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
Privacy Policy |
Site Disclaimer |