's not available,
18 that external events IPE's have not all been completed, so
19 what we have is a piece of the information.
20 One would have to sub-allocate -- make certain
21 arguments about what contribution there might be from
22 shutdown, from external events, recognizing what we have, by
23 and large, are IPE's that deal with internal events only.
24 This is again discussed in our guidance documents,
25 because that recognition is there, but the licensees would
. 11
1 have to pull together some arguments as to what those
2 contributions might be.
3 CHAIRMAN JACKSON: To bound them in some sense.
4 MR. THADANI: To some sense, yes.
5 CHAIRMAN JACKSON: A bounded contribution.
6 MR. THADANI: One needs to have nigh confidence
7 that those don't pop up as being the most significant
8 issues. We're going to be having, I expect, fairly
9 significant interaction with the Advisory Committee on these
10 issues next week. Our intention is to pull together some of
11 these thoughts and get the information to the Commission
12 soon after that.
13 CHAIRMAN JACKSON: Doesn't, to some extent, the
14 question revolves around what is the role of a numerical
15 analysis in a risk-informed, as opposed to a risk-based,
16 framework?
17 MR. THADANI: Yes. Yes. What's the role? We
18 made an attempt when we identified five basic principles in
19 our guidance documents that we would follow. That was an
20 integration of probabilistic approach, as well as
21 engineering analyses and what we call deterministic
22 thinking. How do we integrate that?
23 It seems to me that's a much better way to make
24 risk-informed decisions. It doesn't rely entirely on
25 numerical analysis aa the basis for decision, but that
. 12
1 numerical analyses do play a part in that decision.
2 CHAIRMAN JACKSON: There are two.
3 MR. THADANI: Unless there are some other general
4 questions, I think I'll just go to Tom King.
5 CHAIRMAN JACKSON: Why don't you go ahead.
6 MR. THADANI: Tom.
7 MR. KING: Thank you, Ashok. In fact, let me just
8 mention -- to follow up on your comment, Chairman Jackson --
9 some of the comments we did get from industry were related
10 to the role that we've put forward with using the PRA
11 numbers in a risk-informed fashion.
12 [Slide.]
13 Some of the comments were directed toward -- they
14 thought we've gone too far in using them in some sort of
15 decision criteria and that they really ought to be geared
16 toward looking at risk ranking, looking at trending and so
17 forth, and not hard and fast decision criteria.
18 CHAIRMAN JACKSON: Yeah, but then what do you do
19 if you talk about tech spec changes based on these analyses
20 or inservice inspection requirement changes based on these
21 analyses or inservice testing changes? So it sounds like
22 there's some variability in what the industry wants to do.
23 I mean, yeah, you can do kind of generalized risk-ranking
24 and sensitivity analyses and kind of configuration analysis.
25 That's one use. But if you're actually talking about
. 13
1 changes or relaxations in tech spec requirements or
2 inspection requirements, that's a different kind of thing.
3 MR. THADANI: Yes.
4 CHAIRMAN JACKSON: That's what I meant about what
5 is the use to which things are going to be put.
6 MR. THADANI: Yes.
7 CHAIRMAN JACKSON: And, therefore, that relates to
8 this question of quality, certainty, how much can be
9 tolerated, et cetera, et cetera.
10 MR. THADANI: If I may just add to just that
11 point, it is easier said than done. When I said the quality
12 should be driven by application -- which I think most of us
13 would --
14 CHAIRMAN JACKSON: Agree with.
15 MR. THADANI: -- agree is reasonable -- in some
16 areas -- for instance, inservice testing, quality assurance
17 -- that the issue propagates through the plant. It's not a
18 narrow issue, which means you have to rely on the overall
19 risk assessment. It is not just a small piece that we're
20 talking about. That means one has to have credibility in
21 the overall study, itself. And so, for that application,
22 the demand for the quality, it seems to me, would be very
23 significant.
24 On the other hand, if it has to do with an issue
25 -- let us just say tech spec changes on accumulators and
. 14
1 PWR's -- then I know that's a narrow issue. I know that
2 there are only a couple of sequences where that system plays
3 a part. I can make sure the quality of that analysis is
4 good. And, in fact, if the change in, say, core damage
5 frequency is very small, then one could have pretty good
6 confidence that the overall impact is, in fact, pretty
7 minimal.
8 ISD/QA issues of that type really, I think, go to
9 the heart of the broad issue of quality of the whole study,
10 not part of the study.
11 CHAIRMAN JACKSON: Right. I know Commissioner
12 McGaffigan has a question, but, actually, I'm going to put
13 my commissioner on the spot, because I know he has had some
14 fairly significant perspectives on this.
15 COMMISSIONER DIAZ: Oh, I do agree. I just really
16 would like to ask whether we are actually getting closer to
17 define whether there should be a level playing field on PRA
18 where, no matter what the application is, there is a base
19 quality that we can feel we can use risk information
20 consistently inside and outside, because I think that
21 becomes clearly an issue.
22 Until we can feel that we can use this across the
23 board with a certain level of quality and the industry
24 realizes that that will increase safety and it would also
25 reduce burden, we will always be writing things.
. 15
1 I think there has to be a demand that there be a
2 level playing field at some point, and we look for your
3 guidance in establishing what that level is. There has to
4 be. We can't be looking always at the minor application.
5 You have to have a base.
6 CHAIRMAN JACKSON: What's the based? Commissioner
7 McGaffigan.
8 COMMISSIONER McGAFFIGAN: It's really very closely
9 related, in that my sense is that the big dollar savings are
10 going to be in the complicated cases and where you're going
11 to need some sort of decent quality. They have to see that
12 that investment in having that quality PRA across the board
13 -- or at least for that plant, the IPE/IPEEE -- that that
14 will then result in savings.
15 If they don't see that, then probably they won't
16 go. But if they do see that, then maybe that up-front
17 investment can even be still made in a coming deregulated
18 environment. A good quality PRA -- could you remind me?
19 How much does a good quality PRA cost?
20 [Laughter.]
21 MR. THADANI: If you were starting from a clean
22 sheet -- which, in this case, we're not --
23 COMMISSIONER McGAFFIGAN: Right.
24 MR. THADANI: But if one were starting from a
25 clean sheet -- I would let Tom correct me if I'm wrong --
. 16
1 but the estimates I have got have been anywhere from 2 to
2 6 million dollars, depending on scope and so on, and if
3 we're not doing shutdown, that will save some pieces, as
4 well.
5 COMMISSIONER McGAFFIGAN: And how much does it
6 then cost to maintain that PRA, to keep it up to date -- in
7 an annual O&M cost? Do you know?
8 MR. KING: Well, just as an example, the South
9 Texas project people, they have a PRA group of around four
10 to five people that maintains it and supports risk-informed
11 applications and so forth.
12 COMMISSIONER McGAFFIGAN: So that would be about a
13 half million a year, maybe.
14 MR. KING: To get a ballpark idea.
15 COMMISSIONER McGAFFIGAN: Well, given some of the
16 applications that they want to apply this to, it strikes me
17 that that's a pretty good investment you could sell to a
18 board.
19 CHAIRMAN JACKSON: Right. It's a question of what
20 the saving is, if you could have a configuration. You don't
21 have to shut down some other things.
22 MR. THADANI: And we have heard some estimates on
23 different applications -- including, in South Texas,
24 technical specification changes -- as very significant
25 annual savings.
. 17
1 MR. KING: Right.
2 MR. THADANI: Very significant annual savings. So
3 an area where we're being told that the savings are not very
4 significant, I believe, was on inservice testing, actually.
5 That's what I was told by, I think, South Texas.
6 CHAIRMAN JACKSON: Yes.
7 MR. THADANI: But, in general --
8 CHAIRMAN JACKSON: But I do think that we have to
9 come out with some baseline and then, for the special big
10 applications, what additional is really required?
11 MR. THADANI: Yes. Yes. We have developed
12 guidance in terms of what our expectation was. Industry
13 doesn't agree. We have to deal with those issues next, the
14 significance of some of them.
15 CHAIRMAN JACKSON: All right, I'll adjourn the
16 meeting.
17 [Laughter.]
18 MR. KING: In conclusion, yes. Could I have
19 slide 2, please.
20 [Slide.]
21 What we're going to concentrate on today in the
22 briefing is what has happened over the past six months since
23 the last briefing in May. As you'll see, there has been a
24 lot of work on a lot of fronts.
25 We've made significant progress in the
. 18
1 risk-informed guidance documents. We've got some new
2 initiatives and activities we're going to talk about that
3 have been added to the quarterly update.
4 We've received several SRM's which we're going to
5 respond to specifically today in the briefing. I'm going to
6 focus on the major items, and then, at the end, we'll also
7 come back and talk about where we're going from here in the
8 future over the next few months. Slide 3.
9 [Slide.]
10 Since we briefed you last, you've received two
11 quarterly updates -- one in July and the one yesterday.
12 There has also been several papers and SRM's that have been
13 issued over that time frame.
14 Back in May, we had an SRM that asked the staff to
15 expedite activities on the use of IPE results in
16 prioritizing inspection activities, improving regional
17 capabilities, and providing inspector training. When we get
18 to slide 11, we'll address that specifically.
19 We had issued our draft regulatory guides for
20 comment and held a workshop in August. We'll talk more
21 about that later.
22 [Slide.]
23 Slide 4. On June 5th, there was an SRM that
24 requested our plans for training the NRC staff, which is a
25 very important activity to implement risk-informed
. 19
1 regulation. We've got several slides. We'll talk about
2 that activity, what has happened there and where we're
3 going.
4 On June 13th, we had an SRM that requested
5 progress reports on the voluntary industry effort to provide
6 reliability and availability data. We'll talk about that.
7 And, as I mentioned, we issued a couple of quarterly
8 updates. Slide 5.
9 [Slide.]
10 COMMISSIONER DIAZ: Let me just make a comment.
11 Let me put my professor's hat in here for a minute.
12 MR. KING: Yes.
13 COMMISSIONER DIAZ: And look at this, the training
14 and so forth. I haven't seen all of the final objectives on
15 it, but if I may think of how I used to deal with some of
16 these issues.
17 You know, I think one basic objective is, when you
18 finish this training, anybody in NRC that has to do with any
19 policy, decision-making, ruling, contact with licensee,
20 should have clear in his mind, when somebody says, "This is
21 risk-informed," that you have a picture.
22 When somebody says, "The coals are hot," or "Your
23 coffee is hot," you've got a picture. And when they say,
24 "risk-based," they should have a picture of what it is, and
25 when they say, "risk-informed, performance-based," it should
. 20
1 be clear.
2 MR. KING: Yes.
3 COMMISSIONER DIAZ: Because we keep mixing terms
4 or mixing terminologies and things. I think it has to be,
5 basically, that the NRC has to be, in this country, the
6 agency in which every technical person has a clear picture
7 of what each one of these things means and their
8 relationship to how we regulate.
9 CHAIRMAN JACKSON: Right.
10 COMMISSIONER DIAZ: Does that make sense?
11 MR. KING: I agree, I agree.
12 CHAIRMAN JACKSON: That's the basis. We can't do
13 anything otherwise.
14 MR. KING: Part of our training is directed toward
15 telling the staff, what do our documents say? What's
16 expected? How do we make decisions? Then part of it is
17 directed toward the technology of PRA's.
18 CHAIRMAN JACKSON: Right, but I think he's making
19 another statement, which I think is an important one, and
20 that is to get the definitions straight and that the
21 baseline of training people, before you get to the
22 technology use, is to have a common vocabulary that we all
23 understand.
24 If we're doing risk-informed, performance-based
25 regulation, PRA and the PRA Implementation Plan are tools
. 21
1 along that road, but they have to be put within the right
2 context, and so you need to be thinking about that, and
3 you're going to tell us about the training. But we have to
4 make an assumption, and you're going to hopefully give us
5 comfort in that assumption, that undergirding has to do with
6 clear definition.
7 MR. KING: If we can go to slide 5 and briefly
8 talk about the major accomplishments over the past six
9 months.
10 [Slide.]
11 Most recently, we completed the draft
12 risk-informed guide and SRP section on inservice inspection.
13 The Federal Register notice will be printed today announcing
14 the availability and announcing a workshop in November to
15 discuss these documents.
16 We've also made progress with some pilot
17 activities. We've received two applications so ar for
18 pilots, and I understand there may be up to three more
19 coming.
20 There have been a number of issues raised in the
21 Federal Register notice on inservice inspection that we're
22 soliciting comment on and will be discussed at the workshop.
23 These deal with issues like the scope of submittal,
24 degradation mechanisms and so forth, a number of technical
25 issues.
. 22
1 We received the Commission's October 1st staff
2 requirements memorandum, and we'll continue to work with the
3 industry on the two methodologies that are being developed
4 -- the qualitative and the quantitative -- and continue to
5 work with the pilot programs.
6 As I mentioned, we had held our public workshop in
7 August on the other reg guides and standard review plans,
8 and I'll talk more about those later.
9 We've completed work now on the IPE insights
10 report, what we call NUREG-1560. We had had a workshop in
11 April where we received 25 sets of comments. A number of
12 those required clarifications and expansion of the document,
13 and we've done that, and we're about ready to send it to the
14 printer's.
15 We included in our most recent quarterly update a
16 copy of the executive summary and the comments and responses
17 to the comments that were received. We've expanded our
18 training program -- and, again, we'll talk about that later
19 on. If I can go to slide 6, which now covers the new
20 activities that will show up for the first time in the
21 quarterly update you received yesterday.
22 [Slide.]
23 Over the past several months, we've had a couple
24 of meetings with NEI on an initiative that they've come in
25 with which involves using a full-scope PRA. By "full-scope"
. 23
1 I mean full power shutdown, external events, to look at
2 their plant, get a risk profile of the various systems an
3 components, compare it to the requirements, compare it to
4 their operations and maintenance cost, with the intent of
5 coming back and, in a risk-informed fashion, suggesting
6 changes to their current licensing basis.
7 What we've discussed with them are some of the
8 ground rules by which this study would be done, and they
9 also want to do three pilot projects to try out the review
10 and criteria which would be used to select and make
11 decisions on these on these items. We hope to finalize out
12 discussions and get this underway by December.
13 We've also had some discussions with the American
14 Society of Mechanical Engineering regarding development of a
15 national consensus standard for PRA that would cover scope
16 and quality. This would build upon the work we've done with
17 our draft NUREG-1602. We've worked with them and drafted up
18 sort of a charter for the group. The group involves not
19 just ASME people, but other people from industry, from
20 universities, as well as NRC.
21 COMMISSIONER DIAZ: Any key issues from trying to
22 bring the standards with 1602 together? Any great
23 differences, similarities?
24 MR. KING: Well, I'm not sure we're far enough
25 along. I could ask our person who attended the meetings
. 24
1 with ASME if she wants to expand on where we stand on that
2 activity. Mary. Mary Druin is from the Office of Research,
3 and she has been our representative working with ASME.
4 COMMISSIONER DIAZ: We're driving for consistency.
5 It would be nice to know how consistent we are.
6 MS. DRUIN: The ASME board met last Thursday and
7 did unanimously vote to develop a PRA standard. They
8 understand the criticality of the issue and our need to move
9 forward real swiftly in this area, so there is going to be
10 meetings, hopefully, like, biweekly. Biweekly -- is that
11 twice a month or is that twice a week? Anyway -- I always
12 get those terms mixed up -- twice a month.
13 CHAIRMAN JACKSON: Twice a week, if that is
14 biweekly. Bimonthly is twice a month.
15 MS. DRUIN: That's right. But hopefully it will
16 -- I mean we don't have the standard yet, but it is trying
17 to address the quality, the level of the detail, and get
18 into a lot of the issues that were discussed at the very
19 beginning of the meeting this morning.
20 CHAIRMAN JACKSON: Have they laid out a timeline
21 for their actions?
22 MS. DRUIN: The timeline that has been laid out is
23 to have a standard ready for NRC endorsement by December of
24 1998 -- in a year -- which, if we're successful, will be
25 phenomenal, because, typically, this is a four-year process.
. 25
1 They understand the criticality, and they're putting in a
2 new process to get this through in that kind of time frame.
3 CHAIRMAN JACKSON: And this is a better process
4 than doing it the reverse -- developing the standard
5 in-house and having a professional group review and endorse
6 them; is that right?
7 MS. DRUIN: I can't comment on that.
8 CHAIRMAN JACKSON: Mr. King?
9 COMMISSIONER DIAZ: But 1602 has the standards.
10 COMMISSIONER McGAFFIGAN: Implicit standards in
11 it.
12 MR. KING: Yes. And I think, from the comments
13 we've received, people feel that that's a good input to the
14 standards development effort, that it does have a good
15 foundation.
16 COMMISSIONER DIAZ: And one question was whether
17 there's going to be a significant difference between what
18 they are going to be doing, or do we have significant
19 similarities? In other words, are we on good grounds with
20 our standards or not?
21 MR. KING: I'm not sure we're far enough along to
22 say whether there's going to be a difference.
23 MS. DRUIN: Tom, we had a meeting last September,
24 and at the meeting were representatives from the different
25 owners' group, different utilities. NEI was present. One
. 26
1 of the things that was discussed was where do we start in
2 the standard?
3 Everyone was pretty much in a consensus to start
4 using 1602 work that has been developed by the CE Owners
5 Group and other organizations. There was not a lot of
6 diversity there, and no one felt that there was going to be
7 a big need to go out and create new writing, that there was
8 enough information out there, between all of these different
9 documents, that it was going to be more of a -- it don't
10 want to trivialize it -- but more of a cut-and-paste job.
11 CHAIRMAN JACKSON: Good. Thank you.
12 MR. KING: The third new activity, which is in
13 response to Direction Setting Issue 12 on risk-informed
14 regulation, is an effort by NMSS to develop a framework for
15 the application of risk-informed regulation. I understand a
16 paper will be coming to the Commission by the end of this
17 month providing that information.
18 CHAIRMAN JACKSON: Would you have any additional
19 comments?
20 MS. FEDERLINE: Yes. We would just let the
21 Commission know that we think the Commission's direction was
22 particularly timely in this regard. As you know, we had a
23 wide diversity of regulated systems in NMSS, all the way
24 from predictive 10,000-year analysis to the handling of
25 gauges.
. 27
1 So there are a couple of issues that we're going
2 to be bringing forward to the Commission that we've
3 considered. One is the diversity of the licensee base and
4 the economic motivation for turning to a risk-based
5 standard.
6 Another is the methodology questions. We're very
7 long in the development of waste disposal predictive
8 methodologies. We're less far along in the application of
9 human reliability in the medical applications, so there will
10 be a lot of development that needs to be done in the
11 methodology area. But we will be highlighting in this paper
12 these issues for you and proposing a path forward.
13 CHAIRMAN JACKSON: Good. Thank you.
14 MR. KING: The fourth items was, about a month
15 ago, we provided a paper to the Commission responding to
16 Chairman Jackson's July memorandum which followed up on an
17 ACRS letter that said we ought to consider elevating the CDF
18 to a level of a safety goal.
19 We've taken a look at that. We feel it's
20 certainly an item that might be very worthwhile in doing,
21 but in doing that there is a number of other issues that we
22 ought to look at in an integrated fashion, because the
23 safety goal policy talks about defense and depth; it talks
24 about uncertainties; a number of things that we're dealing
25 with now in trying to finalize these reg guides.
. 28
1 So we came to the Commission with a recommendation
2 to defer our answer on that until the end of March so that
3 we can deal with some of these related issues in the context
4 of finalizing the guides and then take those resolutions and
5 then come back with an integrated recommendation.
6 CHAIRMAN JACKSON: And that will be in the March
7 time frame?
8 MR. KING: End of March is what we proposed.
9 CHAIRMAN JACKSON: Right, because the current
10 safety goal policy, in fact, does not permit plant=specific
11 use.
12 MR. KING: That's correct. That's another issue.
13 CHAIRMAN JACKSON: And the Commission has recently
14 endorsed the plant-specific use.
15 MR. THADANI: Yes.
16 CHAIRMAN JACKSON: And that and this issue of
17 elevating the core damage frequency, which you're basically
18 de facto using on a plant-specific basis. You know, all of
19 these things have to be tied together, and I think you and I
20 have just said the same thing.
21 MR. KING: Yes.
22 [Laughter.]
23 And finally, we've started work now preparing for
24 a more intensive effort next year in looking at low power
25 and shutdown risk. This was prompted by an ACRS letter last
. 29
1 year. What we're doing this year is gathering existing
2 information from overseas, from industry, from other work
3 that has been done at NRC, and then, based upon that, well
4 decide what additional analysis we need to do.
5 CHAIRMAN JACKSON: Right. Now, I understand that
6 the staff, in fact, conducted fairly extensive evaluation of
7 low power and shutdown operations for both a PWR at the
8 Surry Plant and a BWR at Grand Gulf.
9 MR. KING: Correct.
10 CHAIRMAN JACKSON: And so, the question is, were
11 there inadequacies that were identified in those efforts
12 that would require us to embark on the new activity -- you
13 know, it's really related to what kind of scope of activity
14 are we planning for this risk study.
15 MR. KING: What we found out on that was two
16 things. We had done a screening study where we looked at a
17 number of plant states during shutdown and then tried to
18 identify the one that seemed to be most risky for the
19 detailed study.
20 In doing that, it became clear that the thing that
21 drives the risk is what the plant configuration is during
22 shutdown. It's not related to vendor type or containment
23 type or anything like that. It's how do they do the
24 refueling? The thing that we need to do --
25 CHAIRMAN JACKSON: In that was not a focus in the
. 30
1 earlier evaluations?
2 MR. KING: No. For the two plants we looked at,
3 we tried to identify that state, but it became evident that
4 that state does not apply to all the reactors out there.
5 CHAIRMAN JACKSON: I see.
6 MR. KING: So one of the things we need to do is
7 go back and see, are there some generic states that could
8 envelop the type of plants out there in the way they do
9 refueling and shutdown activities and try and take a look at
10 those conditions that we haven't looked at in these two
11 studies. That's one key aspect of what we need to do.
12 CHAIRMAN JACKSON: The focusing on this overall
13 plant configuration.
14 MR. KING: Yes.
15 CHAIRMAN JACKSON: Commissioner Diaz.
16 COMMISSIONER DIAZ: Yes. And, of course, when we
17 do this, we're going to set a standard for the industry.
18 We're going to do it in a very consistent, realistic,
19 thorough basis where apples are apples and oranges are
20 oranges, because it's not,
21 MR. KING: Correct.
22 COMMISSIONER DIAZ: Thank you.
23 CHAIRMAN JACKSON: And that that's the point of
24 trying to genericize some of this.
25 MR. KING: That's the point of trying to
. 31
1 genericize, but genericize in a sense that really reflects
2 the way things area done in plants. I don't know whether
3 we'll end up with six types of conditions or four types, but
4 that remains to be seen at this point.
5 CHAIRMAN JACKSON: Okay.
6 MR. THADANI: I would just make a comment,
7 Commissioner Diaz, that yes, indeed, but, in terms of
8 uncertainties, that becomes, I think, a much more
9 significant issue during shutdown of low power operation
10 because they're largely controlled by, A, human actions; B,
11 some of the models that need some work, like improvements in
12 trying to estimate these risks.
13 MR. KING: Let me go on to slide 7.
14 COMMISSIONER DIAZ: I am not going to respond
15 because it will consume the rest of the meeting.
16 [Laughter.]
17 CHAIRMAN JACKSON: Well, but before you do go on,
18 I guess I'm interested in implementing the Commission's
19 policy statement. Has the staff envisioned any
20 risk-informed applications that would require changes to the
21 backfit rule? If you haven't thought about it --
22 MR. THADANI: The way we have gone forward has
23 been that this is a voluntary option for the industry, and
24 that's how the guides are structured. They're not
25 impositions, but that if the industry wants relaxations,
. 32
1 this is an approach we would pursue. We have not thought
2 through the backfit implications if we were to impose such
3 an approach as being mandatory. Currently, we're pursuing
4 just the voluntary option.
5 CHAIRMAN JACKSON: Well, maybe you need to do some
6 background on that.
7 MR. THADANI: We'll give some thought to that
8 issue.
9 CHAIRMAN JACKSON: Right.
10 MR. KING: One thing we have thought about -- at
11 least the team working on these guides -- was the regulatory
12 analysis guidelines.
13 CHAIRMAN JACKSON: Right.
14 MR. KING: Maybe they need to be updated once we
15 get the framework and the principles and guidelines in
16 place. For example, they don't really talk about defense
17 and depth.
18 COMMISSIONER McGAFFIGAN: It's fascinating.
19 MR. KING: We have had some discussions internally
20 on that.
21 MR. THADANI: I want to make sure I didn't
22 misunderstand your question. Your question was to the heart
23 of the backfit rule, itself.
24 CHAIRMAN JACKSON: Yes.
25 MR. THADANI: To the heart of the backfit rule.
. 33
1 That's what I thought. Thank you.
2 CHAIRMAN JACKSON: You answered the question.
3 Thank you.
4 MR. KING: If I could have slide 7, please.
5 [Slide.]
6 Slide 7 talks about the guides and standard review
7 plans that are out for comment now. That comment period
8 closed a couple of weeks ago. We now have about 30 sets of
9 comments, as Ashok mentioned. Generally, the comments are
10 supportive of the concept, the approach, the structure of
11 the documents.
12 Now, they did have a lot of detailed comments on
13 specifics of the decision criteria and things that were not
14 clear to them, and I think we probably will be making a
15 number of changes, certainly, to clarify things like, we got
16 a lot of questions on the use of NUREG-1602, a lot of
17 misunderstanding, that kind of thing.
18 CHAIRMAN JACKSON: Let me ask you a question. In
19 terms of going back to the issue we talked about a little
20 bit earlier, I want to be sure I understand some of the
21 concerns with respect to required PRA quality. Was the
22 concern having to do with the fact that the required quality
23 was not adequately described, or was the concern with the
24 quality as laid out already in the documents?
25 MR. KING: The concern was that what we asking for
. 34
1 was the gold-plated PRA, that our putting forth NUREG-1602
2 was interpreted as requiring a gold-plated PRA. That was
3 not our intent.
4 CHAIRMAN JACKSON: Right.
5 MR. KING: And we need to clarify that.
6 CHAIRMAN JACKSON: And then, this issue of the
7 acceptance guidelines for very small or negligible risk
8 increases, is that an issue for those plants that may be
9 bumping up against the 10-4, as opposed to -- because it
10 strikes me that you've dealt with it, essentially, in the
11 guidance, as far as I understand it, but there is an issue
12 for those plants that either today are at or maybe even
13 slightly exceed the 10-4 core damage frequency.
14 MR. KING: It's an issue for all plants that want
15 to use risk-informed regulation, but I think it's of
16 particular importance to those that are bumping up against
17 the 10-4.
18 CHAIRMAN JACKSON: And in looking at that, have
19 you looked at issues that may have to do with -- let's call
20 it for those there for the moment, for the purposes of a
21 straw man -- looking at risk neutrality, in terms of where,
22 if there were some risk increase in one place, there might
23 be some compensation somewhere else so that that is a
24 potential way, within some band, of handling things without
25 losing something that you tag your analysis to?
. 35
1 MR. KING: Yes. In fact, that's one area we need
2 to clarify in our documents. We agree with the concept of
3 bundling several changes together -- some risk increases,
4 some risk decreases. We think that, certainly, to do that,
5 that will bring risk down in some areas, which is one of the
6 incentives for allowing bundling.
7 CHAIRMAN JACKSON: Right, but it also allows you
8 do deal with the issue in a more generic way, but
9 particularly for those that are bumping up against some
10 ceiling.
11 MR. KING: Yes.
12 MR. THADANI: Commissioner.
13 CHAIRMAN JACKSON: Go ahead.
14 MR. THADANI: I just want to make sure -- and I
15 believe that is completely consistent with the policy
16 statement.
17 CHAIRMAN JACKSON: Right.
18 MR. THADANI: Because the policy statement said
19 you need to focus on both -- the areas where we need to
20 relax our requirements and the areas it may be appropriate
21 to enhance safety.
22 CHAIRMAN JACKSON: Right. In fact, the lawyers
23 would tell you that you can't go one way without going the
24 other.
25 MR. THADANI: Yes, I've seen this.
. 36
1 COMMISSIONER McGAFFIGAN: How much does a
2 gold-plated PRA cost?
3 [Laughter.]
4 MR. KING: That's probably Ashok's $6 million.
5 COMMISSIONER McGAFFIGAN: Is the $6 million one --
6 CHAIRMAN JACKSON: I don't know that you -- can
7 you really make such a statement? Because it's a plant-
8 specific issue.
9 MR. THADANI: I think it's very difficult for us
10 to sit here and give you that estimate, because their IPE's
11 have already been done, and there is variability, as I think
12 the Chairman noted, in terms of the quality, so the cost of
13 upgrading -- we don't know, for that matter, the quality of
14 the IPEEE's.
15 COMMISSIONER McGAFFIGAN: Right.
16 MR. THADANI: And so the cost of upgrading could
17 be from a small amount to possibly quite significant. I
18 think it's very hard to give you one --
19 CHAIRMAN JACKSON: Well, is the definition of
20 gold-plated that I've done my PRA --
21 COMMISSIONER McGAFFIGAN: And it's good enough.
22 CHAIRMAN JACKSON: And it's good enough, and I
23 don't want to change it? And if you ask me to change it for
24 some application, by definition that makes it gold-plated?
25 MR. THADANI: No. I think industry's roles are
. 37
1 gold-plated, but --
2 CHAIRMAN JACKSON: No, I'm not talking about your
3 point of view. I'm playing the devil's advocate here a
4 little bit, not with you, but in the more generic sense.
5 MR. KING: I don't think changing it to reflect or
6 support an application is an issue. I think it's what's
7 that base level of quality.
8 CHAIRMAN JACKSON: It's the base level issue
9 again.
10 MR. KING: Yes.
11 COMMISSIONER DIAZ: But definitely we want it to
12 be corrosion of the system; right?
13 [Laughter.]
14 MR. KING: True.
15 COMMISSIONER McGAFFIGAN: Yes.
16 CHAIRMAN JACKSON: Gold is good in that regard.
17 Brass, on the other hand, does tarnish.
18 MR. KING: The comments we receive, we'll be
19 discussing those with ACRS next. We'll also, when we come
20 back with the proposed final documents, be summarizing them
21 for the Commission, as well, and what our response is.
22 Let me talk a little bit about the last bullet,
23 the policy issues. There are several items that have come
24 out of the comment process and from internal discussions
25 that we're going to come back to the Commission with to get
. 38
1 a policy decision. We've got a paper under preparation now.
2 We hope to have it up here by the end of this month.
3 Two of the issues are listed here -- treatment of
4 uncertainties, which we talked about earlier, and acceptance
5 guidelines for very small or negligible risk increases.
6 That gets, really, to the question of what's the
7 definition of risk-neutral. Can very small increases in
8 risk essentially, from a practical standpoint, be considered
9 risk-neutral, which would allow more flexibility for plants
10 that are bumping up against the 10-4 CDF, for example, to
11 come in and participate in risk-informed changes.
12 CHAIRMAN JACKSON: What is the complaint about the
13 treatment of uncertainties? You haven't treated them
14 anyway, so -- at this stage of the game.
15 [Laughter.]
16 So what's the complaint or the potential
17 complaint?
18 MR. KING: The complaint is that we've asked for
19 too extensive an uncertainty analysis in the current draft.
20 CHAIRMAN JACKSON: In the current draft?
21 MR. KING: Yes.
22 CHAIRMAN JACKSON: Okay.
23 MR. KING: What we're thinking about at this
24 point, and which we'll talk about in this policy paper is
25 very small changes in risk increases, very small increases,
. 39
1 how extensive an uncertainty analysis do we need? Can we
2 define a small range that we can call risk-neutral where the
3 uncertainties really don't matter, because the change is so
4 small, and, therefore, you don't need as extensive a
5 treatment of uncertainties.
6 It would also allow plants that are bumping up
7 against the 10-4 to come in and propose changes.
8 CHAIRMAN JACKSON: But isn't it rooted in the fact
9 -- and we're not going to sit here and debate it all day or
10 anything -- but isn't it related to the confidence you have
11 in the beginning? With the answer, I mean.
12 It is naive to think that you can just take a mean
13 and say, "Okay. That's it. I don't know how well I know
14 that mean, and it's okay." So it's small, and it may be,
15 though, that there's a band around it where you are that's a
16 factor of 10 higher. And so it is not an issue that you can
17 sweep under the rug.
18 MR. KING: No. And part of the problem is those
19 same concerns apply when you're talking a confidence level,
20 because even with a full-scope PRA that includes external
21 events and shutdown, there are things that are not analyzed,
22 that are unanalyzable at this point -- management and
23 organization factors, aging of components, probably some
24 others.
25 So when you're talking mean or you're talking a
. 40
1 confidence level, you still have to somehow decide how
2 you're going to treat those unquantifiable factors. So
3 we're trying to come up with a scheme that deals with that
4 and that also deals with the application that makes sense
5 from the standpoint of maybe they're proposing a change that
6 really is only affecting full power operation, and they
7 don't want to do a low power and shutdown. How do you deal
8 with that?
9 CHAIRMAN JACKSON: Okay. I understand.
10 MR. KING: It's a complicated problem.
11 CHAIRMAN JACKSON: Right. Go ahead.
12 MR. THADANI: I might note, in our paper that we
13 sent you just a few days ago, there is an attachment that
14 talks to this issue and some of the initial thoughts, at
15 least, on how we might want to consider proceeding.
16 MR. KING: Yes.
17 CHAIRMAN JACKSON: When will the Commissioners
18 receive your formal analysis of the comments and your
19 proposed resolution -- and pulling out the policy issues?
20 MR. KING: What we were doing is pulling out the
21 policy ones and sending them up the end of this month.
22 CHAIRMAN JACKSON: Okay.
23 MR. KING: That's our plan. The others, both the
24 comment and how we've treated it, we're planning to put
25 forward in the package that sends the final documents to the
. 41
1 Commission, which is December.
2 CHAIRMAN JACKSON: Right. All right. So this is
3 all tracking, still, to have the finalization by the end of
4 the year.
5 MR. KING: Yes.
6 CHAIRMAN JACKSON: Commissioner McGaffigan.
7 COMMISSIONER McGAFFIGAN: The second policy issue,
8 the acceptance guidelines for very small or negligible, does
9 this get into things like 10-6 or -- I mean have you chosen
10 a number for what is very small or negligible that you've
11 quantified and said, "Okay. If it's 10-6, I don't care
12 whether they're bumping up against 10-4, because 10-6 is
13 1/100 of 10-4, so therefore I'll consider that negligible in
14 the scheme of things"?
15 MR. KING: We need to define a number.
16 COMMISSIONER McGAFFIGAN: You need to define a
17 number? You do not have a number?
18 MR. KING: I've chosen a number, but that doesn't
19 mean that we have a consensus on it.
20 COMMISSIONER McGAFFIGAN: Is it fair to ask what
21 that number is?
22 [Laughter.]
23 CHAIRMAN JACKSON: No, because it's his number,
24 not their number.
25 MR. THADANI: Yes, right. And I want to be
. 42
1 careful. I think that's not the only variable. The other
2 variable is how many times and the cumulative impacts --
3 CHAIRMAN JACKSON: And what's the cumulative
4 impact?
5 COMMISSIONER McGAFFIGAN: Right.
6 MR. THADANI: And that's the real issue, I think,
7 that we have to pay attention to.
8 CHAIRMAN JACKSON: Exactly.
9 MR. KING: And there may be -- we've listed two
10 policy issues.
11 CHAIRMAN JACKSON: I mean risk neutrality is
12 defined within that context, anyway.
13 MR. KING: Yes.
14 MR. THADANI: Yes.
15 CHAIRMAN JACKSON: Yes.
16 MR. KING: This would define risk neutrality,
17 whatever that number is.
18 CHAIRMAN JACKSON: Exactly. That's right.
19 MR. KING: And there may be other policy issues in
20 the paper, as well. We're talking about, for example, do we
21 need separate guidelines for the shutdown condition? Do we
22 need guidelines to cover temporary increases in risk? So
23 the paper in October will deal with all of those.
24 [Slide.]
25 Let me go on to slide 8 and just quickly talk
. 43
1 about where we stand on IPE and IPEEE. The IPE reviews are
2 done except for Browns Ferry Unit 3, which will be done in
3 December, I believe.
4 We are putting together what we call the IPE
5 follow-up program, which is talking a look at a number of
6 the results from the standpoint of the plants that had a
7 relatively high core damage frequency or containment failure
8 probability.
9 Perhaps we would want to follow up with those and
10 see why haven't they done something to reduce that using the
11 guidelines in the regulatory analysis guidelines document as
12 sort of a benchmark to look at what improvements could be
13 made.
14 There are some generic issues that perhaps we want
15 to work on. You know, probably the most prominent one is
16 the pump seal LOCA for PWR's. A number of plants, that was
17 a dominant sequence. Do we want to do something further
18 there.
19 We had asked plants to specifically answer some
20 what we called containment performance improvement
21 questions, questions that came out of generic studies that
22 were done several years ago.
23 A number of plants answered those; a number of
24 plants didn't. We want to follow up on the ones that
25 didn't, and we want to follow up and see have licensees
. 44
1 really done the things that they committed to do when they
2 submitted their IPE, make the improvements they said they
3 were going to make.
4 So those are the kinds of things that we plant to
5 include in our follow-up program, and we owe you a separate
6 paper, giving more detail on that. I think it's in
7 November.
8 The IPEEE is underway. We're going to give you an
9 interim insights report in November. We've just tried to
10 put a short executive summary in the paper that you received
11 yesterday, and ultimately, when we're done, we'll have a
12 follow-up program similar to the one we're embarking on on
13 the internal events.
14 CHAIRMAN JACKSON: Let me ask this quick question.
15 Are the IPE results becoming obsolete, and do we know how
16 many licensees have been updating their IPE's? Because kind
17 of undergirding this is the issue of, if we are going to
18 make regulatory decisions based on PRA results -- which is
19 what most of these IPE's have turned out to be -- there's an
20 issue there in terms of -- and you've mentioned cumulative
21 impact, for instance, of changes. Are we thinking about how
22 that's going to be tracked and what that implies about how
23 updated the IPE's need to be?
24 You know, I've been to nuclear plants this year.
25 This is 1997. Now, admittedly, what may be in our
. 45
1 residents' offices may be different than what's on the shelf
2 in the licensees' engineering organization or PRA
3 organization, but I've seen dates like 1991. Presumably,
4 there have been changes made to the plant, including ones
5 that could have lowered the estimated core damage frequency,
6 as well as ones where there has been no analysis one way or
7 the other to know what the cumulative impact is. What are
8 we doing about that?
9 MR. KING: The IPE's, basically, are a snapshot or
10 information that's maybe five years old, something of that
11 nature.
12 CHAIRMAN JACKSON: Well, I guess I'm really asking
13 something else, which has to do with, if we're purporting to
14 make regulatory use of the PRA's -- and I want to stay on
15 that plane, but I'll ask it within the context of the IPE's
16 -- is there not a linked question having to do with living
17 PRA's, how they're maintained and how up-to-date they need
18 to be relative to -- otherwise, how do you make the
19 judgment?
20 MR. KING: Our guidance documents --
21 COMMISSIONER DIAZ: Going back to what is not even
22 a dead horse anymore -- it's a skeleton by now.
23 [Laughter.]
24 But it goes back to establishing a base --
25 CHAIRMAN JACKSON: Baseline, yes.
. 46
1 COMMISSIONER DIAZ: Okay. And they need to know
2 what the baseline is, and we need to inform them of what it
3 is so we can say this is what we will find a baseline
4 acceptable to make these changes.
5 CHAIRMAN JACKSON: Right. And included in that is
6 the issue of how do you update?
7 COMMISSIONER DIAZ: Yes.
8 CHAIRMAN JACKSON: I mean we're dealing now -- I
9 mean that's our big lesson learned in the last year and a
10 half with updating and maintaining certain fundamental
11 information vis-a-vis the licensing basis and design basis
12 information. The question is, what are we doing?
13 MR. KING: There is no requirement for them.
14 CHAIRMAN JACKSON: No, no, no. I'm not talking
15 about the requirement.
16 MR. KING: I mean there was no requirement for
17 them to do a specific type of PRA in response to an issue
18 generically in the first place.
19 CHAIRMAN JACKSON: Well, no. That is not the
20 issue. One is an informational question. One, are the
21 IPE's becoming obsolete? The second part of the
22 informational question, do we even know how many licensees
23 are updating their IPE's -- second informational question.
24 Because, three, it has bearing on the third
25 question, which is, if you're going to lay out standards
. 47
1 having to do with what the IPE's have to be for what
2 regulatory use you're going to make of them -- because
3 that's what we're talking about -- you have to include in
4 there how they're updating.
5 I'm not saying you're laying it as a requirement,
6 but you have to have something. You can't take 10-year-old
7 information and make a regulatory judgment on it. That's
8 all I'm saying.
9 MR. THADANI: I think we can answer one out of
10 three questions.
11 [Laughter.]
12 CHAIRMAN JACKSON: Well, that's good.
13 MR. THADANI: I'm pretty comfortable with the
14 second question you have raised, which is, first of all, are
15 there some licensees whose IPE's or obsolete or are there
16 licensees who are keeping up? I think the answer probably
17 is yes to both, but how many I don't think we could say.
18 Clearly, the pilot plants that we've been working with have
19 been keeping up and have been paying attention to the issue
20 of quality and so on.
21 My sense would be that if we keep moving in this
22 direction where there is, in fact, consensus within the
23 industry, more and more of the licensees' IPE's would be
24 along the track.
25 The only other comment I wanted to make was, in
. 48
1 the guides that we have, we have clearly stated that when
2 licensees are coming in for changes that the analysis,
3 itself -- that is the risk assessment -- should reflect the
4 plant as is and not as it might have been.
5 CHAIRMAN JACKSON: So, in a sense, you have that
6 covered in there.
7 MR. KING: It's in there now, yes. Yes.
8 CHAIRMAN JACKSON: That's the baseline issue. Mr.
9 Callan, you were going to make a comment.
10 MR. CALLAN: I was just going to say -- and I'm
11 sure Ashok would agree -- that we can't take comfort from
12 the fact that pilot sites are maintaining their IPE's,
13 because they were selected precisely because they maintained
14 their IPE's current.
15 CHAIRMAN JACKSON: Yes.
16 MR. CALLAN: That they are, in effect, leaders in
17 the industry in that regard. So I don't think that's a good
18 indicator of the rest of the industry.
19 CHAIRMAN JACKSON: You said something very
20 important, though. You've already said, in the guidance
21 documents that you've developed, that you already clearly
22 have stated that when one comes in with a PRA it has to
23 reflect the latest and the greatest.
24 MR. CALLAN: As built, as operated plant is what
25 it says.
. 49
1 CHAIRMAN JACKSON: Right. In order to be able to
2 use it. Commissioner McGaffigan.
3 COMMISSIONER McGAFFIGAN: My only point is that
4 the standard process -- the ASME standards process that we
5 talked about earlier that's going to start with 1602 -- if
6 you have imbedded in 1602 that a good PRA is a living PRA,
7 then that presumably is the standard we're going to be
8 propagating into standard space.
9 I think that's good, but I suspect some of the
10 stuff we're reading in Inside NRC about gold-plated PRA's
11 may well be -- a living PRA is -- I'm trying to still get a
12 definition of gold-plated.
13 [Laughter.]
14 Maybe part of it is that it's living and it has to
15 have been updated since the last time you thought about
16 this. Many of them see that large improvement as too hard,
17 as opposed to the South Texases who you've been working with
18 who, for them it's a small increment, and they are ready to
19 go. But I think it's real important, the notion of a living
20 PRA.
21 MR. THADANI: Yes. And, in fact, what Tom said
22 earlier is significant. South Texas, I think having a group
23 of five or six people, it is that the idea is not just to
24 update the PRA; the idea is to apply it.
25 COMMISSIONER McGAFFIGAN: Right.
. 50
1 CHAIRMAN JACKSON: Right.
2 MR. THADANI: And so if a licensee were to apply,
3 there is obviously the incentive to make sure it's kept up,
4 it's in fact a living PRA.
5 CHAIRMAN JACKSON: Right.
6 MR. THADANI: The issue is going to be --
7 CHAIRMAN JACKSON: Okay. You know, I've asked on
8 several occasions, but I'm now going to get you to commit to
9 this. Are we keeping record of what regulatory use we are
10 have been making of IPE results? I've asked this question
11 for -- two years. So now I'm going to get you to commit to
12 a date. Or we'll give you a date.
13 MR. KING: No. We are committed. You will find
14 it in the table in the quarterly update.
15 CHAIRMAN JACKSON: Ah, okay. Very good. We've
16 been talking about South Texas, and I notice it's on the
17 next viewgraphs.
18 MR. KING: Yes. I'm going to turn it over to
19 Scott Newberry, who will talk about the pilots.
20 MR. NEWBERRY: Yes. Good morning.
21 CHAIRMAN JACKSON: Good morning.
22 MR. NEWBERRY: I'm going to go through the status
23 of the four pilots on the next couple of viewgraphs and then
24 a little bit about insights in the inspection program.
25 Of course, each of the pilots would use a
. 51
1 risk-informed approach to all their NRC requirements and the
2 associated program at the plant in response to those
3 requirements, using a process along the lines of the Reg
4 Guides which are out for public comment, and I know there
5 are issues, and are being finalized, which creates a
6 challenge in the pilot process. But progress has been made.
7 The first pilot, on tech specs, just to remind you
8 that that pilot would extend the outage time for certain
9 equipment, ECCS equipment that would be used to respond to
10 unlikely events, large LOCA's or the safety injection
11 accumulators and low pressure injection systems.
12 The Commission -- I guess early last summer --
13 approved the Arkansas risk-informed tech spec SER. But
14 there was an issue in the SER that we're working with
15 licensees on that has to do with one of the five principles
16 in a risk-informed approach, and that's the configuration
17 risk management program that would be committed to by the
18 licensee.
19 I think the snapshot of our experience to date is
20 that that's really a plant-specific issue. In working with
21 specific licensees on that, it looks like -- San Onofrio, I
22 think, may become the lead plant there -- that we should be
23 able to finalize a position on that issue and get our first
24 safety evaluation out by the end of the year that would
25 approve a plant-specific configuration risk management
. 52
1 program that would serve as a model -- perhaps a simple
2 example here, but a model nonetheless -- that shows that we
3 can make a risk-informed decision with tech specs.
4 Graded quality assurance pilot, we just sent a
5 paper up to the Commission -- 97-222 -- which forwards the
6 draft safety evaluation report for South Texas. This SER
7 would accept Houston Power and Light's risk-informed
8 revision to their operations quality assurance program.
9 This was an interesting pilot. I think it
10 challenged the staff to use the guidance, to learn from the
11 guidance. Many of these issues that came up here were
12 talked about extensively as we moved that safety evaluation
13 report up to the Commission.
14 I think the conclusion in the safety evaluation
15 report is important to point out to you, from the standpoint
16 of including, that we think this program presented an
17 overall safety improvement at the plant.
18 But the issues of -- I think the term, "bundling,"
19 was used here -- clearly, that activity was meant to focus
20 on the most important equipment at the plant so there's
21 intense activity, but to relax the program on the less
22 important equipment at the plant.
23 But then there's an increased feedback mechanism.
24 -- that's principle number 5 in NUREG-1060 -- which we think
25 will really provide us a safety benefit.
. 53
1 I think those conclusions really come from the
2 judgment of the staff. I think there's a lot of qualitative
3 evaluation involved in this SER. So we were happy to get
4 that SER to you finally in that recent paper. Next
5 viewgraph please.
6 [Slide.]
7 IST and ISI pilot. Both pilots, of course, would
8 be intended to improve test programs, inspection programs at
9 plants using risk insights. We're still shooting for a
10 December date on the first IST pump and valve pilot at
11 Comanche Peak.
12 In terms of recent activities, we had a team on
13 site at Comanche Peak dealing with the issue of PRA quality.
14 Looking at the PRA for this particular application, there
15 were some minor issues identified that are being worked
16 through that have to do with elements of the PRA such as
17 treatment of human performance, success criteria, issues
18 like that.
19 I think I would mention that the dates have
20 delayed a little bit on this pilot and others. Utilities
21 were focusing resources on providing comments to us on the
22 Reg Guides and SRP's so that there has been an impact there.
23 But, as I said, we're still shooting for December on that
24 pilot for Comanche Peak. Palo Verde is going to slip into
25 1998.
. 54
1 Inservice inspection of piping. I want to modify,
2 I think, the context of that first bullet there, in terms of
3 "nothing received for review yet." At the time of the
4 viewgraph, nothing had been received for the identified
5 pilots to date. Those would be Surry, ANO-2, and
6 Fitzpatrick.
7 However, just within the last couple of days,
8 ANO-2 did come in with a full proposal, and we understand,
9 in talking to Surry, that they would be in next week. And
10 then, a recent letter from the industry adding two more
11 pilots, ANO-1 and Vermont Yankee, so there would be five
12 plants pursuing ISI initiatives.
13 Now that we have ANO-2 and with Surry coming in,
14 we'll be able to look at schedules and priorities and
15 provide you an update in the next plan. Next viewgraph,
16 please.
17 [Slide.]
18 We've got one viewgraph here on some of our
19 actions in response to your May SRM regarding use of PRA in
20 the inspection program. I think it's fair to say, too --
21 and, Chairman Jackson, you mentioned that -- we're really
22 talking about a philosophy here in terms of implementing the
23 policy statement.
24 So there are other broader actions in our pursuit
25 of opportunities every day, not just in the inspection
. 55
1 program, but, of course, here at headquarters, to try to use
2 risk-informed decision making in our programs, but you hear
3 it particularly in response to your SRM.
4 You asked about prioritizing inspection activities
5 using IPE information, improving the region's ability to use
6 risk insights, and then, of course, staff training. Pat
7 Baranowski will talk in some detail about training after I
8 finish up here in a minute.
9 But in terms of particular staff actions to date,
10 a lot of the information that has been talked about here in
11 terms of the IPE's, the information that we have has been
12 made available now through documents. There will be more
13 documents going out to the regions, but there have been
14 training lectures at all the regions conducted by people
15 familiar with the IPE results.
16 Then there's a continuing effort with the senior
17 reactor analysts program, keeping that program staffed.
18 CHAIRMAN JACKSON: No, do all the regions, at this
19 point, have senior reactor analysts?
20 MR. NEWBERRY: Yes. The exact status on that is,
21 of course, there's 10 SRA positions in the agency -- two at
22 headquarters and, so, two at each region.
23 CHAIRMAN JACKSON: So there's at least a body at
24 each region.
25 MR. NEWBERRY: Yes. Region 3 has two SRA's, but
. 56
1 they're not certified yet. They're in the training program.
2 CHAIRMAN JACKSON: I see.
3 MR. NEWBERRY: That's an 18-month program.
4 CHAIRMAN JACKSON: And to whom are the training
5 lectures targeted?
6 MR. NEWBERRY: As I recall, it's a broad target.
7 We look for opportunities to bring people in from the field,
8 residents as well as regional staff. So the effort here --
9 and you'll see that, I think, in the training.
10 There's training directed toward the review staff
11 and NRR in terms of what these guidance documents say.
12 There's training on PRA technology. There's training for
13 inspectors. There's training for managers. There's a broad
14 spectrum of training, and Pat's going to talk about that.
15 The intent here, and I think you'll see it in Pat's
16 viewgraphs, is to touch on, basically every NRC employee
17 making regulatory decisions here.
18 CHAIRMAN JACKSON: Okay.
19 MR. NEWBERRY: Guidance on the use of PRA in the
20 Inspection Manual. Just last month, an appendix to Manual
21 Chapter 2515 was completed. Just looking at it recently, I
22 think folks will find that very interesting. We will
23 probably get some feedback on it and maybe have to do
24 something else.
25 But there's everything from general advice to
. 57
1 advice to a glossary of terms in that Manual chapter. There
2 are examples of things that have been found and could be
3 found at plants with respect to support system
4 relationships, to front line systems, and also guidance on
5 ranking issues in planning using PRA information. And, as I
6 said, Pat will be talking about training here in a minute.
7 Upcoming actions -- the training effort is a
8 pretty significant investment of resources, both in the
9 training center facilities and also getting employees into
10 those classes and the seminars. It's an important activity.
11 We're just starting to do more in revising core
12 inspection procedures. In particular, the graded QA
13 inspection procedure is being worked on, and that's
14 explicitly an item in the plan, and that's due to be
15 completed this spring.
16 CHAIRMAN JACKSON: Good. Let me ask you a quick
17 question. You talked about the ANO risk-informed tech spec
18 changes.
19 MR. NEWBERRY: Right.
20 CHAIRMAN JACKSON: How long did that take?
21 MR. NEWBERRY: I don't know. Is there somebody
22 here?
23 CHAIRMAN JACKSON: And are we --
24 MR. NEWBERRY: We can get you an answer on that.
25 CHAIRMAN JACKSON: And even though the guidance
. 58
1 documents that you're working on are still being finalized,
2 the question is are you gathering kind of what you've
3 learned so that you kind of begin to develop a better oiled
4 process for doing them to shorten the time?
5 MR. NEWBERRY: Yes. In fact, some of the comments
6 received have been from staff working on those reviews.
7 CHAIRMAN JACKSON: Okay. And also input from
8 licensees to try to draw it all together into a process.
9 MR. NEWBERRY: Yes,. Some of the comments received
10 were derived directly from the pilots -- the tech spec
11 pilot, for example.
12 CHAIRMAN JACKSON: So if I talked to ANO and so
13 on, what would they say? Were they pleased as Punch or --
14 you know, "This is the cat's pajamas"?
15 MR. NEWBERRY: No. I will take a chance here and
16 guess at what they would say. I would say they would be
17 concerned about the risk configuration management program,
18 that we should limit ourselves to, perhaps, the extension.
19 More precisely, I think, for example, the safety
20 injection accumulators now have a one-hour AEOT; we would
21 extend it to 24 hours. They would like to limit that safety
22 assessment or that configuration issue to the latter part of
23 that extension. So they would have an issue. They would
24 have an issue.
25 CHAIRMAN JACKSON: Okay. Training. Pat.
. 59
1 [Slide.]
2 MR. BARANOWSKI: The October 14th update to the
3 PRA Implementation Plan briefing provided an attachment that
4 talked about training, and I have a few highlights here.
5 That attachment responds to the June 5th SRM in
6 which we were requested to discuss plans for training the
7 staff and, in particular, with regard to training for
8 regulatory approaches that would be relevant to the
9 risk-informed regulatory guides and the standard review
10 plans that are in development, as well as overall training
11 for basics in PRA, and, moreover, to focus somewhat on
12 regional inspection training activities.
13 The PRA Implementation Plan does include several
14 tasks related to training. These are modified as we go
15 along to reflect the regulatory program that's in
16 development as it evolves.
17 We have made some changes over the last several
18 months and are continuing to define and implement some new
19 training requirements, and I have a couple of them that I'll
20 mention here, in particular, with regard to the NRR
21 technical staff.
22 First, let me mention that there is a seminar that
23 has been put together that covers the responsibilities
24 associated with risk-informed regulatory activities, and
25 it's designed to familiarize the NRR staff in general, and
. 60
1 hopefully it will cover the kinds of things that we talked
2 about a little bit earlier -- terms and things that we can
3 all talk in a common language.
4 It's a mandatory seminar. It's meant to motivate
5 and familiarize the staff with the uses of risk-informed
6 regulatory initiatives, and it's primarily taught by an NRR
7 senior manager. I think Gary Holohan has done the most
8 recent one.
9 MR. NEWBERRY: Let me just make a comment quickly
10 there, Pat, in response to Commissioner Diaz. Your issue on
11 this mental image of risk-informed comes out very quickly in
12 the dialogue that's created in that seminar. That seminar
13 is built around regulatory policy, the PRA policy statement,
14 and the intent of where we're trying to go.
15 We've got a ways to go, based on the dialogue in
16 those seminars, but I think that's where it's beginning to
17 take place.
18 CHAIRMAN JACKSON: Well, the question becomes, in
19 terms of a metric, on the outcomes, the people who attend
20 the seminars walk away, you feel, with more clarity in that
21 regard? That's a metric.
22 MR. NEWBERRY: More clarity, yes, but my own view
23 is it's going to take continuing attention for a period of
24 time here with on-the-job attention.
25 CHAIRMAN JACKSON: Application and guidance.
. 61
1 MR. NEWBERRY: Yes, very definitely. It's only a
2 two-hour seminar.
3 CHAIRMAN JACKSON: Yes, I understand the point.
4 MR. NEWBERRY: Yes.
5 CHAIRMAN JACKSON: But you're at least opening the
6 minds in this regard.
7 MR. NEWBERRY: Yes. They're lively time periods.
8 CHAIRMAN JACKSON: Yes. Okay.
9 MR. BARANOWSKI: In addition to the seminar I just
10 mentioned, the PRA Basics for Regulatory Applications course
11 -- that's course number P-105 -- has been modified to
12 include some additional information on the regulatory
13 approaches for the Reg Guides and the SRP's.
14 That's also a mandatory course for NRR technical
15 staff, and over the next two fiscal years, we would expect
16 that the full staff should be trained, would have attended
17 that particular course. Next viewgraph, please.
18 [Slide.]
19 The resident inspectors' needs are intended to be
20 addressed by the PRA Technology for Regulatory Perspectives
21 course, P-111, which is mandatory for all full-time NRR
22 inspectors and regional reactor program inspectors.
23 Now, the course curriculum includes extensive
24 practical workshops and case studies applicable to the needs
25 of the inspectors as they would perform risk-informed
. 62
1 inspections.
2 The first presentation of the course was
3 originally scheduled for October of this year. It was
4 delayed until January of next year as a result of what we
5 call a pilot talk-through, where we sort of try the course
6 out on some more experienced people to see if the concepts
7 that were intended to be in there are coming through
8 clearly.
9 So a few modifications are being made, and then we
10 would expect to have several of these courses in Fiscal Year
11 1998 and finish up in 1999. Resident inspectors will be
12 given the highest priority for attendance in this course.
13 CHAIRMAN JACKSON: Will there be different
14 training for NMSS staff? Has anyone thought about that?
15 MS. FEDERLINE: Yes, we have. You'll see we
16 highlight it in the paper as one of our issues. We're
17 trying to decide right now what systems approach is
18 appropriate for each of our individual regulated systems,
19 and that would somewhat dictate the types of training. We
20 have had training, dedicated courses for performance
21 assessment in the waste disposal area.
22 CHAIRMAN JACKSON: Okay
23 MR. BARANOWSKI: Just to make one final point, we
24 would like to have at least one resident inspector through
25 this training by December of 1998 at each site, and so
. 63
1 that's the way that priorities will be worked out. Next
2 viewgraph.
3 [Slide.]
4 At the recommendation of the PRA training focus
5 group, a PRA Technical Managers course was developed and is
6 now being implemented. The course is required for NRR,
7 AEOD, NMSS, and regional technical managers. It's course
8 P-107. The course has been recently updated to include
9 information on the RG's and SRP's, or at least the
10 approaches that we're talking about taking in them, since
11 they're still in draft.
12 We would expect to conduct a number of course in
13 1998, which would get about two-thirds of the agency's
14 technical managers trained and then the balance in Fiscal
15 Year 1999.
16 At the same time, all offices are looking at their
17 technical training needs with regard to risk-informed Reg
18 Guides and SRP's that might result in some additional
19 training or revisions to the courses that I just mentioned.
20 We would expect to enact additional courses in the future as
21 warranted, which I mentioned earlier.
22 I would like to mention that there was a question
23 raised in the SRM regarding a regional representative on the
24 PRA training focus group. There wasn't one on the PRA
25 training focus group. The NRR representative normally
. 64
1 provided representation of inspection personnel in terms of
2 PRA training.
3 We have recently added a regional person on there,
4 but I would like to point out that when the PRA training
5 focus group meets, it's not just four people that meet.
6 There are four specific members, but folks from the
7 inspection staff have, in fact, participated in meetings in
8 the past where we've talked about resident and headquarters
9 inspection personnel training in PRA.
10 CHAIRMAN JACKSON: Right. Well, it's very
11 important, which is why the Commission asked for that,
12 because of the fact that you, for instance, talk about
13 having a resident at each site.
14 I mean you need to have a member who has equal
15 weight at the table that can represent the interests,
16 because those are our folks who are out there on a
17 day-to-day basis, interfacing with the licensees, looking at
18 how they handle configuration management, overseeing outage
19 activities, overseeing on-line maintenance, any number of
20 things.
21 Their interests and what they do -- and it is
22 unique to what they do -- need to be represented fully in
23 what you do, not just as visitors to the meetings. So it's
24 a very important issue. Yes.
25 COMMISSIONER DICUS: Do you believe that our
. 65
1 training is keeping pace with our other activities in PRA,
2 or is training lagging a little bit behind it?
3 MR. BARANOWSKI: That's a good question.
4 COMMISSIONER DICUS: Or do you feel that it
5 perhaps is, and that's a concern?
6 MR. BARANOWSKI: I guess my personal feeling is
7 that you need to have an understanding of what your
8 regulatory program is before you can train people to execute
9 that program. I wouldn't want to use training as an
10 approach for developing the regulatory program. I've seen
11 some of that tried in some of our meetings, and I don't
12 think it works very well.
13 So it may be we're lagging a little bit, but I
14 think we're addressing things pretty rapidly.
15 CHAIRMAN JACKSON: Once you get the pieces in
16 place.
17 MR. BARANOWSKI: There's a very big push to get
18 this training going.
19 CHAIRMAN JACKSON: Commissioner McGaffigan.
20 COMMISSIONER McGAFFIGAN: How long do these
21 courses last? I mean I'm just trying to get a sense of how
22 much one might possibly take away from it.
23 MR. BARANOWSKI: A typical course is running four
24 or five days. Now, there are some plans to have some
25 two-week courses and things like that. I don't remember the
. 66
1 exact length of them.
2 COMMISSIONER McGAFFIGAN: The ones that you have
3 here -- the 107 and 111, et cetera -- they're all four- or
4 five-day?
5 MR. BARANOWSKI: I believe they're four or five
6 days.
7 MR. THADANI: I see three and four days.
8 MR. CALLAN: Three and four days. In classroom
9 hours, something on the order of 18 to 24.
10 MR. BARANOWSKI: We're looking for about 25, I
11 think it was.
12 MR. CALLAN: Twenty-five?
13 COMMISSIONER McGAFFIGAN: I suspect we're going to
14 have to have refresher course.
15 MR. CALLAN: Right. Our concern right now, just
16 to put that in perspective, I think NRR has asked for 400
17 slots of a four-day course, and we're trying to find the
18 money.
19 CHAIRMAN JACKSON: Right. Plus you want to try to
20 get a certain baseline.
21 MR. CALLAN: We're trying to get a baseline, so --
22 I think I know where you're going with the question. It's a
23 survey course -- we understand that -- but it's to
24 accomplish some of the objectives that Commissioner Diaz
25 mentioned and others, to get a baseline. But we are
. 67
1 concerned about the resource expenditure it's going to take
2 to do this training, and we haven't solved that problem yet.
3 CHAIRMAN JACKSON: You wanted to make a comment?
4 Could you identify yourself?
5 MR. COE: Yes. I'm Doug Coe with NRI. I just
6 wanted to correct one thing. P-11, the course that's
7 specifically designed for inspectors, will run the better
8 part of two weeks.
9 CHAIRMAN JACKSON: Thank you.
10 MR. BARANOWSKI: And these courses are also in
11 addition to our other more exacting curriculum, which a
12 lesser number of people are taking.
13 CHAIRMAN JACKSON: Right. Let me just ask a
14 question that is somewhat disconnected. How are we creeping
15 up on -- or are we -- performance-based regulation? What
16 are we doing in that regard?
17 MR. THADANI: There are two parts to that effort.
18 Part one is, as we have indicated before, we're trying to
19 utilize some of the thinking that went behind the
20 maintenance rule. Where areas are amenable to risk
21 analysis, we're folding in the performance-based aspects as
22 feedback for any follow-on actions. So that's embedded in
23 what we're doing in terms of areas which are amenable to
24 risk analysis.
25 As far as the other areas, the Commission directed
. 68
1 the staff in an SRM to also look at how performance-based
2 approaches could be used in areas which are not amenable to
3 risk analysis.
4 We have a paper due to the Commission, I believe
5 in two weeks, in which we're going to address what is it
6 that we're doing to respond to that Commission direction.
7 Basically, we're going to tie it to integrate with the
8 SI-13, which is the role of industry, and we are planning to
9 have some workshops. But those are actions that are going
10 to come.
11 CHAIRMAN JACKSON: So the more formalized one has
12 to do with this response to the SRM --
13 MR. THADANI: Yes.
14 CHAIRMAN JACKSON: -- and working with industry,
15 and the other is on the more ad hoc basis of utilizing
16 approaches a la the maintenance rule when it seems
17 appropriate.
18 MR. THADANI: That's right. That's right.
19 CHAIRMAN JACKSON: Yes, Commissioner.
20 COMMISSIONER DIAZ: As a parallel, how are we
21 doing with the maintenance rule assessments? I guess we've
22 got now -- what? -- 36?
23 MR. THADANI: Thirty-six, I believe we've
24 completed.
25 COMMISSIONER DIAZ: Thirty-six? Can you tell us,
. 69
1 you're happy with the way they're looking at -- the
2 responses?
3 MR. THADANI: Let me ask Scott to give you, up to
4 date, the status.
5 MR. NEWBERRY: I thought we might get that
6 question, so I even conducted a poll of happiness. I think
7 it depends on the expectation. If you were to ask some,
8 they believe there's still a significant way to go. But I
9 think, in the larger picture, we've come a long way in terms
10 of where we were before the maintenance rule, before the use
11 of -- you know, very little use of risk information.
12 Now, you see all plants the use of risk
13 information in terms of qualitative or even very good models
14 and work stations and the like. So I think overall, in the
15 broad context, we've been reasonably pleased with the
16 progress, but, of course, there have been some issues
17 identified.
18 COMMISSIONER DIAZ: And so those 36 plants that we
19 have now -- you know, have assessments completed already --
20 when you say performance-based, they know what it is.
21 MR. NEWBERRY: I can't agree with you. I don't
22 know. I don't know how they would answer that particular
23 question.
24 COMMISSIONER DIAZ: I think Mr. Callan was wanting
25 to speak.
. 70
1 MR. CALLAN: The 36 plants, Commissioner, are the
2 plant that have received the baseline programmatic
3 inspection.
4 COMMISSIONER DIAZ: Right.
5 MR. CALLAN: Which, in my view, is not a real good
6 measure or gauge of the maintenance rule, per se. The
7 maintenance rule is not intended to be a programmatic type
8 of rule. And so, as we gain experience with implementing
9 the performance-based aspect of it, then we'll be in a
10 position to, I think, answer your question. We really don't
11 have that much experience to date in the performance-based
12 aspect of the maintenance rule.
13 CHAIRMAN JACKSON: So what you're saying is the
14 baseline inspections have been looking at what licensees
15 have put into place --
16 MR. CALLAN: Right. Yes.
17 CHAIRMAN JACKSON: -- to begin to fully implement
18 the rule. It's not until you have the chance to begin to
19 inspect against their program that you can really address
20 the question.
21 MR. CALLAN: That's right.
22 CHAIRMAN JACKSON: Is that a fair statement?
23 MR. CALLAN: That's right.
24 MR. THADANI: I think we can say one other thing,
25 and that is, when one gets finished with maintenance
. 71
1 inspections, you do end up with some understanding of what
2 are those so-called performance measures the industry is
3 going to use, so that that level of confidence is attained.
4 But the actual experience, as Joe correctly noted, we really
5 haven't had.
6 COMMISSIONER DIAZ: But going back to the dead
7 horse or the skeleton, we are needing to define, and the
8 industry understands and we communicate on a one-to-one
9 basis of what it is. I mean it's the expectation that we
10 need to be clear on, even if we're delaying the actual
11 looking at the performance.
12 MR. THADANI: And that was one of the goals behind
13 these programmatic inspections.
14 COMMISSIONER DIAZ: Right.
15 MR. THADANI: It was to make sure that expectation
16 was going to be realized.
17 COMMISSIONER DIAZ: And that's why I asked the
18 question. You know, from these inspections, are we getting
19 a sense that people are moving in the direction of really
20 getting performance measures that can be then "regulated
21 according to the maintenance rule," which is risk-informed
22 performance?
23 MR. THADANI: Yes.
24 COMMISSIONER DIAZ: Okay.
25 MR. BARANOWSKI: Okay. Number 15.
. 72
1 [Slide.]
2 Let me briefly mention a few additional things on
3 the risk-based analysis of operating experience. Of course
4 you're aware that we have an agreement now and an SRM of
5 June 13th giving the staff the go-ahead to work on the
6 voluntary approach for obtaining reliability and
7 availability data with the nuclear industry.
8 We recently signed a memorandum of understanding
9 -- or a modification to the memorandum of understanding --
10 which addresses our obtaining that data that we expect to
11 get from industry through INPO, and we should begin
12 receiving that information sometime in Fiscal Year 1998.
13 CHAIRMAN JACKSON: You may recall that, in fact,
14 the staff agreed to characterize for the Commission the
15 scope of the voluntary data arrangement --
16 MR. BARANOWSKI: Yes.
17 CHAIRMAN JACKSON: -- as opposed to the scope of
18 the maintenance rule.
19 MR. BARANOWSKI: Right.
20 CHAIRMAN JACKSON: Now, when will we get that
21 comparative?
22 MR. BARANOWSKI: I would hope you have a
23 memorandum already in your in box.
24 COMMISSIONER McGAFFIGAN: It arrived in our boxes
25 as we were walking downstairs.
. 73
1 CHAIRMAN JACKSON: Oh, you did it so you would get
2 us.
3 [Laughter.]
4 MR. BARANOWSKI: It's that kind of trick we pull
5 on you once in a while.
6 COMMISSIONER McGAFFIGAN: It was brought to me as
7 I was walking downstairs.
8 CHAIRMAN JACKSON: Oh, okay.
9 MR. BARANOWSKI: Sorry. We wanted to get it to
10 you a few days early, but we couldn't quite do it.
11 CHAIRMAN JACKSON: Touche.
12 MR. BARANOWSKI: But just in a quick nutshell, the
13 scope of the maintenance rule's systems and components --
14 not structures -- overlaps quite well with the scope of the
15 equipment in the voluntary approach. They're essentially
16 the same. I can't say they're exactly the same.
17 Now, the level of information that's provided on
18 each component or system varies depending on it's perceived
19 risk significance. We tried to put together a more thorough
20 discussion in the paper, and we would be glad to meet either
21 individually or under other circumstances to go over that if
22 necessary.
23 [Laughter.]
24 CHAIRMAN JACKSON: That's very good. Thank you.
25 MR. BARANOWSKI: Let me also mention a few more
. 74
1 things. The Accident Sequence Precursor program, we have a
2 paper due to the Commission in November, and so I won't
3 cover too much of that since we're running a little bit
4 late.
5 I will just mention that we've finished the '95
6 work and published that. The '96 precursors have all been
7 identified, and they're going through final QA. Most of
8 them have been finalized and released through the PDR to the
9 public, and we're even into doing some 1997 analyses.
10 In general, what we're finding is about 10 to 15
11 precursors identified per year, and the conditional core
12 damage probabilities run up to a maximum about 10-3, and the
13 10-3 we see about once every other year. But we're going to
14 cover this a little bit in our paper, and so I think we'll
15 do a more thorough job in November, if that's okay.
16 CHAIRMAN JACKSON: Sure. And also, there's a
17 question of where are you with respect to developing
18 risk-based performance indicators? Are you going to speak
19 to that in November?
20 MR. BARANOWSKI: I might just mention that the
21 next couple of things that I have listed here, some of the
22 studies that we've done -- reactor core isolation cooling, a
23 special study on fire events, BWR core spray system, and a
24 number of others that we have on progress on auxiliary
25 feedwater systems, reactor protection, initiating events,
. 75
1 and loss of off-site power -- are all part of what I would
2 call the ground work for preparing more risk-based
3 performance indicators.
4 I think, after we get a few more of these studies
5 done, we would be wanting to come back to the Commission
6 with some sort of a conceptual idea and see if we're all on
7 the same page on this.
8 CHAIRMAN JACKSON: Okay. Very good.
9 MR. BARANOWSKI: The last thing I'll say is that
10 we're now about to issue our CCF database. It's in a CD-ROM
11 format. It will be issued to the nuclear industry. It does
12 contain proprietary information, and we have worked with
13 INPO to make sure that that can be released to U.S. nuclear
14 power plant operators.
15 Now, I'll turn it back to Tom.
16 MR. KING: All right. Let me try and summarize
17 with the last viewgraph.
18 [Slide.]
19 We've got a lot of things underway or that have
20 been completed over the past six months, but we still have a
21 long way to go. What we've tried to list on 16 were the
22 major things coming up over the next three months or so.
23 As we mentioned, we've got the framework paper
24 from NMSS, which is due in a few weeks. We've got an
25 intense activity to complete the Reg Guides and SRP's.
. 76
1 That is going to involve a policy paper to the
2 Commission later this month. It's going to involve some
3 meetings with ACRS -- one next and nne in mid-November -- to
4 go through the comments, the policy issues, and the
5 positions as to how we want to deal with the public comments
6 and finalize the guides.
7 MR. THADANI: Tom, on that, this is a very
8 important point. Tom indicated that a number of significant
9 issues have been raised, and we're going to be meeting with
10 the advisory committee.
11 I think, in view of the significance of some of
12 these issues, we do need to take a week or two to really
13 think through these issues carefully, and so I'm really
14 putting a hedge on whether we can get the paper to the
15 Commission by the end of October. That's a big question
16 mark in my mind.
17 CHAIRMAN JACKSON: Well, you work toward that.
18 MR. THADANI: We are going to work towards this.
19 CHAIRMAN JACKSON: Right? Because, you know --
20 MR. THADANI: Yes.
21 CHAIRMAN JACKSON: -- if you let it slip, things
22 tend to slip forever.
23 MR. THADANI: We're going to work towards it, but
24 I just want to acknowledge that there are some tough issues
25 that we better take a little time to think through.
. 77
1 MR. KING: In addition, we've got the pilot
2 activities that we're trying to complete by December, at
3 least in the tech specs and inservice testing area. We've
4 got the ISI package, which is out for comment now, a
5 workshop in November in pilot activities that are just
6 getting underway there.
7 We have the initiative from NEI, where they're
8 going to take a full-scope PRA and compare it against
9 regulatory requirements and operations and maintenance
10 costs, which we hope to finalized and kick off in December.
11 So there's a number of things in the mill. Some of the
12 schedules, as Ashok said, are ambitious, but that's what
13 we're working toward. With that, I conclude.
14 CHAIRMAN JACKSON: Commissioner McGaffigan.
15 COMMISSIONER McGAFFIGAN: I have one question. I
16 apologize. It really goes back to Mr. Newberry's
17 presentation. You said that San Onofrio was out front in
18 terms of the configuration risk management program and that
19 you hoped to have something done by December.
20 Now, do they have a living PRA? And do they
21 calculate, when they take something out, what the
22 conditional core damage frequency is? And do they then
23 adjust for it, take it into account and say, well, that this
24 is too high a risk? And did you even get to the point of
25 discussing what the threshold is for when that might be too
. 78
1 high? How did all that work in that discussion?
2 MR. NEWBERRY: In general, they have a very
3 advanced, or a significant program at that plant, such that
4 they would be able to essentially do what you've just
5 suggested there. That's right.
6 And I only wanted to mention that Arkansas was the
7 lead and, in fact, you got the SER, but as that issue
8 unfolded, right now it appears that SONGS is ready to move
9 out on that amendment. That was really the only thing that
10 I wanted to mention.
11 COMMISSIONER McGAFFIGAN: I'm just trying -- is
12 the commitment that they're going to be making something
13 that's going to be captured in a license condition or
14 something?
15 MR. NEWBERRY: A technical specification.
16 COMMISSIONER McGAFFIGAN: In a technical
17 specification.
18 MR. NEWBERRY: In the administrative section, I
19 believe.
20 COMMISSIONER McGAFFIGAN: And can you tell me what
21 number -- is there a number, like, if the conditional core
22 damage frequency approaches some number, then we will think
23 twice about whether we allow the configuration?
24 MR. NEWBERRY: At this point, I don't think there
25 would be a number.
. 79
1 COMMISSIONER McGAFFIGAN: So it's a qualitative
2 judgment we're leaving to the licensee.
3 MR. NEWBERRY: Yes, with -- I think there's five
4 or six elements -- just like principles -- that would go
5 into the tech spec considerations.
6 MR. THADANI: I might add to what Scott is saying
7 -- and we're going to be addressing this as one of the
8 issues -- while we have talked about core damage frequency
9 of 10-4, it's an average estimate over a period of one year.
10 COMMISSIONER McGAFFIGAN: Right.
11 MR. THADANI: And the next issue is what kind of
12 instantaneous risk or dynamic aspect of it would one want to
13 consider. That issue we're going to address amongst the --
14 I think that's a policy matter, as well. It needs to be --
15 COMMISSIONER McGAFFIGAN: The reason I'm asking
16 the question -- and I apologize; it's late -- is we're
17 considering the should-to-shall issue in A-3 of the
18 maintenance rule at the moment.
19 CHAIRMAN JACKSON: Right.
20 COMMISSIONER McGAFFIGAN: And to some extent, what
21 you're going through in this negotiation with SONGS or in
22 the other combustion engineering plants is a precursor to
23 what happens when "should" gets changed to "shall" and what
24 do we mean by that?
25 CHAIRMAN JACKSON: Right.
. 80
1 COMMISSIONER McGAFFIGAN: So I may just ask the
2 question --
3 CHAIRMAN JACKSON: Well, once "should" is changed
4 to "shall," then they'll be forced to address what, in fact,
5 that means on the ground.
6 COMMISSIONER McGAFFIGAN: Right. But I may just
7 ask separately -- or maybe our TA's may get briefed in more
8 detail about --
9 CHAIRMAN JACKSON: I think that would be good to
10 do a TA briefing on that.
11 MR. NEWBERRY: I'll take the action item to look
12 at that.
13 CHAIRMAN JACKSON: Right.
14 MR. NEWBERRY: And we'll put something together.
15 CHAIRMAN JACKSON: I'm just going to go ahead in
16 the reverse order. Do you have any other comments or
17 questions, Commissioner? And then I'm going to go to
18 Commissioner Dicus.
19 COMMISSIONER DIAZ: Okay. I think I wrote
20 something in here that I think is -- it's going back to
21 philosophy. But having looked at these things for some time
22 and looking at gold-plated or hot-dipped galvanized, we need
23 to remember that, you know, in this case, we've been for
24 some time striving to get to a level of achievement.
25 And in that case I am of the opinion that the
. 81
1 better is the enemy of the good and that the best is the
2 enemy of the better, that we need to define what is it that
3 we can do and do it, rather than keep trying to make it
4 gold-plated or otherwise, and I think it is an important
5 step.
6 I would like to ask the staff the next time that
7 we come back -- and these are questions that are not as
8 simple as they sound -- but the first question is, are we
9 convinced -- we, the NRC -- that we are going to be a
10 risk-informed agency? And if that is so, have we permeated
11 the structure so everybody knows that that is a fact?
12 Second is, have we convinced licensees,
13 stakeholder, or anybody that we are going to be a
14 risk-informed agency? Because if we are and we haven't done
15 that, then we have a job to do. Even it it's information,
16 whatever, we need to be doing.
17 And if the answers to these two things are yes --
18 and I do hope they are -- then we go back to the dead horse.
19 It's a matter of defining how good is good and where the
20 process needs to lie. I think we need to move to make
21 something happen rather than keep waiting for further
22 definition. But in that sense, what we make happen has to
23 be enforceable in regulatory space.
24 CHAIRMAN JACKSON: Absolutely.
25 COMMISSIONER DIAZ: Thank you.
. 82
1 CHAIRMAN JACKSON: Absolutely. Commissioner
2 Dicus.
3 [No response.]
4 I would like to thank the staff for a very
5 informative and, I'll actually say, enjoyable briefing on
6 the agency's PRA activities. Thank you.
7 We commend you for the progress that you've made
8 to date in what is a sometimes difficult area, but at the
9 same time, we encourage you to continue to improve the
10 process -- and we've heard various comments to that effect
11 -- and to provide appropriate -- for yourselves -- review
12 mechanisms and feedback mechanisms to ensure that the PRA is
13 appropriately understood in a risk-informed framework and
14 appropriately used to make your own efforts
15 performance-based in that sense, to have that feedback so
16 that you focus on outcomes.
17 But I think you've made some long steps forward
18 from where we were two years ago, even though the policy
19 statement was there and there was a PRA Implementation Plan,
20 there's a lot more flesh on the bones, and I think you can
21 be proud of that.
22 Clearly, PRA has become an important tool in
23 support of the regulatory process, a risk-informed process.
24 And so we have to strive to enhance the process where
25 necessary, but always to ensure its consistent use where
. 83
1 appropriate.
2 Unless there are any further comments, we're
3 adjourned.
4 [Whereupon, at 11:45 a.m., the briefing was
5 concluded.]
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Home > Electronic Reading Room > Document Collections > Commission Documents > Commission Meetings, Slides, Transcripts, Meeting SRMs, and Full Written Explanation for Closed Meetings > 1997
>
10/15/1997