ML20134D484
| ML20134D484 | |
| Person / Time | |
|---|---|
| Issue date: | 10/16/1996 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| References | |
| REF-10CFR9.7 NUDOCS 9610220351 | |
| Download: ML20134D484 (100) | |
Text
.. .- . . ._
a ., O OmGLNA_
t UNITED STATES OF AMERICA i
NUCLEAR REGULATORY COMMISSION l
Title:
BRIEFING ON PRA IMPLEMENTATION PLAN -
PUBLIC MEETING l 4
2 i
l Location: Rockville, Maryland l
.l
- Date
- Wednesday, October 16,1996 i
f 1
1
., Pages: 1 - 65 9610220351 961016 PDR 10CFR PT9.7 PDR ANN RILEY & ASSOCIATES, LTD. /
3 1250 I St., N.W., Suite 300 Washington, D.C. 20005 L I
[Or/' 1 (202) 842-0034 0 f
- e a i
}
i i
j DISCLAIMER This is an unofficial transcript of a meeting of the United States Nuclear Regulatory Commission held on October 16, 1996 in the Commission's office at one White Flint North, Rockville, Maryland. The meeting was open to public attendance and observation. This transcript has not been reviewed, corrected or edited, and it may contain inaccuracies.
The transcript is intended solely for general informational purposes. As provided by 10 CFR 9.103, it is not part of the formal or informal record of decision of the matters discussed. Expressions of' opinion in this transcript do not necessarily reflect final determination or beliefs. No pleading or other paper may be filed with the Commission in any proceeding as the result of, or addressed to, any statement or argument contained herein, except as the Commission may authorize.
\.
i e 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 11555 Rockville Pike 11 Rockville, Maryland 12 13 Wednesday, October 16, 1996 14 15 The Commission met in open session, pursuant to 16 notice, at 2:07 p.m., the Honorable SHIRLEY A. JACKSON, 17 Chairman of the Commission, presiding.
18 19 COMMISSIONERS PRESENT:
20 SHIRLEY A. JACKSON, Chairman of the Commission 21 KENNETH C. ROGERS, Member of the Comnicsion 22 GRETA J. DICUS, Member of the Commission 23 NILS J. DIAZ, Member of the commicsion 24 EDWARD McGAFFIGAN, JR., Member of the Commission 25 ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
_)
i e l
l
! 2 1 STAFF AND PRESENTERS SEATED AT THE COMMISSION TABLE:
2 JOHN HOYLE, SECRETARY 3 KAREN CYR, GENERAL COUNSEL 4 JAMES TAYLOR, EDO 5 EDWARD JORDAN, Director, AEOD 6 CARL PAPERIELLO, Director, NMSS 7 NORMAN EISENBERG, Senior Advisor, Performance 8 Assessment, NMSS 9 ASHOK THADANI, Associate Director for Inspection 10 and Technical Assessment, NRR 11 GARY HOLAHAN, Director, Division of Systems Safety 12 and Analysis, NRR 13 THOMAS KING, Deputy Director, Division of Systems 14 Technology, RES 15 JOSEPH MURPHY, Special Assistant, RES l l
16 l
\
l I
17 18 19 20 21 22 23 24 l l
25 ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
e 3
1 PROCEEDINGS 2 [2:07 p.m.)
i 3 CRAIRMAN JACKSON: Good afternoon. I am pleased 4 to welcome members of the Staff to brief the Commission on 5 the status of the PRA Implementation Plan.
6 The PRA Implementation Plan was first issued in 7 August 1994, and the Staff provides quarterly written 8 updates and briefs to the Commission semiannually.
9 Previous written updates on the status of 10 activities in the PRA Implementation Plan were provided to l
11 the Commission in March and June of this year. The t
( 12 Commission was last briefed on the plan in April of this l 13 year. The plan is intended to be a management tool that l
l 14 will help ensure the timely and integrated agency-wide use 15 of PRA methods and technology in the Agency's regulatory 16 activities.
17 During today's briefing, the Staff will cover its l
18 recent accomplishments, policy issue recommendations, key i 19 technical and process issues, and its plan for future 20 activities. I am particularly interested in hearing about I
l 21 progress on the PRA regulatory guides and standard review 22 plans, as well as how these activities are being informed by 23 pilot applications. I am also interested in cross-office 24 integration, and my fellow Commissioners and I are looking I 25 forward to your briefing today.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 J
6 n 4
1 I understand that there are copies of the 2 viewgraphs available at the entrances to the room.
3 If no one has any additional comments, Mr. Taylor, 4 please proceed.
5 MR. TAYLOR: Good afternoon. With me at the table 6 from several offices is Norm Eisenberg, Carl Paperiello from 7 NMSS, Ed Jordan of AEOD, Ashok Thadani and Gary Holahan from l 8 NRR, Tom King, and Joe Murphy from the Office of Research. l I
9 I think this represents a good cross-section of people who 10 are working on this particular area. I 11 I would like to preface the presentation, and I 12 think Ashok will bring this up again, that in order for 13 licensees to use PRA and regulatory applications, the design 14 basis and configuration management issues at their plants 15 must be resolved. In other words, the plant design bases 16 must be clearly known and maintained. The plant must have l 17 been constructed in accordance with the design basis, and 18 the plant must be configured and operated in accordance with 19 our NRC requirements and license commitments.
20 With those opening thoughts, I will ask Ashok 21 Thadani to continue.
22 MR. THADANI: Thank you, J '.m .
23 May I have viewgraph Nc. 1, please?
24 Good afternoon. We thought it was probably useful 25 to go 'hrough fairly quickly some of the background ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
a e 5
1 information, and I will do that, hopefully, quickly.
2 Then Tom King from the Office of Research will 3 pick up on the recent accomplishments, as well as discussing 4 some of the key technical and process issues and the types 5 of questions that we need to make sure we can address.
6 Part of these key technical issues are also some l 7 policy matters, and then Gary is going to go over the four l 8 key policy issues, as well as the next activities that we .
9 are going to embark on in the next few months.
10 May I have the next viewgraph, please?
11 I know most of you know all of this information, 12 but again, che final policy statement was published over a 13 year ago, and following the policy statement, it was clear 14 that we needed to have a more detailed set of task schedules 15 that needed to be laid out, and with the special focus, as 16 you indicated Chairman Jackson, it was to accelerate 17 development of regulatory guides and standard review plans 18 as part of the activities.
19 The Staff has been providing quarterly progress 20 reports to the Commission and a semiannual briefing on the 21 progress that we have made as we go forward.
22 In the March '96 status report, we identified four 23 policy issues, and in the SRM that came out in May of 1996, 24 the Commission asked the Staff to provide its 25 recommendations to the Commission on each of the policy ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 i
i i 6
1 issues.
2 Last week, we had the last update report, and that 3 does get into the issue of where we stand on a number of 4 activities in the Implementation Plan, and there is a 5 special discussion of each of the policy issues.
t i 6 In addition to that, there is an attachment in the 7 last update which does identify the type of technical
, 8 process issues that we need to make sure to address.
9 May I have the next viewgraph, please?
- 10 We keep having to remind ourselves that the policy 11 statement has certain constraints and boundary conditions 12 that we have to keep in mind as we go forward, and it is 13 hard to capture everything on one chart, but I think this 3 14 chart does capture some of the important aspects of what is 15 in the policy statement.
16 The desire, clearly, is to use probabilistic 17 safety assessments or risk assessments, use them in all of 18 our regulatory activities, and the key there was all, but j 19 then there are qualifiers, obviously, with those 20 applications.
21 Certainly, as long as the methods are appropriate, 22 the database is there to support decisions in those areas, 23 and two important elements are that the decisions were not 24 to be based on risk analysis alone; that they had to 25 complement the traditional deterministic considerations as ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
s n.
7 1 well. The idea, then, was to integrate the deterministic 2 and probabilistic considerations before making any final 3 decisions on regulatory matters.
4 An issue that had to be dealt with was the issue 5 of uncertainties, not just the uncertainties that one can 6 quantify and develop distributions, but there are issues 7 that are very difficult to quantify as a matter of fact.
9 So specific focus had to be given to maintaining 9 defense in depth; that is, preserving the barriers that are 10 there, multiple barriers.
11 The next element was very important to make sure 12 that when the policy statement went out that the industry 13 did not misunderstand the statement itself; that licensees 14 had to meet all current rules and regulations, even if there 15 were rules and regulations which may have low safety 16 significance, but the idea there was that if there are 17 insights from risk assessments that point out that some 18 other requirements may not be properly in tune in terms of 19 risk significance, that the process would be to first change 20 the requirements and not presume that they didn't have to 21 meet the requirements.
22 The second part, and this is the one that Mr.
23 Taylor mentioned, is that the risk assessments are not very 24 useful if they don't really represent the plant itself. If 25 the documents don't reflect the plant design and the risk ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
e i 8
1 assessment is based on those documents, then, clearly, the i 2 risk assessment doesn't really represent the plant. To what 3 degree depends on the differences between the actual plant
- 4 configuration design procedures versus what is in the 5 documents. So that is an important element that needs to be 6 recognized, and it is clearly a lesson that we have learned 7 from some of the millstone activities.
8 The third bullet there refers to in June of 1990, l 9 the Commission gave us guidance through an SRM, indicating 10 that we ought not to be using the Commission's safety goals 11 and the subsidiary objectives, which relate to core damage, i
12 frequency, and containment performance; that we ought not to l 13 be using these on a plant-specific basis, but they should be 14 used in generic matters for things like future rulemaking i 15 activities and so on.
16 One of the policy issues that you will hear about 17 later on is, in fact, should we use these objectives on a i l 18 plant-specific basis, but we will come back to that issue 4
19 when Gary gets into those policy issues. )
- 20 The next viewgraph, please.
21 This really highlights that the PRA can make a 22 pretty significant role in regulatory activities, and this 23 chart is really representing a reactive program in a very 24 broad scope manner.
25 As our resources go down, there are budgetary ANN RILEY & ASSOCIATES, LTD. I Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
4 4 9
1 constraints. With time, it becomes even more and more 2 important to focus our activities in areas that are more 3 important to safety. So the idea here is to show that the 4 acope can be pretty broad in terms of where these techniques 5 can be applied.
6 The reason for that is if one were to use these 7 techniques in conjunction with our deterministic 8 assessments, the end results are going to be much better 9 decisions, much more effective safety decisions. There 10 would be obviously much more effective use of resources, li both in terms of the Agency resources, but also in terms of 12 the industry resources. So, again, this chart supports the 13 major thrust of the policy statement that we should, in 14 fact, go forward and apply these approaches in our 1 15 regulatory activities.
16 CHAIRMAN JACKSON: Before you go ahead, can you i 17 give us some sense of the status of licensees' 18 implementation of accident management strategies?
19 MR. THADANI: Yes. I will give you just a general I 20 sense.
21 Many of the licensees, as you know, most of them 22 have completed individual plant examination. As a result of 23 the IPEs, they had identified a number of procedural 24 enhancements that could be made, and by and large, the ones 25 that they identified, they have gone forward, but the broad ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 1
l 1
10 1 scope accident management program that we have been working 2 on with the industry for some time, which really goes into 3 not only in terms of prevention of accidents, but also 4 following core damage events, what are reasonable things to 5 do , and all the way into communication with different 4 6 groups. That is broadly included under what we call the 7 accident management program.
8 The industry through the owners group, they have 9 done essentially all of the technical work. There are minor 10 issues that need to be dealt with on the BWR plants, but by 11 and large, much of the technical work is completed, and the j~
12 utilities now are going to be converted that information, 13 which is generic, a fair amount of good technical 1
14 assessment, converting them into their plant-specific either l 15 emergency operating procedures or guidance for technical 16 support groups, which would be called upon to provide 1 17 guidance in case of an accident.
18 The schedule currently calls for all the licensees 1
19 implementing accident management by December of 1988. Some 20 of the licensees would have implemented accident management ,
1 l
21 early '98, and very few late '97. Most of them in 1998, and 22 the last ones, end of 1998, would have implemented accident 23 management.
24 Could I have the next viewgraph, please?
25 As I have said, the policy statement is to I ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
e e 11 1 incorporate all activities, which meant that it was very 2 important to capture these activities in detail, and that is 3 where the Implementation Plan comes in.
4 In the plan, which is very comprehensive and broad 5 scope, there are a number of tasks. I forget the exact 6 number, but certainly over a hundred distinct activities are 7 involved.
8 From those tasks, some of which we have not met 9 the schedule or we think we won't be able to meet the 10 schedule, there will be some delays, but the highest 11 priority we have given is to regulatory guides and the 12 standard review plans. There, the schedule was to get 13 drafts completed by the end of this year, and that is, in 14 fact, the schedule we're still on.
15 You will hear about where we are in terms of the 16 pilot applications. There have been some delays in the 17 projected completion dates for pilot plans.
18 One reason for delays is resources, but I think 19 that is a smaller reason. The larger reason has been trying 20 to do a fairly thorough job, which means a fair amount of 21 information that is needed from individual pilot 22 participants, and in some cases, it has taken longer to get 23 answers to some questions, but nevertheless, the key point I
24 is that we are getting sufficient information from these 25 pilots, so that we can, in fact, go ahead, get the reg guide ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 4 l
]
- _ . . .- ~ . . .. .. . - .- .-_ . . ~_
0 b 12 1 and the standard review guide out for public comment, and 2 finish up the pilots perhaps even during that comment 3 period.
4 The scope of the Implementation Plan goes well 5 beyond what NRR does, of course. It includes a number of 6 activities that AEOD is involved in, NMSS, and of course, 7 Office of Research has been working with NRR on some of 8 these activities I have already described.
9 Now, unless you have other questions, I was going 10 to go to Tom King, so he can get into the real substance of 11 the issues.
12 MR. KING: Thank you, Ashok.
13 What I wanted to cover was to briefly summarize 14 the recent accomplishments since the last status report in 15 June, and then to discuss the review process and the key 16 technical and process issues that have come out of 17 developing the reg guides and SRPs to date.
18 If I could have Slide 6, please.
19 Slide 6 summarizes the recent accomplishments.
20 The first bullet talks about the draft reg guides and SRPs, 21 but I think what I will do is when we get to Slide 7, we 22 will talk about that in more detail.
23 We are continuing to review the industry-initiated 24 pilot applications, as Ashok mentioned, the pilot 25 applications in four areas, ISI, IST, QA, and tech specs.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
.- - -. - -..-.- - ..-.. - ._ .-- .--. ~ -. . . - - . - - - . _ . - . - . - - - ~ .
13 1 It involves seven or eight plants that are participating in 2 the pilot process. We expect to complete those reviews and t
3 send to the Commission.a recommended decision over the next i
4 two to eight months, starting in December with the tech 5 specs and then through June of next year with ISI and IS*r.
6 CHAIRMAN JACKSON: To what extent have those 7 industry-initiated pilots informed the development of the 8 guidance documents that you are talking about?
9 MR. KING: They have provided input. We have gone ,
10 through and taken our list of issues that we have developed l 11 in drafting the reg guide and SRP and looked at the pilots 12 as to how they were addressing those to get some feedback, 13 and we have actually gotten some feedback that has been 14 incorporated.
15 CHAIRMAN JACKSON: And it is somewhat of a. lag, 16: also, you are saying, . relative to when the final outputs of 17 the pilots will be available. Is that a fair statement?
18 MR. KING: I am not sure.
19 CHAIRMAN JACKSON: Well, what I am saying is, do 20 you feel you have gotten all out of the industry-initiated 21 pilots that you can relative to how it propagates into the ;
l 22 development of the guidance documents?
23 MR. KING: I suspect we will probably. As they 24 continue to respond to request for additional information, i 1 25 we will continue to learn some more, i
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
, i
.. . . = . . . - _ - .~ - . . _ - . .. .- =.
14 1 CHAIRMAN JACKSON: Okay.
$ 2 MR. HOLAHAN: I can remind you that I believe at 3 the last commission meeting, we presented a matrix 4 identifying 10 or a dozen issues and which ones we learned 1
5 stuff from on the various pilots and a couple of areas where 6 we needed to do more work.
1 j 7 CHAIRMAN JACKSON: And those ones where you have 1
j 8 identified that you have learned some things from the pilot, 9 is that what you mean when you say these things have or --
j 10 MR. HOLAHAN: Yes, yes.
11 CHAIRMAN JACKSON: -- what you have learned has i 12 been incorporated in these guidance documents?
i 13 MR. HOLAHAN: Yes.
, 14 MR. KING: Yes.
15 CHAIRMAN JACKSON: Okay.
)
16 MR. KING: The backup viewgraphs to the package l 17 'you have has some more details on the pilots in terms of the 18 plants and the schedules and so forth. I wasn't going to 1
19 cover those specifically.
j 20 The third bullet talks about the IPE and IPEEE.
21 We are continuing to review in both of those areas. We 22 currently have 19 IPE reviews to go until we are complete.
23 We expect 16 of those 19 to be done by December. Three will l l
l 24 probably carry over until next year, probably spring or so. l 25 Those are three where we have had problems with the IPE, and l ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
. . 1 15 1 we have requested that parts of it basically be redone. We 2 are waiting for a resubmittal.
1 3 CHAIRMAN JACKSON: Let me ask you a question. '
4 These IPEs are essentially PRAs; is that correct?
5 MR. KING: Yes.
l 6 CHAIRMAN JACKSON: Will you be coming out of that l 7 review with some assessment of how strongly coupled they are 9 to the design basis or how well known the design basis is l 9 for those plants relative to what these IPEs, in fact, are 10 showing?
11 MR. KING: Not through the IPE program. We are 12 not doing that. We are not trying to go back and confirm j 13 the design basis through the IPE program. We would expect 14 licensees in doing their IPE actually reflect the as-built 15 and operated plant. We have not checked that.
16 CHAIRMAN JACKSON: Mr. Thadani, you look like you 17 must say something.
18 MR. THADANI: No. I think that is the answer.
19 As you know, we have 50-50 4F letters out now, and 20 depending on what results come out as a follow-up to those 21 letters, there may be an action that we may have to follow.
22 CHAIRMAN JACKSON: Okay. I've got you.
23 MR. KING: The IPEEEs, we have 24 of those under 24 review. None have been completed at this point that we 25 would expect early next year that they would start coming ANN RILEY & ASSOCIATES, LTD.
I Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
a . j l
16 1 out, the staff evaluation reports on those.
2 The other thing I want to mention on IPEs is that 3 we have prepared an insights report. A copy was sent to the 4 Commission last week. We have also been going to the ;
5 regions and briefing them on the insights coming out of the 6 IPEs, both the generic insights and the plant-specific 7 insights, so they can factor them into their inspection 8 programs and other interactions with licensees. So that is 9 continuing to go on.
10 CHAIRMAN JACKSON: I hate to keep dwelling on the 11 same thing, but let me ask you this question. Based on wnat 12 you may get out of the 50-50 4F responses, the letter 13 responses, are you going to do some juxtaposition of any-14 sample of the IPEs, what comes out of that to have some 15 sense? In a sense, these insights are based on acceptance 16 as is, right?
17 MR. THADANI: That is correct.
18 CHAIRMAN JACKSON: So is there going to be any 19 kind of a sampling?
20 MR. THADANI: If certain plants are identified 21 which may, in fact, have differences, then I think we would 22 go back to those plants --
23 CHAIRMAN JACKSON: And review the IPE?
24 MR. TRADANI: -- and ask them --
25 CHAIRMAN JACKSON: To review.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
. o 17 1 MR. THADANI: -- to address those.
2 CHAIRMAN JACKSON: Redo.
3 MR. THADANI: Yes.
4 CHAIRMAN JACKSON: I got your point.
5 MR. THADANI: Yes.
6 CHAIRMAN JACKSON: Thank you.
7 MR. KING: The fourth and fifth bullet really 8 address the proposed Reliability Data Rule. AEOD conducted 9 a public workshop in June and received a number of comments.
10 They are continuing to look at those and work on resolution.
11 In parallel, I understand industry recently 12 submitted some sample data to demonstrate a proposed 13 voluntary alternative to the data rule.
. 14 CHAIRMAN JACKSON: Let me ask if I may, Mr. Jordan 15 --
16 MR. JORDAN: Yes.
17 CHAIRMAN JACKSON: -- where do things stand with 18 regard to our review of that sample data?
19 MR. JORDAN: We have a dataset that represents the 20 data elements from the safety system performance indicator 21 that INPO uses, and we are applying those data elements into 22 our reliability data scheme. We are still in the process of 23 assessment to identify what elements might be needed in 24 order to assure that we have training level system 25 reliability.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
. , i 18 1 CHAIRMAN JACKSON: Okay.
2 MR. KING: And the backup viewgraphs have some l 3 additional information on the Reliability Data Rule. I 4 CHAIRMAN JACKSON: Okay.
5 MR. KING: And finally, AEOD has completed l
6 development of PRA training guidance document, NUREG BR0228, l 7 and that was issued in July. They have also developed a 8 prototype PRA for a technical managers course, which they l 9 had a dry run on several months ago and I understand will be 10 available, be offered to the Staff in the next several 11 months.
12 If I could have Slide 7.
13 CHAIRMAN JACKSON: I hate to do this to you, but 14 given that this PRA training guidance docunent has been )
15 completed, how is it being used?
1 16 MR. JORDAN: Okay. I can answer that. It is the 17 basis for managers identifying appropriate courses for staff 18 members. So it is a road map in order to provide the right 19 level of qualification for staff members.
20 CHAIRMAN JACKSON: So it identifies some 21 qualification level and associated training program for a 22 given function in a job that someone has?
23 MR. JORDAN: Correct. That is correct. So it 24 identifies the various levels of qualification and then the 25 scheme of courses that, of course, can be looked at with ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
19 1 respect to that individual's experience, education, and 2 training to pick the right courses.
3 MR. KING: Slide 7, please.
4 Slide 7 gives a little more detail on the 5 regulatory guides and standard review plans that are being 6 developed to support risk-informed regulation. The 7 regulatory guides are really the guidance for licensees in 8 terms of what their submittal should contain, and then the 9 standard review plan is guidance for the staff as to how to 10 review that submittal.
11 Early in 1996, we had put together inter-office 12 teams to draft the reg guides and standard review plans, and 13 the ones being worked on are listed here. We had also put 14 together an inter-office PRA coordination committee to 15 provide some oversight and direction into that effort.
16 overall, those activities have been working well.
17 Currently, there are drafts for all of the reg 18 guides and SRPs. The ISI one has slipped three months, as 19 noted in the SECY paper that came up, primarily because of 20 late start on the pilot programs, but the others are 21 underway. They are under various stages of review. We plan 22 to get them to ACRS.
23 We have also developed a draft NUREG 1602 which is 24 a key reference document in the general reg guide in terms 25 of the standards for a PRA, in terms of the level of detail ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
l l
20 l 1 and quality and so forth. ;
i 2 CHAIRMAN JACKSON: So you have laid that out?
3 MR. KING: That has been sent to ACRS for review.
4 We have had numerous interactions with industry on both the 5 pilots and generic topics, as well as ACRS. We have had a 6 number of meetings with them. We have the next one coming 7 up on October 31st and then another one after that on 8 November 21st where we will be reviewing the reg guides, the 9 SRPs, the draft NUREG, and the issues that are coming out of 10 these things.
11 If we could move on to Slide 8.
12 Slide 8 shows the review process around which the 13 reg guides and SRPs are being developed, which we are trying 14 out on the pilot activities. It is a six-step review 15 process that we have defined to try and provide some 16 consistency and structure to the evaluation and review, and 17 we would expect licensee submittals and the Staff review 18 would follow these six steps as much as possible.
19 The six steps are shown on Slide 8, and the 20 feedback loops are shown. We thought it would be useful to 21 put it in this presentation as background because, as we get 22 into the discussion of the technical and policy issues, this 23 will illustrate the sequence of the logic in the evaluation 24 and I think will help in understanding where the technical 25 and policy issues fit in the evaluation process.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
-- -- . . - . -. . . . . - ~ - . .. . .---.--- . - - . .- -
, c.
1 21 1 The steps we believe are consistent with the PRA q , 2 policy statement, they are set up such that risk assessment 3 complements the deterministic evaluation and defense in 4 depth.
5 There is a step that specifically was put in on i
6 performance monitoring, which is related to one of the 4
7 policy issues we are going to talk about, and even though 8 there is not a feedback loop shown, if you go through this
! 9 process, you could end up coming back to step one and 10 redefining the scope of your proposed change, depending on l 11 how the outcome of the evaluations were.
12 If we could go to Slide 9.
i j 13 Pages 9 through 13 contain a list of what we call 14 the key technical and process issues. These are things that 15 we are addressing as part of the reg guide and standard 16 review plan development, and they were identified as part of 17 drafting the reg guide, standard review plan and interaction 18 with the pilot projects.
19 I don't plan to discuss all 27 of them, but what I 20 wanted to do is highlight the ones that are related to the 21 policy issues that are going to come up later on in the 22 briefing, as well as any others that are of particular 23 importance.
24 We thought it would be useful to present these in 25 this briefing because they do provide some key background ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
22 1 regarding the six-step review process, as well as the 2 background for understanding the policy issues.
3 CHAIRMAN JACKSON: Let me ask you, before you go 4 through them, can you say to what extent these questions 5 will be addressed in the guidance documents being developed, 6 and if not, are they dependent upon the Commission 7 addressing the policy issues, and if they are not dependent 8 upon that, how are you working on answers? So it is a 9 three-part question.
10 MR. KING: All of them will be addressed in some 11 fashion in developing the reg guides and standard review 12 plants.
13 CHAIRMAN JACKSON: Okay.
14 MR. KING: We are proceeding on the ones that are 15 related to the policy issues. Th? path we are proceeding 16 down is consistent with what we aue recommending on the i
17 position on the policy issues. If the Commission decides 18 otherwise, we will have to revisit those. i l
19 CHAIRMAN JACKSON: Okay. I 20 MR. KING: Before I get into some of the example 21 issues, I did want to say a couple of things about how we 22 are using this list.
23 We put it together for several reasons; one, to 24 help focus attention on the more important items, both Staff 25 and management intention. Two, it is a good way to track
)
ANN RILEY & ASSOCIATES, LTD.
Court Reporters l 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 I (202) 842-0034
t . .
I 23 1 ' progress as to how close we are to getting these things 2 resolved, and three, it does provide, as I mentioned 3 earlier, a systematic way to go through and get some 1
4 feedback from the pilot plants, ask what the pilot l 5 activities -- find out what they are doing and to address 1
- 6 each of these things. So it is being used in several 7 different ways. l 8 Also, I want to mention thac some of these issues 9 have sub-elements. We didn't list all of the sub-elements 10 because it would get too complicated. ,
1 11 Also, some of these issues, the answers may have 12 -- there may be several options in the way to deal with some l
13 of these issues, and in some cases, we will probably 14 recommend -- go to ACRS with some options, and we may want 15 to go for public cos.v.ent en some options and make a final 16 decision after we get feedback from the public comment l 17 process.
18 So we are not planning at this point to pick just l
19 one option for each one. Where it makes sense to list 20 several options, we would plan to do that.
21 Our next meeting with ACRS is going to focus on 22 these issues as part of reviewing the reg guide and standard 23 review plant.
24 Let me start with page 9. These are laid out in 25 accordance with the six steps. Roman Numeral I is step one ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 1
. - - - - -. _ - . - - _ . - _- .- =- - . . . . - _.
O ,
1 24 1 1 on the flow diagram, and I- (a) , what information does the 2 licensee need to submit to characterize the change, this 3 addresses right up front the point that Mr. Thadani brought 4 up. Unless a plant knows its current licensing basis and 5 has the plant built and operated in accordance with the risk 6 evaluation and the deterministic evaluation, it may not be 7 very useful. So we want to establish right up front that a 8 licensee has confirmed its current licensing basis and that 9 the plant is built and operated in accordance with it, so i
10 that the rest of the analysis is consistent with that.
11 Issue II- (b) , what are the acceptance guidelines 12 for the deterministic evaluation, this is one we have been 13 struggling with quite a bit. Again, the PRA policy 14 statement says PRA is to be used to complement the 15 deterministic evaluation. Deterministic terms like " defense 16 in depth" and " design margins" and so forth are used quite a 17 bit, but when you go to write the standard review plan and a 18 regulatory guide to find exactly what is meant by those 19 things and what are the acceptance criteria, it gets a 20 little tougher. So we have been struggling with this. I am 21 not here to say we have an answer yet, but it is going to be 22 one item that is going to involve a lot of discussion over 23 the next several months.
24 CHAIRMAN JACKSON: Well, one could argue that that 25 is an interesting statement, but one could also say that ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
. . 1 1
25 1 maybe this exercise, then, in developing a PRA framework 2 helps us focus on what we mean --
l 3 MR. KING: Yes. '
4 4 CHAIRMAN JACKSON: -- by defense in depth and
, 5 design margin. ,
1
- 6 MR. KING
- Yes.
7 MR. HOLAHAN- As a matter of fact, on that 8 subject, we have an interoffice meeting this afternoon at 4
- 9 o' clock to see if we can come a little closer to figuring 10 out exactly what this ought to be.
11 MR. KING: Slide 10, please.
12 Item III- (g) and (i) are directly related to two 13 of the policy issues, the policy issues associated with 14 plant-specific application of the safety goals and the risk 15 neutral versus risk increase. (g) is how should the 16 acceptance guidelines be structured, and that gets into 17 issues like what metrics should be used, should it be core 18 damage frequency, condition of containment failure 19 probability, large early release frequency, some other 20 aspect, how do we pick the values to be consistent with the 21 safety goal, considering the fact that we are talking 22 plant-specific application, how do you account for less than 23 full scope PRA.
24 The safety goal policy statement was fairly clear 25 that the risk that it was talking about was from all aspects ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
.. -~ . -. . _ . - . ~ . . . ._. .-.
26 1 of plant operation, and a lot of the risk analysis that is 2 out there, including the IPEs, are focusing on full power 3 operation only.
4 CHAIRMAN JACKSON: I have to play the devil's ,
5 advocate here again. If we look at issue (a), what 6 determines the extent to which risk analysis can be used, 7 did the Commission's policy statement itself address that 8 question?
9 MR. HOLAHAN- It has.
10 MR. TRADANI: Yes. I think, in fact, that was the 11 very first bullet when I went through.
12 CHAIRMAN JACKSON: Right.
13 MR. THADANI: To that extent, it is supported by 14 methods and data and to be used as complement.
N 15 CHAIRMAN JACKSON: So what you are trying to do is 16 pin down in some more quantitative way what that is?
17 MR. THADANI: More details and what does it really 18 mean, supported by methods and data, what does that mean.
19 CHAIRMAN JACKSON: I just want to go through a 20 couple of them, not all of them.
21 MR. KING: Sure.
22 CHAIRMAN JACKSON: With (b), where you say what 23 determines the required quality of the risk analysis, will 24 the guidance documents answer that question?
25 MR. KING: Yes.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
o .
27 1 CHAIRMAN JACKSON: Then, my favorite topic is (d),
2 how is uncertainty to be addressed. There are two 3 questions. One is, how has uncertainty been treated in the 4 past, in past uses of PRA insights in the regulatory 5 process. So, if you could give me an answer to that, and 6 then the other one, which is that I note that in your SECY 7 96-218, the Staff indicates that it intends to use the mean 8 value for comparison with numerical guidelines associated 9 with absolute measures, such as core damage frequency, and 10 this is my favorite topic.
11 So the question becomes answering how an 12 uncertainty to be addressed, referencing it to how it has 13 been addressed in the past. Does this imply that if you 14 have equal mean values -- this is where the rubber meets the 15 road -- with big differences in uncertainty, would that lead 16 to the same regulatory decision?
17 MR. KING: Not necessarily.
18 CHAIRMAN JACKSON: Okay. So can you maybe 19 illuminate or amplify on that a little bit?-
20 MR. KING: We haven't settled exactly on how we 21 are going to treat uncertainty at this point either.
22 CHAIRMAN JACKSON: Okay.
23 MR. KING: The Commission safety goal policy said 24 use mean values in assessing against the goals.
25 CHAIRMAN JACKSON: Do you feel that is enough?
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
- - _ --- . . - - . -. - _ _ =
28 1 MR. KING: In some cases, it is enough, but it 2 does require a full uncertainty analysis be part of the l
3 analysis and evaluation by the licensee and by the review by 4 the staff.
5 MR. THADANI: I think that this is obviously a 6 very tough issue. Mean values relate to where you are able 7 to quantify a number of things and you are able to actually 8 draw some sort of distribution and so on, but there are many 9 elements where the uncertainties are really not quantified, 10 organizational cultural issues, some other things, for 11 example, millstone issues, some of the millstone issues.
12 So there are areas where uncertainties are not 13 quantified, programmatic weaknesses or problems. So what it 14 really boils down to is when you get an issue where let's 15 say a licensee wants to use these techniques, it seems to me 16 we are going to have to look at that specific issue and try 17 to use some judgment on what are some of those so-called ;
18 unquantified uncertainties and should we, in fact, use a 19 mean value, then. Maybe not.
20 So it seems to me that there has to be some 21 balance brought into this process to recognize that we 22 cannot answer all the questions up front, I don't think, but 23 the process should allow those considerations whenever there 24 is an application to be made. So I am hoping that is how we 25 can move forward.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
- . .- . . - - _ _ _ . _ . . -. _ ._ _ - _ _-. .~ -
i i '
29 l 1 CHAIRMAN JACKSON: If you look through this list
! 2 of A through I, have you identified which ones minimally 1
3 have to be answered at some level in order to develop 4 realistic guidance documents? Have you answered that?
l 5 MR. KING: I think it is all of them. Our intent j l 6 is all of them. '
7 CHAIRMAN JACKSON: To have some answer?
i 1
8 MR. KING' To have some answer, yes.
9 CHAIRMAN JACKSON: For all of them?
10 MR. KING: Yes.
11 MR. HOLAHAN: Yes. l l
12 MR. THADANI: Yes. ;
13 CHAIRMAN JACKSON: So, when you say, then, that 14 these draft documents are available, that means that you 15 have some answers to all of them relative to development of 16 those guidance documents? ;
17 MR. KING: The drafts have some answers. Whether 18 there is consensus on the Staff regarding those answers is l 19 another question. They are under review.
20 CHAIRMAN JACKSON: That is what you mean when you 1
21 say under Staff review?
22 MR. KING: Yes.
23 CHAIRMAN JACKSON: I see. Okay. Now I 24 understand.
25 COMMISSIONER ROGERS: Well, before we leave this, ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
30 1 I'd like to just pursue one little aspect of it, and you 2 touched on it a bit, the question of quantitative measures 3 or lack of quantitative measures.
4 It is kind of my intention that some areas of risk 5 analysis that are being used, particularly in the fuel cycle 6 facilities, are not really being carried out using 7 probabilistic analysis, but they are risk analyses, and it 8 seems to me that we are going to have to deal with that 9 issue of risk analyses which are not really based upon a 10 strictly probabilistic calculation, and nevertheless, do the 11 job in some way. l l
12 This may or may not fit into the reactor area. It 13 probably does to some extent, but it may_be very important 14 in the nonreactor area.
15 MR. TRADANI: Yes, yes.
16 COMMISSIONER ROGERS: So I do think that while I 17 am very high on numbers, I do think we have to recognize 18 that there are other ways of analyzing risk that are not 19 strictly based on probabilistic calculations, but '
20 nevertheless are something closer or a little bit to the 21 usual traditional deterministic, but nevertheless are a risk 22 analysis rather than a straight engineering calculation of 23 some sort.
24 I hope somehow we keep that in mind here for those 25 situations where that is the only way to go.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
31 1 MR. THADANI: Yes, indeed. In fact, I was going 2 to say some people don't like to hear this, but I think when 3 you go through the process of risk analysis, the first parts 4 are probably the most robust in the sense of logic models.
5 The event trees and fault trees, by and large, I think, are 6 the most robust.
7 When you get into quantification is where one has 8 to be cautious, and no matter what application, one needs to 9 look hard, I think.
10 MR. HOLAHAN: I think in many cases, this is not a 11 go and no-go sort of decision. There are stages, as in the 12 policy statement suggestion is, first, does the 13 state-of-the-art support the kind of issue you are trying to 14 deal with, and I think after that, if the issue is amenable 15 to a probabilistic risk assessment, you still want to choose 16 the proper-sized tool for the job. So, if it is a 17 relatively easy question or, in fact, a qualitative risk 18 assessment, it would convince you that this is a net 19 improvement, and why go through an elaborate uncertainty 20 analysis to figure out how sure are you or how big is that.
21 CHAIRMAN JACKSON: It strikes me, though, that 22 there is kind of-a baseline question that in doing 23 everything that you have just described, you have to 24 address, which is kind of -- let me see if I can articulate 25 it. It is essentially saying how much do I have to know and ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
32 1 be bale to quantify to make a judgment here, and if I want 2 to make a judgment somewhere else at some other level, that 3 it requires that much more, and if I can't get that, then 4 the decision-making has to be done a different way.
5 Now, will the kind of guidance that you are 6 working your way up on allow those kinds of assessments to 7 be made?j 8 MR. HOLAHAN: Clearly, that is our goal. It is 9 early on in this process to identif y Whn is the proper tool 10 for the given issue. Can you make a acrtain type of 11 decision with a qualitative analysis? Does it take a 12 quantitative analysis, but not necessarily an elaborate 13 uncertainty analysis, or in some cases, are we making 14 sufficiently complicated decisions that a full scope, full 15 uncertainty analysis is needed?
16 I think, in each case, what you are trying to do 17 is to say do I have confidence in the decision that I am 18 making.
19 CHAIRMAN JACKSON: I appreciate what you are 20 saying, and I guess all I am really asking is will the 21 guidance documents be such that one proceeds along a path 22 and comes to come bifurcation point that says I can go 23 further down this PRA path or I can't, and if I can't, then 24 it kicks over into something else. I mean, that is 25 presumably where you are trying to go.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
33 1 MR. HOLAHAN: We are working on that very subject.
2 As recently as yesterday's meeting, we were going to divide 3 up regions in which more details on certainty analysis was 4 appropriate and where less is needed. That is the kind of 5 thing that belongs in a guidance document.
6 CRAIRMAN JACKSON: Okay. l 7 MR. THADANI: We had identified in an earlier 8 paper, actually, that generally we were looking at three 9 categories of applications. One was what we called 10 prioritization which is, by and large, NRC activity, and 11 that one could go with something fairly simplified. You i 1
12 don't want to spend a lot of resources to see how to i 13 prioritize things, but that you can use etter understanding 14 of risk importance to make those kinds of decisions. That 15 was probably the simplest type of application in terms of I 16 the quality of analysis. I 17 The next one was where the decision was not really !
18 eliminating a requirement, so to speak, but that you are 19 just shifting importance, so to speak, high safety 20 significance and medium safety significance and low aafety 21 significance. That would require a certain type of 22 analysis.
23 Whereas, if you are really completely walking away 24 from what today's requirement might be, then one has to do a 25 very thorough analysis before saying that that makes sense.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
34 1 So those are the categories that we have been looking at, 2 and then, of course, the toughest issue, I think, is the 3 issue of how to deal with uncertainties in all of this.
4 CHAIRMAN JACKSON: Okay, thanks.
5 MR. KING: Just to quickly highlight item (i) at 6 the bottom of page 10, should the acceptance guidelines 7 apply to proposed changes individually or as a package, the 8 topic there is when someone comes in with a proposed change, 9 can they group changes together, look at risk changes due to 10 changes, proposed changes in tech specs versus ISI versus 11 graded QA and add them all up and get a net reduction or net 12 increase, whatever it turns out to be, or do we want to 13 limit it to just a single topic. So that is the issue that 14 is being talked about there.
15 On page 11, issues associated with implementation 16 and monitoring, this is tied to one of the policy issues.
17 This step was explicitly added in the process so that we 18 would use performance monitoring as much as practical to 19 check the assumptions and provide feedback into the 20 evaluation and the changes that were being made.
21 If assumptions are made regarding equipment 22 reliability or so forth, this is a step that would hopefully 23 check to see whether those assumptions are becoming true, 24 and if not, provide the appropriate feedback into the 25 process.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
1 35 1 CHAIRMAN JACKSON: Now, a natural question that ,
2 arises is this. Now we have a maintenance rule that just 3 became effective. Presumably, each one of these questions 4 have to be addressed in implementing that rule for the SSCs 5 that we mean for it to cover. What are the answers to those 6 questions within the context of the maintenance rule, and 7 then how does that flow into this and vice versa?
8 MR. KING: It may very well be the maintenance 9 rule is accomplishing this for whatever proposed change they 10 are making.
11 CHAIRMAN JACKSON: Well, I guess what I am trying l
1 12 to say is that it strikes me that that is something you have l l
13 to come and tell us; namely, how are these four questions I l
14 being answered within the context of the maintenance rule, 15 and how, then, does that tie back into what you are doing 16 and how is what you are doing affect how these questions are 17 answered.
18 MR. THADANI: I think that there are two parts 19 that we need to be sure about, i
20 The first part would be depending on what i 21 performance criteria one sets up. If those are really 22 related to reliability analyses, so to speak, then, clearly, 23 one has to have some guidance document on how to assess and l 24 interpret what has been done.
25 As far as the maintenance rule is concerned, some ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
36 1 licensees may have used some reliability guidelines that may 2 have come out of the PRAs. Some may not have. That was not 3 strictly necessary under the maintenance rule.
4 We have initiated our inspections, what we call 5 baseline inspections, under the maintenance rule. As I had 6 indicated to you in the past, I am hoping that by February 7 time frame we will have done enough inspections, 10 or 12 or 8 some number like that, that we can probably draw some 9 inferences and some potentially generic insights.
10 Our intention is to then step back. What we learn 11 from those inspections would be considered, if it is 12 . appropriate for these guides, but that I don't have the 13 answer today as to what we are going to find.
14 CHAIRMAN JACKSON: Let me, then, say this. I am 15 going to be explicitly asking you this. Since you would be 16 coming back in the March-April time frame to briefing the 17 Commission again, that you come back as part of that brief 18 with answers to these four questions in the context of the 19 maintenance rule --
20 MR. THADANI: Yes.
21 CHAIRMAN JACKSON: -- and how that ties into the 22 answers to these questions --
23 MR. THADANI: Yes.
24 CHAIRMAN JACKSON: -- within the context of what 25 you are doing --
l l
l ANN RILEY & ASSOCIATES, LTD. !
Court Reporters I 1250 I Street, N.W , Suite 300 .
Washington, D.C. 20005 (202) 842-0034 l 1
i
. . l 37 1 MR. THADANI: Yes.
2 CHAIRMAN JACKSON: -- because it is very 3 important, okay? Because first of all, we shouldn't be 4 going down a path relative to the maintenance rule that is 5 somehow different than the path we are going down in the 1
6 overall PRA Implementation Plan. l 7 Two, we say that the maintenance rule is our first 8 example of a risk-informed performance-based rule, and if it l l
9 is, then it better tie into the PRA framework that we are 10 developing.
11 MR. TRADANI: Yes.
12 CHAIRMAN JACKSON: And I understand your point 13 about doing these baseline inspections, but since you 14 indicated that sometime after the first of the year --
15 MR. THADANI: Yes.
16 CHAIRMAN JACKSON: -- you will have more data, 17 then along around March-April, you should be able to put it 18 together, and you will be further along in these reviews of 19 your reg guides because I think this le very important.
20 MR. THADANI: It is critical, I agree, and we will 21 do that.
22 MR. HOLAHAN. I think it serves the same role as 23 some of the pilot applications, but there are differences in 24 the scope and the intent of the maintenance rule versus the 25 general --
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
38 1 CHAIRMAN JACKSON: No, I appreciate that, and that 2 is, in fact, what you have to come back and tell us because, 3 in fact, we need to understand how the scope differs.
4 MR. HOLAHAN: Right.
5 CHAIRMAN JACKSON: We have talked about this 6 before within the context of the reliability data rule or 7 putative reliability data rule, but it is very important 8 because it is important in terms of consistency in how we do 9 things.
10 MR. HOLJJUW : Yes.
11 CHAIRMAN JACKSON: Okay, thanks.
12 MR. KING: Let me move on to Slide 12, issues 13 associated with integrated decision-making. This is where 14 the deterministic and the probabilistic evaluations come 15 together and a decision has to be made. Again, it relates 16 to what we talked about earlier, what are the deterministic 17 decision criteria. A number of these items are directed .
18 toward that.
19 Let me just mention item (g) at the bottom of the 20 page, the role of 50.109. This actually came out of one of 21 the pilot programs. If the Staff has conducted the review 22 and feels that something else needs to be done over and l 23 above what the licensee has volunteers to do, do we have to 24 follow 50.109 to get that in place.
25 Counter to that or the reverse of that, we have ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
39 1 also discussed, and that is a licensee comes in and proposes 2 a change that causes some increase in risk, why shouldn't we 3 apply the backfit rule in a reverse way. Is the cost i l
4 savings associated with that sufficient to justify the 5 increase in risk? So that is what we have been talking I
6 about. I am not here to give you an answer, but that is '
7 what that item means.
8 CHAIRMAN JACKSON: Let me just make one other ;
9 comments. Aren't B, C, and D on here linked? That is, if 10 one really had a process for addressing uncertainty, then 11 this issue of the extent to which the existing degree of 12 defense in depth should be maintained is more addressable, 13 as well as the issue of the margin of safety.
14 The only reason I keep bringing this up, you say 15 it to me and I am saying it back to you. Somehow we have 16 got to really get our hand around where we can get our hand 17 around. I will put it that way, get our hands around where 18 we can get our hands around the uncertainty issue, because 19 if we don't somehow get that bullet bit and at least know 20 where we can and cannot do something. I understand we 21 cannot do it everywhere. I don't see how we are going to 22 answer (c) and (d).
23 MR. KING: Slide 13 has to do with what actually 24 has to be part of the submittal. The documentation needs to 25 be submitted. Do we need the full PRA or just some summary ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
40 1 of what was done that describes in enough detail?
2 Regardless of whether the full PRA comes in or 3 just some summary information, item C is a process issue.
4 Will our explicit use of risk information and plant-specific 5 decisions now require PRAs to be put on the docket and 6 litigated? It is an item that just remains to be seen at 7 this point.
8 With that, Gary Holahan is going to talk about the 9 four policy issues.
10 MR. HOLAHAN: Could I have Slide No. 14, please?
11 The four policy issues that we identified are shown here, 12 the role of performance-based regulations, the use of the 13 safety goals or guidance, the decision process derived from 14 the safety goals on the plant-specific basis, whether 15 in' creases in risk should be allowed at all or under what 16 circumstances increases are appropriate, and then something 17 of a process question on how should changes in the ISI and !
1 18 IST program be -- )
i 19 CHAIRMAN JACKSON: Mr. Holahan, you know I can't 20 let you slide. How clearly do you feel the Staff knows what 21 performance-based regulation means, how clearly do you feel 22 you know, and what degree of concurrence is there on a 23 definition? If not, how do you go about -- what are you 24 doing to clarify that?
25 MR. HOLAHAN: Well, I think there is not a I
ANN RILEY & ASSOCIATES, LTD. I Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
- -. - . .- = . . _ - - .- - . - - . - _ . _ _ _ _ _.
41 1 unanimity of understanding as to what the definition of 2 performance-based regulation is.
3 I have seen lots of different definitions. I l
4 think we understand common features that performance-based 5 regulations have.
6 I remember that we were told that I
7 performance-based fire protection requirements are being put 8 in place around the world. So a number of the staff met 9 with the National Institute for Standards and Technologies.
10 They were involved with those things, and I guess they 11 confirmed our view that each country, each application has a 12 slightly different definition, but with some common 13 elements.
14 So I think we are starting out with maybe buzz
- 15 words, but we are developing guidance documents which I hope 16 will clarify the situation.
17 CHAIRMAN JACKSON: Is there a utility, too, and 18 have you been able to garner any input from other types of 19 industries or regulatory bodies that have gone at this? Are 20 we on the cutting edge?
21 MR. HOLAHAN: Well, I think there are some other 22 areas. We have had some input from the industry. There is 23 an industry white paper on the subject.
24 A number of recent PRA conferences have identified 25 this as an issue, and so it has been discussed. We ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
. ._ . -. - ..- - - ._ _. . _.. ~ -.-. . . _ - . . _. -
42 1 discussed it with the ACRS.
2 CHAIRMAN JACKSON: But is it all the nuclear 3 people talking to each other as opposed to --
4 MR. HOLAHAN: I would say, largely, it is. In the 5 fire protection area, obviously, it goes well beyond the 6 nuclear area, but there are probably numerous other 7 industries that we haven't fully tapped.
8 CHAIRMAN JACKSON: Okay.
9 MR. HOLARAN: Can I have Slide 15, please?
10 The issue on Slide 15 being the role of a 11 performance-based regulation in the PRA Implementation Plan, 12 the Staff identified three options. I guess it is also 13 important to note that as a result of the strategic 14 assessment, there is, in fact, a paper on the subject which 15 also identifies three, I would say, similar, not identical 16 options.
17 So the Staff's moving ahead on these options, I 18 think, is also tied to the decision in the strategic 1
19 assessment arena.
20 The first option we identified basically is to 21 continue our current practice, and our current practice 22 being what Tom King showed, which, in fact, is to have 23 developed what we called step four, which is as part of 24 risk-informed regulation, actually searching out 25 opportunities for monitoring in a plant application ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
9
- 43 1 information that would validate the assumptions that went 2 into a risk analysis or the assumptions that went into a 3 deterministic engineering analysis. So the first option is 4 to continue with that process.
5 The second option is a bit more aggressive in that 6 it would within the context of the PRA Implementation Plan 7 solicit additional areas in which the industry was 8 interested in pursuing examples.
9 The third option would, in effect, be to create 10 something akin to the PRA Implementation Plan, which you 11 could name the performance-based regulation implementation 12 plan and collect together all those related topics and sort 13 of give it a life of its own.
i 14 Staff has recommended option one, but I think it 15 is fair to say that option one with a leaning towards option 16 two because there is some receptiveness to additional 17 initiatives, and the Staff did send a letter earlier this 18 year to NEI suggesting that at least some additional options 19 as a learning or pilot-type experience would be appropriate.
20 We have discussed this issue with the ACRS. At 21 the bottom of the page, you will see a quote from their 22 August letter. I think they were definitely very supportive 23 of doing at least what the Staff had recommended; that is, 24 to find a constructive place in each risk-informed decision 25 for a performance-based strategy to be included as a ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
44 1 verification or validation step, but I think the important 2 thing at this stage is that we are going ahead at least with 3 option one, and we need to do that for the development of 4 the regulation guides and the SRPs. If the strategic 5 assessment process should have the Staff to do more, I think 6 it is highly unlikely that it will be asked to do less.
7 So I think in the context of the reg guide and the 8 SRP, it is fairly clear what we should be doing. What is 9 not entirely clear and is a policy issue for the Commission 10 to decide is how much more should we do.
11 COMMISSIONER McGAFFIGAN: What was the response 12 from NEI to your letter? Did they have initiatives that 13 they would like to --
14 MR. HOLAHAN: I don't recall them coming back with 15 a specific example. It seemed to me that it was between the 16 time when they send us a draft of their white paper on 17 performance-based regulation and when they finalized it. So 18 I might say they were at least encouraged enough to go 19 forward and finalize their views. I don't think they have 20 identified specific examples to follow up since then.
21 MR. THADANI: I think it is important to make a 22 point. The PRA Implementation Plan is really focused on 23 risk-informed activities, and in some cases, where some 24 reliability guidelines are developed or to be used, then the 25 Implementation Plan can give guidance on how one would ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
-- _ _____ _ _ _______.____ __-_____.__U
45 l
1 assess that. I 2 Frankly, for many systems where you establish very 3 high reliability, steam-generated tubes, reactor and 4 pressure boundary, pressure vessel itself, many of the 5 complements, the expected reliability is very high, and so 6 the performance criterion one establishes cannot be first 7 failure. It cannot be that. It has to_be something else.
8 It has to be some engineering consideration that goes in, 9 how much thickness of a pipe, how much thickness can you 10 afford to lose.
i 11 PIA can tell you how important that component is, 12 and that is an important part. That is the risk-informed 13 part, but the performance-based part is non-numerical 14 because of what confidence one is trying to ascertain 15 reliability of a component.
16 That is why I think we need to be very clear on 17 what is it that we mean by performance-based because, in 18 many cases, going forward in the risk analysis approach 19 cannot answer some of the concerns that we might have. It 20 is that element of performance-based aspect. That would be 21 very difficult for this plan to address. It requires a lot 22 of thoughtful experience and understanding.
23 Really, that is why we said in our letter to NEI 24 that we need to learn from experience and we need to move a 25 little slower in this area is basically what we said. It is ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
1 46 i 1 not to say that we shouldn't go in this direction, but let's !
2 make sure where are we going, learn from whatever experience 3 we have as we go forward.
4 MR. HOLAHAN: I think, as also pointed out in the 5 Commission paper, resources are an important element here.
6 We are trying to maintain an aggressive schedule on the reg 7 guide and the SRP development, and there is a concern of 8 diverting resources from that activity if we take on a 9 little bit more than we can manage all at once.
10 Can I have Slide No. 16, please?
11 The second issue relates back to, as Mr. Thadani 12 mentioned, the June 15, 1990 SRM in which the Commission 13 instructed the staff not to use the safety goals for 14 plant-specific purposes, but to use them in a generic 15 decision-making, and that policy was, I would say, restated 16 in the PRA policy statement, but I think it was restated in 17 the context that the Staff needs to come back to the I 18 Commission if it proposes to do otherwise than the 1990 l
19 directions. I think we have read it not to be an absolute 20 prohibition, but if we develop this to be a worthwhile idea, l l
21 we need to bring it back to the Commission for Commission's l l
22 approval to go forward.
l 23 COMMISSIONER ROGERS: Well, if I could just say 24 anything?
25 MR. HOLAHAN: Yes.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters !
1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
47 1 COMMISSIONER ROGERS: It seems to me, as I recall, 4
2 when the Commission first took that position back in 1990, 3 PRA was not a very attractive method of making regulatory 4 decisions around here, and there was considerable resistance 5 to it up until later than that, I believe. So, at that 6 time, the Commission felt that we weren't on very solid 7 grounds in going beyond a kind of generic approach to using 8 the safety goals, out in the meantime, the kind of 9 development of PRA for nuclear applications and the data 10 that have developed have given us a great deal more 11 confidence that you can begin to think about the possibility 12 of using the safety goals themselves in some way for 13 plant-specific applications.
14 I can't speak for the whole Commission, but it 15 certainly seemed to me that what we were saying there was to 16 open the door to that possibility cautiously and not just 17 say that it is still locked.
18 MR. HOLAHAN: Okay. In fact, we developed two 19 options. One option would be to develop guidelines for 20 plant-specific decisions and to have those guidelines 21 derived from safety goals and the subsidiary objectives.
22 The second would be to derive plant-specific guidelines, but 23 to try to preserve the generic national average nature of 24 the safety goals by coming up with a scheme for relating an 25 individual plant regulatory decision, the effect it would ANN RILEY & ASSOCI.'sTES, LTD.
l Court Reporcers 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
48 1 have on the nationwide risk.
2 What we found is that the second approach is in 3 addition to being rather complicated, since one would have 4 to have a full understanding of the risk assessment at all 5 plants to make a decision on any plant, and I think it also 6 raises a number of rather complicated social and policy 7 issues about decisions, about is it appropriate to make a 8 risk decision at one plant where you are averaging out that 9 local effect nationwide.
10 CHAIRMAN JACKSON: Right. It is a lot harder to 11 raise an industry average than an individual number.
12 MR. HOLAHAN: Yes. So I think it is fair to say 13 we found the second option untenable.
14 The first option is very much desirable in the 15 sense that if we are going to use risk assessment in the 16 decision-making process, certainly the output from the 17 individual calculation seems to be a natural part of the i 18 calculation to use in that decision process.
19 CHAIRMAN JACKSON: Mindful of addressing all the 20 issues we have been talking about all afternoon.
21 MR. HOLAHAN: Absolutely, yes. Yes. 1 22 CHAIRMAN JACKSON: Okay.
23 MR. HOLAHAN: With a full understanding of scope 24 and uncertainties, et cetera, but as we say, without having ,
I 25 a locked door. l l
ANN RILEY & ASSOCIATES, LTD. ;
Court Reporters j 1250 I Street, N.W., Suite 300 1 Washington, D.C. 20005 (202) 842-0034 !
49 1 So the Staff is recommending option one which 2 would be a change in Commise, ion policy.
3 MR. THADANI: If I might, I think it is important 4 to note this is, in a way, one of the difficulties in pilots 5 because the Staff is pushing to get a lot of information, 6 and in some cases, that means a fair amount of additional 7 work on the part of those volunteer pilot licensees, and 8 that has caused some delays, trying to generate that 9 information or the need, the discussion back and forth as to 10 why is it really needed to generate this information. We 11 will wait and see how it all works out, but currently, we 12 have a number of outstanding questions to those licensees.
13 MR. HOLAHAN: I think there is also a related 14 aspect to it, and that is, in some sense, the genie is out 15 of the bottle already. '
16 We know that the maintenance rule is being 17 implemented in many cases with licensees using 18 plant-specific risk assessment.
19 CHAIRMAN JACKSON: That is why I asked you about 20 the implementation and the monitoring part.
21 MR. HOLAHAN: Yes.
22 CHAIRMAN JACKSON: That is exactly why you have to 23 have that issue addressed.
I 24 MR. HOLAHAN: Yes. And it is clear that licensees 1 25 are making other day-to-day prioritization and other ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 l (202) 842-0034 )
)
50 1 decisions based on the results of risk analysis.
2 So, to confine the Staff to not use a similar 3 approach in plant-specific decision-making, it seems -- I 4 think, as Commissioner Rogers suggested, perhaps that had a 5 basis at the time, but I think that time has gone.
6 CHAIRMAN JACKSON: We heard you, Mr. Holahan.
7 MR. HOLAHAN- And the ACRS appears not only to 8 agree with us, but I think they were considerably ahead of 9 us on this issue, encouraging this view for several years, I 10 think.
11 Can I have the seventh slide?
12 The seventh slide is really contingent upon the 13 answer to the sixth, and that is, if you are going to use 14 the results of a risk analysis in plant-specific decisions, 15 should those results, in effect, be a proof that no risk f
16 change has occurred or only an improvement has occurred, or 17 should increases be made under some circumstances.
18 I think what is fair to say is this is a policy 19 matter to the extent that explicit changes in risk would be 20 identified and approved because, in an unquantified way, I 21 think it is clear that Staff does through the normal license 22 amendment process, under some circumstances where we feel it 23 is appropriate and the Commission's regulations are met, 24 that we do allow small risk increases.
25 This would say that in the risk-informed l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
l 1
51 1 regulatory context, we would consciously, knowingly, and 2 with some numerical analysis make such a decision. So, 3 basically, the two options that we have identified as to 4 allow small increases under certain circumstances and the 5 regulatory guidance and the review plan would be the 6 guidance document to identify how small is small and what 7 are those circumstances, or we could say no, it is not 8 appropriate, you should use your risk analysis to hold the j 9 plant risk at some value where you think they currently are.
10 We looked at the pros and cons of these options.
11 The Staff has recommended the first to allow increases under 12 certain circumstances. We think that is appropriate. We l l
13 think w'e can identify how small is small and what is 14 appropriate. The reg guides will help to balance any small 15 changes with deterministic engineering margins to give us l 16 additional confidence that the decision we are making really I 17 makes sense.
18 The ACRS spent some time also reviewing this topic 19 and also agreed with the staff. I would say, as a matter of 20 principle, we haven't brought an example to the ACRS yet as 21 to how those guidelines would be developed.
22 COMMISSIONER DICUS: I want to ask you a question 23 about this. I don't necessarily disagree at least on the 24 surface with the recommendation, but if you were to, for 25 example, allow a small risk increase on one circumstance and ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
. . . - - _ . . _ . . - -- ._. ._ - _.- -. . . . - . _ . = . . . . _ _ .
52 1 later there is another circumstance and then later another 2 one, the way to follow these small risks that accumulated, 3 to have a point in time, you say --
4 MR. HOLAHAN: Yes.
5 COMMISSIONER DICUS: -- and you track this and you 6 know this.
7 MR. HOLAHAD: We will certainly address that 8 issue.
9 One of the techniques we have considered for that 10 -- and in fact, it is fair to say that industry in its 11 PSA application guide, I think, has addleased it to a 12 certain extent in saying that a plant would develop a 13 baseline risk analysis and then any changes that it made, 14 either risk increases or decreases, at either a certain time 15 interval or when a next major change would be anticipated, 16 the analysis would be updated, in effect, if they had moved 17 closer to some ultimate goal, that that would be reflected 18 and understood before the next change would be made.
, 19 CHAIRMAN JACKSON: At the risk of preaching to the 20 choir, let me just reference Mr. Taylor's beginning comments 21 and Dr. Thadani's comments; that it still tracks back to the 22 licensing / design basis issues because if you don't know what 23 you are building the PRA on and if changes aren't 24 appropriately captured and documented as of now, then it is 25 very difficult to talk about moving forward in terms of ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
53 1 looking at how a risk profile of ar. individual plant may 2 change; that these are inextricably linked issues. Do you i 3 disagree?
l l 4 MR. THADANI: No, not at all.
5 MR. TAYLOR: We agree 300 percent.
6 MR. THADANI: Totally.
7 In addition to that, I would note that you tasked l
8 us, the Staff. The Staff should track cumulative changes 9 for those indi"idual plants. So in the Implementation Plan, 10 we have an activity. Not only do we have expectation that 11 the PRA should reflect the plant design and operation; that 12 the plant should track if they are going to be using that 13 tool -- they need to keep track of cumulative effects, out 14 that the Staff will also be tracking that information.
15 MR. KING: Commissioner Dicus, in our list of 27 16 issues, it was item III-(e) on page 10, the issue you 17 brought up. We didn't miss it.
18 (Laughter.]
19 COMMISSIONER DICUS: You are on top of it.
20 MR. HOLAHAN: Let the record show the choir says 21 amen 22 Could I have Slide 18, please?
23 The fourth policy issue was a little different 24 from the first three in that it is more of a procedure and 25 process question than a technical or real technical type ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
54 1 policy matter, and that is where we are considering 2 risk-informed changes in the in-service inspection and 3 in-service testing programs, how should those be treated in 4 the context of the current regulations.
5 We identified three options. One would be to 1
6 consider them exemptions to the current requirements in 10 '
7 CFR 50.55a. If we were to review these as being quite i I
8 different from previous changes and the kind of alternatives I 9 the Staff has allowed to change in the past, then it would be most appropriate to treat those as exemptions.
10 11 The second option would recognize the fact that 12 the regulation currently allows for authorized alternatives 13 in Section 50.55a (a) (3) (i) , and the third option would be to 14 defer any such changes until the national consensus 15 standards, the ASME standard process had actually adopted 16 those changer.
17 We have looked at these options, and we have 18 considered whether the risk information and the kind of 19 decision we would be making would be consistent and 20 appropriate, similar to decisions we have made before, and j 21 our recommendation is to treat the code alternatives, to 22 treat the ISI and IST alternatives to the normal code 23 requirements as authorized alternatives under that element 24 of the regulations.
25 However, we think that carrying those for a long ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
55 1 period of time as authorized alternatives is probably not 2 the clearest and the best approach. So, in parallel with l
l 3 that, we would be working with the ASME to move these 4 alternative approaches into the national codes and make them 5 a part of the -- that would draw them into the normal 6 coverage of the regulations.
7 MR. THADANI: I might just note again, on that 8 one, the Staff is looking. For example, in ISD, there are 9 two approaches that are being looked at. One approach to 10 sponsor is ASME is the sponsor.
11 MR. HOLAHAN: I think in that context, if we were 12 to approve both options, then, perhaps, the ASME part would 13 be taken care of because the regulations refer to the ASME 14 code, but perhaps a role change would be appropriate to 15 reference the other methodology.
16 CHAIRMAN JACKSON: Is this a technical i 17 recommendation or a legal?
18 MR. HOLAHAN: This is really a legal and 19 procedural matter.
20 CHAIRMAN JACKSON: I mee.n, this is including the 21 legal staff analysis of this?
22 MS. CYR: We concluded that, in our understanding 23 of where they have looked at alternatives in the past, that 24 it is an alternative under the 50.55a (a) (3) .
l l 25 MR. THADANI: Correct. Otherwise, we couldn't l
l l
l l ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I St.reet, N.W., Suite 300 l Washington, D.C. 20005 (202) 842-0034 l
56 1 proceed under that.
2 CHAIRMAN JACKSON: But greater clarification would 3 come from this dual path.
4 MS. CYR: Ultimately to adopt that. Now there 5 will be an approved alternative that the ASME adopted as a 6 consensus standard. Then you would want to reflect that 7 essentially as your main part of your --
8 MR. THADANI: That is right.
9 MR. HOLAHAN: And largely, because this was a 10 legal and procedural matter, we did not ask the ACRS to 11 comment on it.
12 I will cover the last two slides quickly. We have 13 a number of activities over the next six months. We have 14 meetings with the ACRS, October, November, and December.
15 Those are largely focused on the regulatory guide and the 16 standard review plans.
17 We are still striving to issue the reg guides and 18 the SRPs by December 31st. We have a couple of major 19 activities to go in order to achieve that. I think both the 20 ACRS views and the CRGR in Novernber will be challenges to 21 the Staff to get those out on the current schedule.
22 We will be continuing our review over the pilot 23 applications, with the IST pilot in March of '97 and the 24 technical specification pilot at the end of this year.
25 We will be moving over the next six months, as Tom ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
l l
57 l 1 King mentioned, to complete the IPE reviews. The draft IPE 2 insights report is, in fact, to the Commission, and I guess 3 it will be sent out for public comment shortly.
4 MR. KING: Yes.
5 MR. HOLAHAN: I think we already covered the 6 reliability data rule as an ongoing activity.
7 CHAIRMAN JACKSON: When is that evaluation 8 expected to be completed? When are you going to be 9 completed?
1 10 MR. JORDAN: I will ask Pat Baranowsky to give me 4 l
11 advice.
t l
12 MR. BARANOWSKY: I believe we are going to try to I 1
]
13 have something to the Commission giving the status of that 14 evaluation either late February or early March.
15 CHAIRMAN JACKSON: So March of '977 16 MR. BARANOWSKY: '97, yes.
17 CHAIRMAN JACKSON: Okay.
18 MR. HOLAHAN: Of course, having developed the 19 number of new training programs, we will be continuing to 20 use those.
21 May I have the twentieth slide, please?
22 This is just a summary of our next commitments to 23 the Commission, with a December update and a briefing plan 24 for next April. I think by the time we prepare the December l 25 update, we will have a much clearer view of where we stand l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
o . ,
58 1 with respect to the regulatory guide and the standard review 2 plan because we will have been through the ACRS and CRGR and 3 we will know how close we are to having a version available 4 for public comment.
5 CHAIRMAN JACKSON: Okay, thank you.
6 Commissioner Rogers?
7 COMMISSIONER ROGERS: Just in the December 8 briefing, do you have essentially a topics list for that yet 9 as to what you think you will be discussing?
10 MR. HOLAHAN: I think the current plan is we would 11 have briefing -- you mean the Commission briefing?
12 COMMISSIONER ROGERS: Yes.
13 MR. HOLAHAN: I think the Commission briefings 14 have been set on six-month intervals.
15 So, although we would produce an update of the 16 report, right now there is not a --
17 CHAIRMAN JACKSON: The report comes in three-month 18 intervals. The briefings are six months.
19 COMMISSIONER ROGERS: Right, but just what will be 20 the emphasis of that?
21 MR. HOLAHAN: Just looking at the topics that are 22 ongoing, I would say the regulatory guide and the standard 23 review plans would be the dominant issues.
24 COMMISSIONER ROGERS: When will you be able to 25 talk to us a little bit about how we expect to use PRA in ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
59 1 inspections?
2 MR. HOLAHAN: I think we could do that at almost 3 any stage since it is sort of in the process. We have heard 4 some progress.
5 COMMISSIONER ROGERS: Well, I would 'e very 6 interested in hearing that because the standara review plan 7 is important, but how do we expect to actually employ the 8 use of PRA out in the field, particularly, for instance, 9 with resident inspectors?
10 MR. THADANI: Actually, we have s:arted to move 11 slowly in that direction.
12 CHAIRMAN JACKSON: Why don't you speak to it 13 specifically in your briefing to the Commission.
14 MR. THADANI: We will do that. WE will do that.
15 MR. HOLAHAN: And there are two or three examples, 16 at least two, in the current Commission paper.
17 The one thing that I would say, the thing that I 18 am most optimistic about is the senior reactor analyst 19 program where we have taken about 10 of the experienced 20 senior inspectors largely from the field offices and put 21 them in two-year training programs for PRA, and that looks 22 like it is working very effectively. That is a mechanism 23 for getting experienced inspectors with risk insights, 24 putting them back into the regional offices to be the local 25 experts.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
i 60 1 COMMISSIONER McGAFFIGAN: I got stuck back on page 2 13. I kept going back to it. If the answer to VI-(c) is 3 yes, that the licensee's PRA would be put on the docket and J
! 4 subject to litigation, what are the implications of that for 5 this whole effort?
6 MR. HOLAHAN: Well, at first, it might make the 7 licensees a little bit reluctant to submit those, but I 8 think, in practice, what is likely to happen is not that the 9 whole PRA is subject to litigation any more than every code 10 analysis and every code run shows up in the litigation, but 11 the information that is extracted from it and summarized and 12 is actually in the licensing decision process. Frankly, if 13 that element of the analysis is really what is at least in 14 part convincing the Staff that this is a good regulatory 15 decision to make, then I think it ought to be subject to 16 public scrutiny, 17 CHAIRMAN JACKSON: Commissioner Dieus, any 18 questions?
19 COMMISSIONER DICUS: No.
20 CHAIRMAN JACKSON: Commissioner Diaz?
21 COMMISSIONER DIAZ: Yes. I just have a comment.
22 When I read these documents, I was sure I was confused. Now 1
23 your pages 10 to 13 assure me that I have good cause to be l
24 confused because there are a lot of good questions in there. !
25 However, looking back a little bit over the 1
ANN RILEY & ASSOCIATES, LTD. l Court Reporters l 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034 1
a s 61 1 history, to be able to make an informed risk decision, I 2 think we already realize we have got to take a risk, and I 3 think we took a risk with the maintenance rule. Is that 4 correct? Was it something we did that we weren't sure how 5 it was going to come out and we took a risk? And I think 6 that's great.
7 In all these studies, sometime soon we are going 8 to have to come with another risk. We are going to have to 9 take a risk, and that is going to be an informed risk to i
10 implement risk-informed performance-based decision.
11 Has the Staff identified any particular area where 12 we are closer to a definite answer to say we are going to be 13 able to do this at a definite time? I mean, is any of the 14 multiple series of issues resolved to a point that we can 15 say in a year we can do that or whatever?
16 MR. HOLAHAN: I think it is fair to say that the 17 pilot activities are the areas that we are getting the most 18 experienced, and I would say I think we are pretty 19 optimistic in each of those.
20 Certainly, perhaps with the ICI, in-service 21 inspection area, it requires a bit more technological 22 development than the others, but I think the Staff is 23 optimistic about using these approaches and coming to 24 agreement with the industry on the in-service testing. I 25 think there is very good reason to think that, certainly, ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
1 l
l 62 ;
i 1 within a year we will be ready to deal with technical 2 specification changes, and I think in the graded QA area, we l
3 will come to an understanding of the appropriate uses.
4 One of the difficulties in that area is it is hard ,
i 5 to quantify the value of doing quality assurance, but ,
6 certainly, the use of risk input in deciding what is more or 7 less important equipment in the plant, I am very optimistic ,
8 that all of those will be successes. 1 9 COMMISSIONER DIAZ: Is the critical path dependent l
10 on the database that you have established on how to track 11 that database?
12 MR. HOLAHAN: I guess I don't see the data role as 13 absolutely essential in the sense that nothing can be done 14 without it.
15 COMMISSIONER DIAZ: I see.
16 MR. HOLAHAN: I think it is a very important ;
1 17 element. I think it would reduce the uncertainties. It 18 would make it much more practical to be making decisions.
19 Perhaps you can't make as good and maybe, 20 therefore, you can't make as certain a decision or maybe you 21 need to put in a little extra margin if you don't have as 22 much data, but I am still optimistic that improved decisions 23 can be made.
24 COMMISSIONER DIAZ: Certainly, if the rule is l 25 imposed, we will be able to track it.
l t
ANN RILEY & ASSOCIATES, LTD.
I Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
- _ _ . . _ ~ . ___- -, . .-_ . . . .. _ -.
63 1 MR. HOLAHAN: Yes.
2 MR. THADANI: Yes.
3 If I may expand on what Gary has said, l 4 Commissioner Diaz, actually we at the Agency are using these 5 techniques today in many of our decisions. In fact, we have 6 a regulation that is called the backfit rule that calls for 7 us to make two determinations before we can impose any new 8 generic requirement.
9 The first one is that it should lead to j 10 substantial improvement and safety, and then we do use the 11 Commission's subsidiary objectives derived from the safety 12 goals and do, do risk analysis to see how the issue might 13 relate to risk before we would go back and do cost benefit 14 analysis, which is the second element of the backfit rule.
15 As far as we go forward, we are even today using 16 what I would call risk insights to complement our I 17 deterministic evaluations and some of the changes that the 18 licensees come in and propose in terms of technical 19 specification changes.
20 What we do not do today, we do not have fixed 21 numerical criteria or what I would call the infrastructure, 22 the regulatory guide standard revien plans, but we do use 23 risk insights in these decisions as we go forward. In fact, 24 it is an important element that we need to consider.
25 So I want to be sure that you know that we are ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
64 1 actually using these concepts and these decisions, but we 2 just don't have the fixed criteria and we don't have 3 guidance on how far to review these things called quality 4 methods and so on. It is sort of ad hoc today, I would say.
5 COMMISSIONER DIAZ: I understand.
6 CHAIRMAN JACKSON: I would like to thank the Staft 7 for what has been actually a very informative briefing on 8 the Agency's PRA activities. Thank you.
9 We commend you, in fact, for the progress you have 10 made to date in this sometimes difficult area. I know some 11 of you have lost a few hairs along the way, but at the same 12 time, we encourage you to continue to improve the process 13 and to provide appropriate review mechanisms to ensure that 14 PRA is used appropriately in our regulatory processes.
15 Clearly, PRA has become an important tool of the regulatory 16 process, and therefore, we have to strive to enhance the 17 process, when necessary, but to ensure its consistent use 18 where appropriate, and that is where the development of what 19 you call, Dr. Thadani, the infrastructure is very important.
20 MR. TRADANI: Yes.
21 CHAIRMAN JACKSON: It is also the reason for the 22 fact that I am explicitly asking you to address the 23 implementation and monitoring issues within the context of 24 the maintenance rule and so forth in the next briefing.
25 We, the Commi'ssion, owe you decisions on the ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
l 65 )
1 policy issues as soon as possible; for example, the one we 2 have been discussing, the use of the Commission safety goals 3 for plant-specific applications. l l
4 Then, as long as we understand what is really 5 needed or how far one can go in the use of these I i
6 methodologies for a given decision, then I think we will 7 start out on more solid ground.
l 8 So, unless my fellow Commissioners have any 9 further comments, we are adjourned.
10 [Whereupon, at 3:42 p.m., the briefing concluded.]
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 I
l i ANN RILEY & ASSOCIATES, LTD.
I Court Reporters 1250 I Street, N.W., Suite 300 Washington, D.C. 20005 (202) 842-0034
l . -
l l
l l
CERTIFICATE This is to certify that the attached description of a meeting of the U.S. Nuclear Regulatory Commission entitled:
TITLE OF MEETING: BRIEFING ON PRA IMPLEMENTATION PLAN -
PUBLIC MEETING PLACE OF MEETING: Rockville, Maryland DATE OF MEETING: Wednesday, October 16, 1996 was held as herein appears, is a true and accurate record of the meeting, and that this is the original transcript thereof taken stenographically by me, thereafter reduced to typewriting by me or under the direction of the court reporting company Transcriber: Ot. n5ub LfOn[Cnu
[ Q Reporter: Mark Mahoney I
( }
United States
/ Nuclear Regulatory Commission t
STATUS LPDATE OF PROBABILISTIC RISK ASSESSMENT (PRA)
IMPLEMENTATION PLAN Ashok C. Thadani Gary M. Holahan, Thomas L. King ,
October 16,1996 f
e OVERVIEW
- Background
- Recent Accomplishments ,
- Summary of Key Technical .
and Process Issues
- Policy Issues Recommendations
- Future Activities i
t 1
BACKGROUND t
- Final PRA Policy Statement published on August 16,1995
- January 3,1996, staff submitted action plans for pilot applications and the development of RGs and SRPs
- Staff provides quarterly written updates on the progress of the PRA Implementation Plan and briefs the Commission semi-annually e May 15,1996, SRM requesting staff recommendations on policy issues
- In early October,1996, staff submitted status update for the PRA Implementation Plan and policy issue recommendations 2 'l
KEY ELEMENTS OF THE PRA POLICY STATEMENT
- Increase PRA applications in all regulatory decisions t
- supported by methods and data ~
- complements deterministic approach i
- supports defense-in-depth philosophy ;
r
- PRAs are not substitutes for meeting current rules, regulations and requirements. Current rules and .
regulations shall be complied with unless these rules and regulations are revised
- Safety Goals and subsidiary numerical objectives are to be used for generic requirements 3
INTEGRATION OF PRA INTO REGULATORY ACTIVITIES
- Research Enhance Safety Decisions (e.g.,
PRA -
Prioritize Safety Issues Configuration Severe Accident Control, Accident Assessment Management)
- Efficient Use of
,,, NRC Resources
=== (e.g., IPE
- - Insights, Risk-Based inspect,on) i
- Current Rules and Requirements Tech Specs
- Inservice Testing Reduce Industry Burden (e.g.,
Graded QA,
- Regulatory Oversight Licensing Risk-Based IST)
Inspection
,j Analysis of Experience 4 SALP
PRA IMPLEMENTATION PLAN e Inter-office activities - comprehensive, broad scope, ongoing activities i
e Develop risk-informed Regulatory Guides and Standard Review Plans i
e Industry pilot applications (e.g., inservice inspection and testing, graded quality assurance) e Analysis of operational data including accident sequence precursors e Staff PRA training
- Nuclear waste management s
RECENT ACCOMPLISHMENTS
- Draft RGs and SRPs for internal staff review e Review of industry-initiated pilot applications e Review of Individual Plant Examination (IPED and Individual Plant Examination-External Events (IPEEE) submittals
- Public workshop to discuss the proposed Reliability Data Rule and Draft Regulatory Guide e in parallel, industry is providing sample data to demonstrate a proposed voluntary alternative to the Reliability Data Rule a Completed PRA training guidance document and "PRA for Technical Managers" course
~
6
.[ '
RGs AND SRPs DEVELOPMsNT J Inter-office teams established to develop h e
General RG and SRP Application-specific RGs and SRPs
- Inservice testing i:IST:i
- Inservice inspection 1:ISI)
- Graded Quality Assurance I:GQA)
- Technical Specifications i:TS:l 1
I e Interactions with the industry and the ACRS
under staff review ;
)
e Draft General SRP and IST, ISI SRPs are currently under staff review ,
, 7 l
- . i l
Principal Licensee Steps in Risk-Informed, l '
i Plant-Specific Regulatory Process i t
l l
t l
! i f i f i f Step 5:
Step 2: Step 3: Step 4: Step 6:
I Conduct Int rated Conduct Develo isionma ing: Documen '
Define the Conduct
+ + Implemenb + +
l Proposed
+ Deterministic Probabilistic
& Monitoring Determine and Subm Engineering Engineering Acceptability of Request Chnge Evaluations Evaluations Strategies Proposed Change r
>L J L h 6 l i
_______._-_____.___-,,,-,g, _ - _ _ - - . _ _ _ _ __ _ __ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ - _ _ . . _ . - _ _ _ , . _ _ . - _ _ - , . _ _ _ . . - - - - - _ _ - . _ _ _ _ . _
Summary of Key Technical and Process Issues I) Issues Associated with Defm' ition of Proposed Change:
a) What information does the licensee need to submit to characterize the change?
b) Should the proposed change be required to meet at least one of the three goals of the PRA Policy Statement?
II) Issues Associated with Deterministic Evaluation:
a) What deterministic evaluations are required?
b) What are the acceptance guidelines for the i
deterministic evaluation?
9 i
HI) Issues Associated with Risk Evaluation:
a) What determines the extent to which risk analysis can be used?
b) What determines the required quality of the risk analysis?
c) How is the appropriate quality assured?
d) How is uncertainty to be addressed?
e) How are cumulative changes in risk accounted for?
f) Should the acceptance guidelines be based upon total plant risk?
g) How should the acceptance guidelines be structured?
h) What is the role of importance analysis?
i) Should the acceptance guidelines apply to proposed changes ,
individually or as a package?
10
~
IV) Issues Associated with Implementation and Monitoring:
a) What are the appropriate performance characteristics to monitor?
b) How should the SSCs to be monitored be selected?
c) How should the SSC performance be monitored?
d? How will feedback from the monitoring be used to make adjustments in implementation?
s 11
l V) Issues Associated with Integrated Decision Making:
a) What are the important factors in integrating deterministic and probabilistic considerations?
b) How are uncertainties to be treated?
c) To what extent should.the existing degree of defense-in-depth be maintained?
d) To what extent should the existing margins of safety be maintained?
e) What should defense-in-depth be based on?
f) What is the role of an expert panel? -
g) What is the role of 10CFR50.109?
12
VI) Issues Associated with Documentation and Submittal:
a) What documentation is to be submitted? .
b) What level of detail of risk information should be submitted?
c) Will explicit use of risk information in plant specific regulatory decisions require the licensee's PRA to be put on the docket and subject to litigation?
i
. 13
?
I POLICY ISSUES l
- Role of " performance-based regulation" in the PRA
! Implementation Plan l
l
- Plant-specific application of Safety Goals
- Risk neutral vs. increases in risk
4
POLICY ISSUES 3
(
- Issue: The role of performance-based regulation in the PRA Implementation Plan
,
- Options: ,
l I
l 1) Implement performance-based regulation in the context of the current l PRA Implementation Plan through the current process
- 2) Implement performance-based regulation as an explicit element of the l i PRA implementation Plan by actively soliciting industry initiatives
- 3) Implement performance-based regulation outside the context of the PRA Implementation Plan
- Staff Recommendation: Option 1
- ACRS letter dated 8/15/96: "We agree with the staff that, where practical, performance-based strategies should be included in the implementation and monitoring step of the risk-informed decision-making process."
, 15 .
i
- POLICY ISSUES (Continued) i e Issue: Should the Commission's Safety Goals and subsidiary objectives be referenced or used to derive guidelines for plant-specific applications and, if so, how?
e Options:
- 1) Develop guidelines for plant-specific decisions that are derived from the Commission's current Safety Goals and subsidiary objectives
- 2) Relate plant-specific risk changes to industry average population goals e Staff Recommendation:
Option 1 e ACRS letter dated 8/15/96: "We believe the safety goals and subsidiary objectives can and should be used to derive guidelines for plant-specific applications."
~'
. 16
- i POLICY ISSUES (Continued) ,
e Issue: Should requested changes to the current licensing basis be risk-neutral or should risk increases be permitted?
l.
e Options:
- 1) Small increases in risk be allowed under certain conditions
- 2) Only risk neutral changes are permitted
- Staff Recommendation: ,
Option 1
- ACRS letter dated 3/15/96: "We agree with the staff and industry that .
increases in risk should be permitted in some situations."
17 i
[ --..-
POLICY ISSUES (Continued)
- Options:
- 1) Approve requested changes to risk-informed ISI and IST requirements as exemptions to current regulations
- 2) Approve requested changes to risk-informed ISI and IST requirements as authorized alternatives under 50.55a (a)(3)(i)
- 3) Defer approval of requested changes to risk-informed ISI and IST requirements until after completion of ASME consensus process and modification of 10 CFR 50.55a i e Staff Recommendation:
Option 2 e ACRS did not offer a view on this issue i
. 18 -l
FUTURE ACTIVITIES (NEXT 6 MONTHS) ,
i
- ACRS full and subcommittee meetings
- Continue review of the pilot applications (staff recommendation to the Commission regarding IST and technical specification pilots)
- Complete evaluation of industry's proposed voluntary approach to reliability data submittal l
- Continue staff PRA training !
19 l
FUTURE ACTIVITIES (NEXT 6 MONTHS, Continued)
- l.
- PRA Implementation Plan updates to the Commission in December 1996 e PRA Implementation Plan briefing for the Commission in April 1997 20 '
s- . 4 -an a ** .m mh wm -* . - , _ _ -aa-....s-C+6+ .4 a E.4 m - N * --. Ee-- -- M- _. MA~-*4h 4-A J84 Am. -4a.2AA---. - - -**. --rw a--A4J a 3i88-*---AA.--m-l e s J
1 l
1
)
7
(
1 1 l
< l 1
2 1
4 l
l I l
i
! M -
i N
i R \
4 i
u \
4 A 1 i
i i
d 4
i d
! i l
- 1 1
a i
4 l
r .
PILOT APPLICATIONS e Inservice testing (IST) l
- Reviewed Palo Verde and Commanche Peak's j responses to Request for AdditionalInformation (RAI)
- Additional information from licensees needed to complete staff evaluation
- Staff recommendation to the Commission delayed i
until 3/97 e Inservice inspection (ISI)
- Reviewed initial submittal from Surry
- ANO-2 and Fitzpatrick submittals delayed until 10/96
- Reviewed risk-informed ISI guidance from industry l i
22
PILOT APPLICATIONS (Continued) L e Graded QA
- Evaluated graded QA programs from South Texas, Palo Verde and Grand Gulf
- Incorporated elements of the pilot methodology in the risk-informed graded QA RG and SRP
- Slower progress than planned due to staff's higher priority on RG/SRP development and licensee's higher priority on other plant tasks
- Staff recommendation to the Commission delayed until 6/97 23
PILOT APPLICATIONS (Continued) e Technical Specifications
- Reviewed Combustion Engineering Owners' Group (CEOG) request to extend allowed outage time (AOT) for LPSI and SITS
- Staff awaits response to third RAI from lead plant
- Staff recommendation to the Commission delayed until 12/96 l '
,i
, 24
- I
IPE/IPEEE REVIEWS AND INSIGHTS ,.
I e Draft NUREG-1560 to the Commission in 9/96 providing perspectives from reviewing 75 IPEs ,
i e Briefing regions on plant-specific IPE results and insights e IPEEE review is continuing 1
i 25
IPE VS RISK-INFORMED REGULATION
- Current IPE results do not by themselves always provide I a complete basis for supporting risk-informed regulatory decisions
- NRC will require PRA quality commensurate with the proposed application
- NRC is working with the industry to develop standards for using PRAs in regulatory applications 26 .!
RISK INFORMED REGULATION L l
s
-+ -+ Plant w Severe Accident GL-88-20 IPE Specific Vulnerabilities PRA a ,
IPE "
Program' Limited Scope IPE/PRA NRC Review Insights k es es es
~ es t
~
9F Licensee : Potential Pilot Application Proposed PRA % +
Risk m
' Applications implemented Applications Upgrade informed, a Regula- ,
tion '
- i Application ;
Guidelines ;
h
RELIABILITY DATA RULE e Staff is resolving comments for the proposed rule and the draft Regulatory Guide
- e Conducted a public workshop in June 1996 e Parallel effort to obtain necessary data
through an industry voluntary program as well as through rule making i
l
- Expect to complete an evaluation of industry's voluntary proposal in about 6 months e Once a decision is made on the voluntary approach, issuance of the final rule would proceed if appropriate
. 27 '!
a f
PRA STAFF TRAINING ,
,'
- Issued NUREG/BR-0228, " Guidance for Professional Development of NRC staff in Regulatory Risk Analysis"
- Staff has completed the "PRA for Technical Managers" i course development and conducted a successful dry run
- Turnover in the Senior Reactor Analyst program due to promotions; training and rotational assignments near completion i
i i
t 28
REVISIONS TO PRA IMPLEMENTATION PLAN e Pilot application target dates delayed due to
- licensee's delay in responding to RAIs
- staff effort focused on RGs and SRPs development j l
e Tasks to incorporate risk insights into inspection programs delayed due to resource diversion to other t
critical short term activities l
l e Implementation of the revised operator licensing l
examination standards incorporating risk insights contingent upon Commission approval i
( e Task 1.10 added to use IPE insights to identify l appropriate staff and industry followup actions and to
( track regulatory use of IPE/IPEEE results :
i 29 <l i