ML21055A742

Transcript of the Advisory Committee on Reactor Safeguard 682nd Full Committee Meeting - February 4, 2021, Pages 1-278 (Open)

ML21055A742
Person / Time
Issue date:	02/04/2021
From:	Advisory Committee on Reactor Safeguards
To:
Burkhart, L, ACRS
References
NRC-1369
Download: ML21055A742 (442)

Text

Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION

Title:

Advisory Committee on Reactor Safeguards Docket Number: (n/a)

Location: teleconference Date: Thursday, February 4, 2021 Work Order No.: NRC-1369 Pages 1-278 NEAL R. GROSS AND CO., INC.

Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.

Washington, D.C. 20005 (202) 234-4433

1 1

2 3

4 DISCLAIMER 5

6 7 UNITED STATES NUCLEAR REGULATORY COMMISSIONS 8 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 9

10 11 The contents of this transcript of the 12 proceeding of the United States Nuclear Regulatory 13 Commission Advisory Committee on Reactor Safeguards, 14 as reported herein, is a record of the discussions 15 recorded at the meeting.

16 17 This transcript has not been reviewed, 18 corrected, and edited, and it may contain 19 inaccuracies.

20 21 22 23 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com

1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 + + + + +

4 682ND MEETING 5 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 6 (ACRS) 7 + + + + +

8 OPEN SESSION 9 + + + + +

10 THURSDAY 11 FEBRUARY 4, 2021 12 The Advisory Committee met via Video 13 Teleconference, at 9:30 a.m. EST, Matthew W. Sunseri, 14 Chairman, presiding.

15 COMMITTEE MEMBERS:

16 MATTHEW W. SUNSERI, Chairman 17 JOY L. REMPE, Vice Chairman 18 WALTER L. KIRCHNER, Member-at-large 19 RONALD G. BALLINGER, Member 20 DENNIS BLEY, Member 21 CHARLES H. BROWN, JR. Member 22 VESNA B. DIMITRIJEVIC, Member 23 JOSE MARCH-LEUBA, Member 24 DAVID A. PETTI, Member 25 PETER RICCARDELLA, Member NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

2 1 ACRS CONSULTANT:

2 MICHAEL CORRADINI 3 STEVEN SCHULTZ 4

5 DESIGNATED FEDERAL OFFICIAL:

6 CHRISTIANA LUI 7 QUYNH NGUYEN 8

9 ALSO PRESENT:

10 DON ALGAMA, RES 11 KENNETH ARMSTRONG, RES 12 JOE ASHCRAFT, NRR 13 MICHELLE BALES, RES 14 ANDREW BARTO, NMSS 15 SUSAN COOPER, RES 16 JAMES CORSON, RES 17 ERIC FOCHT, RES 18 RAY FURSTENAU, RES 19 CHARLES HECK, GE-Hitachi 20 DAVID HINDS, GE-Hitachi 21 MATT HISER, RES 22 COLIN JUDGE 23 JEFF JULIUS, Public Participant 24 NECDET KURUL, GE-Hitachi 25 LUCAS KYRIAZIDIS, RES NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

3 1 CHANG LI, NRR 2 LOUISE LUND, RES 3 WAYNE MARQUINO, Public Participant 4 WILLIAM McCAUGHEY, DOE 5 SCOTT MOORE, Executive Director, ACRS 6 SEAN PETERS, RES 7 DAVID RAHN, NRR 8 PAUL REBSTOCK, RES 9 JAMES SHEA, NRR 10 JOHN STETKAR, Public Participant 11 DINESH TANEJA, NRR 12 JOHN TOMON, RES 13 DANIEL WACHS, Idaho National Laboratory 14 GEORGE WADKINS, GE-Hitachi 15 KIMBERLY WEBBER, RES 16 FROSTIE WHITE, GE-Hitachi 17 JING XING, RES 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

4 1 C O N T E N T S 2 Roll Call . . . . . . . . . . . . . . . . . . . . 5 3 Agenda and Opening Remarks . . . . . . . . . . . 6 4 IDHEAS-G: An Integrated Human Events 5 Analysis System - General Methodology: . . . . . 13 6 Advanced Reactor Computer Codes Vol. 4 and 5 . 127 7 Post-Halden Plans . . . . . . . . . . . . . . . 188 8 Adjournment . . . . . . . . . . . . . . . . . . 278 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

5 1 P R O C E E D I N G S 2 9:30 a.m.

3 CHAIR SUNSERI: Okay. It's 9:30. The 4 meeting will now come order. This is the second day 5 of the 682nd meeting of the Advisory Committee on 6 Reactor Safeguards. I'm Matthew Sunseri, the Chair of 7 the ACRS. I will now call the roll to verify quorum 8 and that clear communications exist. Ron Ballinger?

9 MEMBER BALLINGER: Here.

10 CHAIR SUNSERI: Dennis Bley?

11 MEMBER BLEY: Here.

12 CHAIR SUNSERI: Charles Brown?

13 MEMBER BROWN: Here.

14 CHAIR SUNSERI: Vesna Dimitrijevic?

15 MEMBER DIMITRIJEVIC: Here.

16 CHAIR SUNSERI: Walt Kirchner?

17 MEMBER KIRCHNER: Here.

18 CHAIR SUNSERI: Jose March-Leuba?

19 MEMBER MARCH-LEUBA: Here.

20 CHAIR SUNSERI: Dave Petti?

21 MEMBER PETTI: Here.

22 CHAIR SUNSERI: Joy Rempe?

23 VICE CHAIR REMPE: Here.

24 CHAIR SUNSERI: Pete Riccardella?

25 MEMBER RICCARDELLA: I'm here.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

6 1 CHAIR SUNSERI: And myself. I know we 2 have quorum. The designated federal officer for this 3 meeting is Ms. Christiana Lui. During today's 4 meeting, the Committee will consider the following 5 presentation oral report on IDHEAS-G: An Integrated 6 Human Events Analysis System - General Methodology, 7 the presentation and letter report on Advanced Reactor 8 Computer Codes, Volumes 4 and 5, and info briefing on 9 Post-Halden Plans.

10 A phone bridge line has been opened to 11 allow members of the public to listen on presentations 12 and Committee discussions. We have received no 13 written comments or requests to make oral statements 14 from members of the public regarding today's session.

15 There will be an opportunity for public comment, and 16 we have set aside time in the agenda for comments from 17 members of the public listening in or members of the 18 public attending or listening in to our meeting.

19 Written comments may be forwarded to Ms.

20 Christiana Lui, the designated federal officer. A 21 transcript of the open portion of the meeting is being 22 kept. And it is requested that speakers identify 23 themselves and speak with sufficient clarity and 24 volume so that they may readily be heard.

25 Additionally, participants should mute themselves when NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

7 1 not speaking.

2 So I have a few opening remarks here. Let 3 me get myself situated here. Okay. I'm going to turn 4 on this camera for a few minutes. I just wanted to 5 call attention today that Christiana Lui was a long-6 time member of the ACRS staff and an even longer time 7 employee of the NRC. During her time with the ACRS, 8 she has been involved in a number of areas, most 9 recently supporting members with the review of PRA 10 related activities.

11 When the pandemic restrictions on 12 gathering came down, Christiana took on an additional 13 assignment, along with a few other members of this 14 ACRS staff, to develop the protocols and implement the 15 technology to allow the Committee to hold its meetings 16 virtually. This took extraordinary effort and was a 17 resounding success.

18 On an individual level, Christiana has 19 shared her knowledge and experience with me on several 20 topics, allowing me to make better decisions and be a 21 better member. Christiana has gotten an opportunity 22 to further advance her career and will be taking on a 23 position in research. So today is Christiana's last 24 full Committee meeting as an ACRS staff member. I 25 look forward to a future full Committee meeting where NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

8 1 we might get to work with her in a different role.

2 So, Christiana, on behalf of the ACRS 3 members thank you for your support to us and our 4 mission. At this time, I would like to call on the 5 members to see if anybody has any additional comments 6 or anything they would like to say about Christiana.

7 MEMBER DIMITRIJEVIC: Hi. This is Vesna.

8 Yes. I was really blessed to be welcomed by 9 Christiana when coming to ACRS. And I don't really 10 know what kind of member I will be, but she was an 11 incredible support and very knowledgeable. And 12 without her, I told her I will feel like half of the 13 member. Christiana didn't do anything half-heartedly 14 so whoever works with her will be blessed. And I feel 15 blessed that I worked with her.

16 CHAIR SUNSERI: Thank you, Vesna. Anybody 17 else?

18 MEMBER BALLINGER: Pretty short and sweet.

19 Christiana is just flat out a nice lady.

20 CHAIR SUNSERI: Thank you. Others?

21 VICE CHAIR REMPE: Sure. This is Joy.

22 And I also wanted to say thanks, Christiana. I've 23 known you since you were a project manager over some 24 research I was doing back at INL many, many years ago.

25 But I have enjoyed my interactions with you over the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

9 1 years, and I was really happy when you joined us at 2 ACRS and best of luck in your new assignment.

3 MEMBER BLEY: Yes. This is Dennis. I've 4 known Chris a long time as well, even before she was 5 busy here. And I wish her great success, and I know 6 she'll be going back more to her roots over in 7 research. They'll be lucky to have her. The one thing 8 others haven't said, it's great working with Chris.

9 But she also really makes sure we are well prepared 10 and have thought things out thoroughly. And she's 11 been great help all the way through. Thanks, Chris.

12 CHAIR SUNSERI: Thank you, Dennis. Anyone 13 else?

14 MEMBER BROWN: Yes. This is Charlie.

15 Back about a year ago was when we started struggling 16 in this pandemic routine in our remote meetings. She 17 recognized my limitations as the Neanderthal-18 troglodyte as I struggled to be able to get my 19 computer to work. So I had several calls with her and 20 many interactions as we started trying to get it up.

21 And she did a marvelous job with her patience in 22 helping me out. I much appreciated that. It's a 23 testament to her quality. Thanks, Christiana.

24 CHAIR SUNSERI: She does have a lot of 25 patience.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

10 1 MEMBER BROWN: That's an understatement 2 when it comes to me and computers.

3 CHAIR SUNSERI: The same applies to me, 4 too. She's been very helpful in that regard. Anybody 5 else?

6 MEMBER KIRCHNER: I would just like to add 7 that thank you, Chris, for your thoroughness and your 8 professionalism on many, many different matters and 9 certainly, as Charlie said, getting us computer 10 literate in this new world that we're working in. So 11 thank you ever so much.

12 CHAIR SUNSERI: Thanks, Walt.

13 MEMBER MARCH-LEUBA: I also would like to 14 say good luck in research, Christiana. We all like 15 you and wish you great progress there.

16 CHAIR SUNSERI: Thank you, Jose. Anyone 17 else?

18 MEMBER BROWN: I'll make one other 19 observation. I think she's going to get bored in 20 research. There's too much excitement over here.

21 That's a positive comment by the way. Take care, 22 Christiana.

23 CHAIR SUNSERI: Okay. Last call. Well, 24 I think, you know, this virtual restriction, I think 25 this is as good of a sendoff as we can give you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

11 1 virtually. I wish we were in person where we could 2 really do a proper sendoff. But I think from the 3 members' comments you heard, we really appreciate what 4 you've done for us, and we'll miss you. So I'm going 5 to put you on the spot and say is there anything that 6 you would like to add before we get started with the 7 meeting?

8 MS. LUI: Well, I really appreciate the 9 feedback. And I really enjoyed with all the members 10 and also the ACRS staff during the time I've been with 11 ACRS. And I do want to highlight that the success is 12 not by running the virtual. We all worked together as 13 the team to make the transition as easy as possible, 14 and we all play our part. Because I was the lead so 15 I get a lot of the credit, but I really want to make 16 sure that the whole team gets recognized for the 17 transition because without their effort, I don't 18 believe that this would have been as smooth as 19 possible. So given that I have some brownie points 20 with the members, next time when I return to make 21 presentation in front of the ACRS, I will expect to 22 get some passes when I get there. Hopefully, that 23 will be the case.

24 CHAIR SUNSERI: Well, that's a big ask.

25 I don't know about that one. But we'll take it under NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

12 1 advisement.

2 MS. LUI: I do have one thing to add, 3 Matt. We did receive a request to make comments from 4 Mr. John Stetkar at the end of the IDHEAS presentation 5 today. So we will be switching on the public bridge 6 line to at least accommodate Mr. John Stetkar's 7 request.

8 CHAIR SUNSERI: Okay. All right.

9 Understood. That request probably came in after I put 10 my remarks together. So thanks for the heads-up on 11 that. Okay. Well, thank you, Chris. I do have one 12 other announcement on the list right now. So I'm 13 going to call on Peter Riccardella.

14 MEMBER RICCARDELLA: Yes, hi. On a sadder 15 note, some of you may have heard that Keith Whitman 16 passed away recently. Keith was a long-term NRC 17 material guy. Perhaps you'll remember the silver hair 18 and the big handle bar mustache. I think he retired 19 from the NRC about 10 years ago and has been doing 20 some consulting. But unfortunately he had a fall 21 about four years ago and was in a nursing home and 22 contracted COVID there. So, anyway, for those of you 23 who knew Keith, I'm sure you'll recognize that he'll 24 be sadly missed.

25 CHAIR SUNSERI: Thank you, Pete. All NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

13 1 right. So any members have any comments or questions 2 about our agenda for today? We will proceed on then 3 with the IDHEAS-G Integrated Human Events Analysis 4 System General Methodology presentation. So, Dennis, 5 I'll turn it over to you.

6 MEMBER BLEY: Thank you, Mr. Chairman.

7 Before I turn it over to Sean Peters and Jing Xing, 8 I'd like to make a few introductory remarks. We have 9 a fairly long period today to let Sean and Jing go 10 through their presentations and then a fairly thorough 11 outline of their talk. So it's going to cover 12 everything from the original issuance of the SRM back 13 15 years ago. But I had mentioned the work that led 14 to this started even 10 years before that. So it's 15 been a long evolution. There's been a lot of work 16 along the way, a lot of -- some missteps and a lot of 17 cleaning things up.

18 Before I give it to Sean to -- I want to 19 apologize for something I couldn't help. But I missed 20 the September meeting and hadn't had a chance to fully 21 prepare for that but I wasn't able to be there. If I 22 had, some of the comments you will hear today you 23 probably would have heard some time back. In the last 24 three weeks, I've had a chance to really dig into the 25 five reports they sent us. No, actually the one NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

14 1 before that says three reports. And I have a few 2 comments that will come up as you go ahead, Sean.

3 Sean mentioned in their FLEX expert 4 elicitation report a white paper that wasn't sent to 5 us, but it is available publicly on expert 6 elicitation. And I would commend that to all the 7 members. There's a lot of good information there. At 8 this point, I'm happy to turn this over to Sean. We 9 look forward to your presentation.

10 MR. PETERS: Yes. Thank you, Dennis.

11 This is Sean Peters, branch chief of Human Factors and 12 Reliability Branch in the Office of Research. And I 13 wanted to jump in also. I'd like to also thank Chris 14 Lui for all of the work that she's done. She is one 15 of the main drivers for the IDHEAS program back when 16 she was my division director in the Office of 17 Research. And she also was the one that hired me into 18 this branch chief job and got me into the IDHEAS 19 program. So I really appreciate that.

20 It will be nice to have her back because 21 I think she had a lot of great ideas. And I think she 22 will be able to help guide us a little bit more from 23 the Office of Research than she was from ACRS. So I'm 24 really happy to have Chris back. And I'd just like to 25 also thank the members of the ACRS. I know, as Dennis NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

15 1 said, that we've been going on for a very long time on 2 this program and have had -- and you'll see as we go 3 through the presentation how much progress we made.

4 And I do want to apologize a little bit in 5 advance because you're going to see a lot of the same 6 presentation here from me that you saw back in the 7 subcommittee meetings. And the main reason is because 8 we've had some members who weren't in that 9 subcommittee meeting who are here today. So I wanted 10 to give them an outline of the IDHEAS program. And 11 then after I talk a little bit about the outline of 12 the program, Jing Xing is going to come, Dr. Jing Xing 13 is going to come in, and present on some more of the 14 details of each of the pieces of the program.

15 And then I'm going to have a really short 16 wrap-up, kind of what we see as the future of the 17 IDHEAS program on the back end of that. So I'm just 18 going to go ahead and proceed to our slides. So the 19 reason why we're here was back in 2006, Dr. George 20 Apostolakis, as a member of the ACRS, he convinced the 21 Commission to write a one sentence SRM on HRA.

22 And when I've had discussions with Dr.

23 Apostolakis recently and kind of picked his brain 24 about why we went down this path, and what Dr.

25 Apostolakis was saying was that he was concerned NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

16 1 associated with misapplication of HRA methods. That 2 there were a ton of methods out there that were built 3 for very specific purposes. And people started using 4 these methods beyond that original intent. And he 5 wasn't certain that they were built for that.

6 So what he wanted the staff to do and 7 wanted the Committee to do was look at these various 8 methods and determine which ones should be used in 9 which circumstances or maybe develop one or propose of 10 one that should be used for all circumstances from the 11 NRC's perspective so the NRC should be using one or a 12 set of methods. So that was kind of the driver for 13 where we went down with the IDHEAS program. And I'll 14 show you a little bit of a timeline on it. So I'm 15 going to talk more about these little bullets here.

16 But as Dennis was saying, we had already started 17 several years before the SRM came out. The SRM is in 18 red on the bottom of the screen there.

19 And we had evaluated methods versus the 20 best practices. But post-SRM, we began looking at, 21 okay, let's do an evaluation for our methods. So we 22 developed an international human reliability analysis 23 study. We worked and also did a U.S. empirical study.

24 I'll talk a little bit more about that on the back 25 end. We developed a technical basis, an HRA cognitive NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

17 1 basis report. We also developed IDHEAS At-Power. And 2 then IDHEAS-G, IDHEAS-ECA, we performed expert 3 elicitations on FLEX and also did FLEX HRA using 4 IDHEAS-ECA to calculate the HEPs.

5 So if you look at that, all the items in 6 red are the products that we've developed since that 7 SRM came out. So I just breezed by a couple slides 8 that were associated with -- the references associated 9 with the report. I just wanted to keep those in there 10 for the public record. But now I'm on Slide 6. The 11 IDHEAS development process -- you know, after the SRM 12 was developed, we reached out with the Halden Reactor 13 Project and the teams of international researchers.

14 And we performed experiments on international 15 operators at the Halden Reactor Project. And we had 16 teams of HRA analysts come in and try to predict the 17 performance of those operators.

18 And we used that as a way to compare the 19 methods versus, you know, operational performance.

20 And after that, we had some questions from the ACRS 21 associated with that. And when we had presentations 22 of the ACRS on our results and the questions, some of 23 them came up, some of the big ones came up, is that 24 these are like Swedish operators in a French digital 25 simulated plant at the Halden Reactor Project. How NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

18 1 would this actually work if we were in the United 2 States on a more, like, knobs and dials type scenario 3 like we have in the United States?

4 So we followed-up that international 5 experiment with a U.S. experiment. We ran our U.S.

6 operators through their own simulator. And we also 7 saw this opportunity as a way to narrow our focus on 8 the HRA methods that were more of interest to the NRC.

9 And what that allowed us to do was have multiple teams 10 use one method and then multiple teams use another 11 method. And so what we were able to do was actually 12 see not just method to method variability but also 13 team to team or analyst to analyst going through those 14 methods.

15 So based upon that, we found that, you 16 know, there wasn't one method that we thought that was 17 ideal for all situations, that they all had strengths 18 and weaknesses. And we decided to try to take those 19 strengths of those methods and then incorporate them 20 into one method, so basically try to negate the 21 weaknesses of some of the methods and try to go 22 basically for the strengths. And so one of the things 23 that we determined in that review was that we needed 24 to develop a new updated cognitive basis for the 25 methodology.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

19 1 We performed a very extensive literature 2 review and developed a scientific basis and structure 3 for HRA. This came out as NUREG-2114. And we 4 presented to the ACRS. And the ACRS, I believe this 5 was about 8 or 10 years ago, that was very high 6 accolades from the Committee members on the cognitive 7 basis report. So we used that as our basis, and we 8 worked with industry because our SRM indicated that we 9 needed to work with industry to develop a methodology 10 for the Agency to use. And so we came up with IDHEAS 11 At-Power. And more of the main issues we tried to 12 address in IDHEAS At-Power was this variability 13 between analysts and between methods.

14 So we worked really hard on that, the 15 IDHEAS At-Power methodology. But in the middle of the 16 development of that method, the Fukushima event hit.

17 And that got the focus of HRA to change at the Agency.

18 The Agency was no longer just concerned with internal 19 events, at-power applications. They were controlled 20 with ex-control room, things that are out there that 21 may have environmental effects. And a lot of other 22 methodologies really didn't consider environmental 23 effects because almost all of them were built for 24 those in control room activities.

25 I'm going to the next slide. So we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

20 1 decided we needed to take a more human centered 2 approach. And we also needed to have a little bit of 3 a broader concept from development that we couldn't 4 just look at, you know, control room activities. We 5 needed to look at things outside. And we also needed 6 to kind of imagine what the future can be for the NRC.

7 We know we have spent fuel storage and 8 transportation. We know we have long-term waste 9 disposal. We know we have, you know, mining 10 operations, well logging. We have medical 11 applications. And so we developed a general framework 12 that is human centered. So what it does is it allows 13 us to look out at those various frameworks and select 14 factors that would influence performance in those 15 domains.

16 So IDHEAS-G, we began that development 17 process. And this is the overall guidance document 18 for how to build those specific methods for those 19 domains. It gave us a framework to generalize and 20 integrate human error data into our program. And also 21 it's a structure that can be used not just for HRA but 22 can also be used for analyzing human events and 23 looking at root causes and human failures.

24 So it's a very general -- we have this 25 general framework and scope that we're using to assess NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

21 1 human reliability here and in the future. And so from 2 that, looking at the majority of applications at the 3 NRC, most of what we utilize is we utilize SPAR-H as 4 a tool for significance determination process and 5 accident analysis. And then, like, in the Office of 6 Research, we do some development outside of that 7 significance determination process and accident 8 analysis framework. We do some development outside of 9 that. But most of the brunt of the work we do is with 10 that SPAR-H in those domains.

11 And so we began to develop IDHEAS-ECA to 12 replace SPAR-H and to think of it as a way to replace 13 SPAR-H to give it a broader breadth, to give it the 14 ability to calculate in domains outside of the control 15 room. So we built the IDHEAS-ECA from IDHEAS-G, and 16 we built it to handle all NRC applications. And when 17 I say all applications, this includes medical events.

18 It includes spent fuel transportation. So we think 19 we've included in IDHEAS-ECA all of the relevant 20 influencing factors that can be used throughout 21 domains.

22 As I said in the second bullet there under 23 IDHEAS-ECA, it can be used for in and ex-control room 24 activities and other nuclear, non-nuclear domain 25 because it's human centered. And the nice thing about NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

22 1 a human centered methodology is when you implement new 2 technologies or new procedures or new control concepts 3 that you still are looking at it as how did that 4 affect the human? So what it does is it allows us to 5 have a method that's not outdated when those new 6 technologies come in. And one of the best-selling 7 features of IDHEAS-ECA is we integrated the 8 quantification model into our software tool. And that 9 software tool is a very easy to use tool, and it has 10 high accolades from our users.

11 So next slide. And then on the -- I think 12 this is my last slide. But the other thing that we've 13 done with IDHEAS is we have a data structure that ties 14 into our existing data collection activities. So this 15 is a strong database for the IDHEAS quantification.

16 It's constantly evolving as we collect information 17 through our SACADA Program. And we presented on the 18 SACADA Program in the past to the ACRS. And we would 19 be happy to present again in the future.

20 We do collect data out of our human form 21 assessment facility that is rolling into the IDHEAS 22 Program. And we also collect a lot of data with the 23 Halden Reactor Project that we rolled into that.

24 There are other sources of data and Jing will talk a 25 little bit about that in her upcoming slides.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

23 1 I'm going to double-check to make sure 2 this is my last slide. Oh no, one more here. So just 3 giving you a brief overview of the development of 4 IDHEAS-G. We've had multiple ACRS subcommittee 5 reviews of IDHEAS-G. My best guess is we're averaging 6 about one ACRS subcommittee per year through the 7 development process. And we've had multiple rounds of 8 comments from ACRS subcommittee members both current 9 and former members.

10 With the three formal external peer 11 reviews, we had two extensive internal peer reviews.

12 And we used IDHEAS-G. In the development process, we 13 kept testing it on things like Fukushima, our U.S.

14 benchmarking events. And we also used it to help do 15 some fuel cycle facility analysis for user offices in 16 NMSS. I did DCA. We used it on various FLEX scenarios 17 in the NRC studies, which we published and you guys 18 got to review. The industry also developed their own 19 studies using IDHEAS-ECA. And they presented on those 20 at the last subcommittee meeting. And they we will be 21 presenting on them in the upcoming RIC in March.

22 So we should be able to see a little bit 23 more from the industry on what they were able to do 24 with the IDHEAS-ECA. And we got very high accolades 25 from both industry and our internal users on those.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

24 1 And we are currently using IDHEAS-ECA to analyze AST 2 and STP events, as this is rolling in to become more 3 of a common practice at the NRC to use the 4 methodology.

5 And in April of this year, we're going to 6 be having a public meeting to officially take in user 7 comments on the document. We're open at all times to 8 user comments. But at that April meeting, it will be 9 kind of a driver for the industry and for users of the 10 methodology to provide comments to us. And we're 11 going to be -- and our plans are with IDHEAS-ECA right 12 now, it is published as a rule for use. We plan to 13 take those comments and incorporate them into a NUREG 14 report.

15 And the last major product, we have a 16 draft IDHEAS Data Report that is out there publicly 17 available. It is currently being reviewed. So we 18 have a contract with Pacific Northwest National 19 Laboratories. They're doing an extensive data review 20 looking at the structural report and how we 21 incorporated the literature into it for accuracy 22 purposes and for recommendations for improvements.

23 And we also, as we continue to collect 24 data through our program, we plan to have regular 25 updates of the IDHEAS Data Report. And so that is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

25 1 correct. That is my last slide. Are there any 2 questions before I pass this over to Dr. Jing Xing?

3 MEMBER BLEY: Yes, Sean?

4 MR. PETERS: Yes.

5 MEMBER BLEY: This is Dennis. You went 6 through a great number of peer reviews, internal and 7 external. And I would comment that IDHEAS-G, the 8 document on IDHEAS-G has really moved from a lot of 9 almost scattered ideas into a very coherent 10 presentation of some very extensive work. Are there 11 any -- well, are you putting any published summaries 12 of the comments you've received on all of these items, 13 G, ECA and DATA?

14 MR. PETERS: That's a great question. I 15 haven't really -- we're so much in the development 16 process, we haven't really considered that internally.

17 I definitely am open to that idea to publish the 18 comments that we have gotten. Jing, do you have any 19 thoughts on this?

20 MEMBER BLEY: I know you've been running 21 like crazy. So it's harder to keep that. But --

22 MR. PETERS: Yes. We --

23 MEMBER BLEY: -- it's very useful 24 information that's been codified in those reviews.

25 MR. PETERS: Yes. Jing, do you have any NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

26 1 thoughts on that?

2 DR. XING: Yes. I think that's a great 3 suggestion. Actually, a couple weeks ago when Theresa 4 asked for some reviewer's comment, I went through the 5 old files. I, myself, was very much impressed again 6 how many comments we saved and how helpful those were.

7 So I think it will be very useful to document them.

8 And there are too many. Like, I estimated the 9 comments we received just on IDHEAS-G alone probably 10 would add up to 500, 600 pages. But I think we can 11 find a way to summarize -- I mean, actually already we 12 have recorded summarized the comments and how we 13 addressed them. So we just need to do some final 14 summarizing work.

15 MR. PETERS: So, I guess, from Jing and 16 myself, that sounds like a really good idea. And I 17 think we would look into incorporating that.

18 MEMBER BLEY: One thing that is not clear 19 to me. You talked about reviews on the three. You 20 did two reports associated with FLEX. Have you had 21 any peer review of those?

22 DR. XING: The FLEX Report -- one FLEX 23 report was the FLEX expert elicitation. That's why we 24 had the internal staff review but not external review.

25 So FLEX evaluation was recently developed on December NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

27 1 27 something last year. So it hasn't been externally 2 reviewed yet.

3 MEMBER BLEY: Okay. Thank you.

4 MR. PETERS: Yes. So, Dennis, real quick 5 on the FLEX, we were under an Agency metric to get 6 that out the door by December 31. It just didn't 7 support the timelines for a peer review.

8 DR. XING: Yes. Just one more item. On 9 the FLEX, the expert elicitation reporter had peer 10 reviewed it. That's because we invited EPRI to 11 observe the activity. I thought this should be 12 classified as external review.

13 MEMBER BLEY: I agree.

14 MR. PETERS: Any other questions? Okay.

15 Jing, I guess we're ready for your presentation.

16 DR. XING: Okay. Thanks, Sean. Thanks 17 for the nice introduction and that really made my part 18 easier. Okay. I'm going to share my screen. Can 19 everyone see my screen?

20 MR. PETERS: We can, yes.

21 DR. XING: Okay. Thank you. Okay. So 22 I'm Jing Xing working for Sean Peters in the event 23 process of IDHEAS. So it's my pleasure on behalf of 24 the IDHEAS team today to present IDHEAS to ACRS' full 25 committee. So for today I will talk about the first NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

28 1 section. First, an overview of IHDEAS program from a 2 technical perspective and then I will introduce the 3 three IDHEAS products, IDHEAS-G, IDHEAS-ECA and 4 IDHEAS-DATA. After that, I will talk about the two 5 examples of using IDHEAS, which is what we just said, 6 the FLEX expert elicitation and the FLEX evaluation.

7 At the end, I would like to summarize our revision to 8 IDHEAS report up to the last ACRS subcommittee 9 meeting.

10 Okay. So starting from where we were in 11 the beginning of IDHEAS in later 2011 or beginning of 12 2012, so what you see on these slides are the 13 technical areas we want to pick on. On the top three 14 major areas, one, the scope of the HRA application.

15 Sean Peters just talked about in 2011 after Fukushima, 16 the Agency needed an expanded scope of HRA 17 application. On the top right, we have seen ones that 18 we can better and more use of the human performance 19 data used in Chart A to enhance HRA credibility.

20 The middle one is the big one that we 21 prepared in the SRM HRA data team. So at that time 22 from what we learned in the two HRA benchmarking 23 studies that Sean just talked, we can see in the slide 24 what caused the variability. So I summarized that 25 into three boxes. The first one is the uncertainty in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

29 1 the scenario. We don't always have perfect scenarios.

2 It always comes with uncertainty.

3 And the second one is the analyst 4 practice, even using the same method for the same 5 scenario, analysts had a different experience, 6 different focuses. And, of course, the big error we 7 want to improve is the HRA method as the benchmark 8 studies find that even there's a bigger analyst-to-9 analyst differences caused by the HRA methods.

10 So we do HRA method to benchmark the 11 studies, identifying a number of sources for HRA 12 variability. So the four major areas are, the first 13 one is quality of analysis guidance. And even when 14 you have a good qualitative analysis guidance from the 15 method used, that would still not address the issue of 16 how you transform the outcomes of qualitative analysis 17 to HRA quantification. And events performance 18 influencing factors or PIF because a big part of HRA 19 is estimating the human error probability, HEPs. So 20 HEPs depend on the PIFs. And we want to explain a 21 better description of the PIFs so they can be assessed 22 more confidently among analysts.

23 And the last error, probably the most 24 important error, is the cognitive basis embedded in 25 the HRA method because HRA wasn't just about getting NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

30 1 the HEP number. It needs to tell how human can fail 2 and the why they fail, how we can prevent them from 3 making errors. So thus we need a cognitive basis to 4 give the explanation. So both --

5 MEMBER BLEY: Jing?

6 DR. XING: So both -- yes. Question?

7 MEMBER BLEY: I'm going to interrupt you 8 for a second. And I'm not sure of the best place to 9 do this so I'm going to start here. Are you going to 10 say more in a later slide about the uncertainties in 11 the scenario?

12 DR. XING: Yes.

13 MEMBER BLEY: Okay. Then I'll wait on 14 that one. On the cognitive basis, I'm going to sneak 15 in a couple comments now because I have nowhere else 16 to do it. I think the work you did really is 17 excellent and makes this basis to the literature in a 18 way that hasn't really been done. The two reports, 19 the cognitive basis document NUREG-2114 and the 20 IDHEAS-G itself, both delve into this. I was bothered 21 in Chapter 2 of the IDHEAS-G report because in many 22 places the text says the figures make various elements 23 together. And when you look at the figures, they 24 really don't. I'll mention a few examples.

25 Figure 2-3 is fine. It's identified as a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

31 1 causal tree. And it's kind of okay as a cartoon. It 2 shows how one would lay out such lengths. And the 3 cognitive basis document actually does lay out such 4 links. One example is Figure 2-7 in that document 5 that ties together the performance influencing 6 factors, their mechanisms and what were called 7 proximate causes in that report. I'm going to come 8 back to that report.

9 But in your figures in Chapter 2, 2-5, 2-10 6, 2-8, the words say the figures show the links 11 between cognitive activities, processors and cognitive 12 mechanisms, and they don't show any links. They just 13 show that they're connected. I think you ought to 14 either change the text or change the figures to look 15 more like what was in the cognitive basis document 16 because they don't agree. It jumps at you once you 17 spot it. It's easy not to see it. I didn't see it 18 the first time through.

19 And then I had a -- well, let me go to the 20 proximate causes. In IDHEAS-G, you changed the 21 language from proximate causes to processors. But I 22 might have missed it. I don't think you ever 23 explained why you did that. I'm interested in why you 24 did that. And it took me several years to get used to 25 the first language and now there's new. And I'm not NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

32 1 sure why you made that jump.

2 And the last one about Chapter 2, in a 3 series of Sections, the first one, I think, is 2.3.1.3 4 on cognitive mechanisms for detection -- that's one of 5 the macro cognitive functions -- you said you opened 6 some bullets with some public capacity limits and 7 that's true. But you don't show any links between 8 those cognitive mechanisms and the elements of 9 detection, D-l, D-2, D-3, D-4. And I'm not sure why 10 not. It seems obvious. I tried to map them a little 11 bit. So that one is kind of a minor one. But the 12 other one, I think, is more important. If you can say 13 anything about the proximate cause change of language, 14 I would appreciate it. And I'll be quiet and listen 15 for a while.

16 DR. XING: Okay. So thanks for pointing 17 out those places. I think our team will look at the 18 transcript of what you just said and discuss how we 19 can change it, how we can better address that. And 20 quickly why we changed from proximate causes to 21 processors. Proximate causes was an earlier term when 22 the pre-IDHEAS team tried to say what are the 23 accompanying processes for human failures? And that's 24 why that word was put in 2114. Later on when we moved 25 to IDHEAS-G, we wanted to lay out more structure to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

33 1 accompany the model, to accompany to basic structure.

2 And so first we look at why people fail. We look from 3 the success phase first.

4 Okay. It's human need to perform 5 detection. What are the things that they have to do 6 in order to perform detection? And so the data is the 7 same. If they fail, they will fail detection. So 8 because we described the combination model from the 9 success basis, how people do the work so, therefore, 10 the process, even they appear at the same level as 11 previously we called proximate causes, they are talked 12 about the success factor. And the motivation for 13 change was from one ACRS meeting, I think it was 2013.

14 And there were two audiences that came to talk to me 15 after the meeting. They really liked her. We talked 16 about the success path first. So you need to 17 understand how humans assess their work. They talk 18 which is why they fail.

19 There was one bigger motivation we like to 20 use the term processor instead of jump to failure, 21 which is what a proximate cause means.

22 MEMBER BLEY: Thanks, Jing. I appreciate 23 that. You know, for somebody who picks up your 24 reports and reads them, I don't think that comes 25 across. And it might be worth a couple words in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

34 1 Chapter 2 of IDHEAS-G to clarify that. But go ahead.

2 DR. XING: Thank you.

3 MEMBER BLEY: Yes.

4 DR. XING: Okay. So we were in 8 or 10 5 years ago. Now this slide shows where we are now. So 6 the top blue color box represents the areas that we 7 are confident that we made an improvement in these 8 areas. And the two areas, above that top blue color, 9 I will talk about that later. So look what we have 10 achieved. So in the HRA method of scope, IDHEAS in 11 the HRA method is really for all nuclear HRA 12 applications.

13 The next bullet, use of the human 14 performance data. The human error data was basically 15 used in IDHEAS because the method and the basic 16 structure are based on the same cognitive basis model 17 such that data can be generalized and used by the 18 method. Previously, the HRA variability had issues 19 that the data doesn't match the method so we couldn't 20 use the data and the HRA variability. So IDHEAS 21 improved HRA method of variability and enhanced the 22 four areas that were identified in the HRA benchmark 23 studies. Because it offered a systematic qualitative 24 analysis guidance and the links between qualitative 25 analysis outcomes to quantification of human error NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

35 1 probabilities.

2 And for performance influencing factors, 3 every performance influencing factor has a set of 4 explicit attributes.

5 MEMBER BLEY: Jing, can I sneak in 6 something here?

7 DR. XING: Yes.

8 MEMBER BLEY: I'd like to go to your 9 bullet about data, about having it match the cognitive 10 basis arrangement. Something came up reading the 11 FLEX, and I think it was the expert elicitation part 12 of FLEX. When we get into the analysis in that 13 report, the names of the cognitive functions, not 14 their cognitive functions; detection, understanding 15 and so on don't get linked into this. And I'm 16 wondering why. It makes sense on the data. It seems 17 it would have made sense in your analysis of FLEX.

18 DR. XING: Are you talking to the expert 19 elicitation or the FLEX evaluation using IDHEAS-ECA?

20 MEMBER BLEY: I'm pretty sure it was the 21 expert elicitation.

22 DR. XING: Okay.

23 MEMBER BLEY: Whichever way you go, you 24 ought to be anchoring what you're doing back to that 25 cognitive basis in my opinion.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

36 1 DR. XING: Mm-hmm. I will talk about that 2 later. But to quickly answer your question, for 3 expert in the FLEX expert elicitation were operational 4 people. So we organized the data package for them in 5 a way that was more fixed to their thinking process.

6 We talked about it. They defined macrocognitive 7 functions but that wasn't a major part. We asked them 8 to do their measure. Maybe that's why you think 9 there's a disconnection between the data and the 10 macrocognitive function.

11 MEMBER BLEY: It makes sense to put things 12 in their language. But since you've got this 13 structure, it seems to me it would make sense to adapt 14 their language and link it your basis, but you didn't 15 do that. And --

16 DR. XING: We think --

17 (Simultaneous speaking.)

18 DR. XING: Yes. We picked the other part 19 because one major purpose of that expert elicitation 20 we want to have a better understanding of the PIFs, 21 whether we can use those PIFs for FLEX and whether we 22 can only need to subset them. So therefore, we 23 actually itemize the data by different PIF back to the 24 expert. And the expert found that that was very 25 helpful. And they actually made the recommendation we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

37 1 should use that data from the HRA method to do the HEP 2 estimate, stand-up behind the expert to do that.

3 MEMBER BLEY: Okay. Well, let me go just 4 a little further. You're going to come to the FLEX 5 stuff later.

6 DR. XING: Yes.

7 MEMBER BLEY: But since we've got this 8 going right now, I want to dig a little further.

9 Volume 1 and Volume 2 are the FLEX, the expert 10 elicitation in the ECA, the event and condition 11 assessment using your computer tool. You explain a 12 little bit about why things are a little different 13 between the two, but the reason there is Volume 1 and 14 Volume 2 kind of implies that they all get linked 15 together for their utility. And even if you change 16 the language for the experts, I would think in your 17 exposition and the expert elicitation report, adding 18 words that would tie their language back to your 19 structure would be very helpful in showing how the 20 pieces all fit together. But, go ahead. Don't dwell 21 on that now. It's a comment for your consideration.

22 DR. XING: Thanks. I really appreciate 23 that comment. That was a very good comment. At the 24 time we wrote the FLEX expert elicitation report back 25 in 2018, we were still focused on the FLEX part NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

38 1 because we could see this big picture. Thank you.

2 Okay. So we talked of a change that was 3 made in Phase 3 error. And the model I want just to 4 mention, the other sources, the other two sources of 5 HRA variability. The first one, uncertainty in the 6 scenario. The uncertainty would result in different 7 analysis assumptions. So that kind of uncertainty, 8 you can't do the scenario. HRA method cannot and 9 should not eliminate the uncertainty. But what we can 10 do is IDHEAS can provide a guidance on systematically 11 identifying uncertainties in the scenario and the 12 tracing for assumptions in the HRA process.

13 MEMBER BLEY: Can I interrupt you a second 14 again?

15 DR. XING: Sure.

16 MEMBER BLEY: This is one that leaps at me 17 because I've played with this one a lot. And I think 18 the method to give more help to someone, especially if 19 they're doing an expert elicitation, but even if 20 they're not, if you're looking at difference in the 21 result, we've often found when you see different, 22 either an elicitation or people using some more 23 prescriptive method, that when you see very broad 24 differences in the answers and you get people to 25 explain why they got their answers, you find that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

39 1 Person A and Person B have really picked different 2 subscenarios within the uncertainty of the definition 3 we've given them.

4 And, you know, sometimes that means you 5 might want to break one scenario into multiples to 6 examine them but at least acknowledge where the 7 uncertainty in the results are coming from. And then 8 it's not so much uncertainty in performance of the 9 method. It's variability in the thing you're 10 analyzing, the particular scenario. And I think 11 that's a real crucial one and one that deserves more 12 exposition, both about expert elicitation and whatever 13 tool you're using to quantify it.

14 MEMBER KIRCHNER: Dennis?

15 MEMBER BLEY: Yes. Go ahead.

16 MEMBER KIRCHNER: This is Walt. I'm sorry 17 for interrupting. But, yes, on this one, I also am 18 thinking at it. Now I'm an outsider. This is not my 19 area nor is PRA, but this looks a lot like PRA. And 20 so in this critical area of uncertainties in the 21 scenario, would it be feasible to use the PRA of 22 entries or, you know, that's also often done with the 23 assistance of expert elicitation and so on, so that 24 there's some -- I don't want to make this -- it's 25 already fairly complex. But isn't there some way, at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

40 1 least for the at-power kind of scenario since some of 2 the other, you know, going through the Level 1, 2 and 3 3 of the PRA for the existing fleet that you could 4 mine that information as input into the scenarios --

5 MEMBER BLEY: Sure.

6 MEMBER KIRCHNER: -- and reduce the 7 uncertainty? Because you would then start with what 8 was also a -- pardon. I got a phone call. Sorry.

9 I'll turn it off. You could use all that effort that 10 went into the PRA to be the basis for at least 11 starting on the scenarios.

12 MEMBER BLEY: Sure. And when you're 13 saying you're not an expert, but you had some 14 operating time and from that you can get this idea 15 pretty well. But this method and the others all ought 16 to be using the information in the PRA. But within a 17 particular PRA scenario, they haven't looked at all 18 the other things. You know, the things that get 19 modeled there are the particular pieces of equipment 20 that can challenge the core -- not challenge the core.

21 But there is a whole world of other things going on 22 and flavors within that one that can affect what the 23 people do.

24 And, you know, one way is to identify them 25 all in advance, which is a very big job. Another way NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

41 1 is to do that to a reasonable amount. And one is 2 you're quantifying when you find things that look 3 wildly different and try to figure out why are they 4 wildly different? Is it because just the people did 5 it differently? Their heads weren't in the right 6 place or the method didn't help them? Or was it 7 because they were looking at things that were 8 different but were within the definition of a single 9 scenario or call it a scenario group.

10 And I think guidance in that area can 11 really help and can also explain when you see a couple 12 of orders of magnitude difference in different 13 analyses that in many cases that's because they're 14 looking at different detailed scenarios. Anyway, 15 Jing, we took it away from you for a while. I'll put 16 it back to you. But I think that's an area, for which 17 there is plenty of experience to give better guidance.

18 Now maybe this is done in the derivative methods or 19 maybe it's done in some later improvements through 20 IDHEAS-G. But go ahead.

21 DR. XING: Thank you. I appreciate the 22 discussion. And the next item is related or similar 23 to what we just said. In practice all HRA needs an 24 analyst to interpret the information and to enter that 25 to the HRA method. So it's largely relying on analyst NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

42 1 interpretation. And the different analysts practices 2 will result in different interpretation of the same 3 scenario.

4 And the IDHEAS SPAR mutation, it has been 5 also used as a structure process to make a clear 6 documentation of each step. That provides a good 7 possibility of transparency of analyst interpretation.

8 We cannot eliminate the difference. They can still 9 come up with different interpretations so we can say 10 why they interpret it differently. So that will help 11 us to reconcile the resulting variation. I will have 12 an example of each of these items later on.

13 Okay. So we're ready to look at 14 individual IDHEAS products. And you have seen these 15 products in Sean's earlier slides. This one kind of 16 recaps the process of how these products are related 17 in such ways that go back to the combination basis for 18 HRA. From that, we developed the IDHEAS-DATA 19 methodology. And IDHEAS-DATA methodology is intended 20 to developing application specific method. The first 21 one we developed was the IDHEAS Internal At-Power 22 Application. But I would like to say chronologically, 23 IDHEAS Internal At-Power Application was completed 24 before the IDHEAS-DATA methodology. We keep evolving 25 and developing the DATA methodology.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

43 1 From the general methodology used in the 2 structured, we developed IDHEAS-DATA, use data to 3 document various sources of human error data. And 4 using IDHEAS-G and the data together, we developed the 5 IDHEAS-ECA. And I should also put an arrow here so 6 IDHEAS-ECA was also developed from IDHEAS with the 7 input from 2018 FLEX expert elicitation. So on the 8 bottom are a bunch of test team or pilot team 9 applications of the products that Sean talked about 10 earlier. For the rest of the presentation, I will 11 give a relatively high level introduction of the three 12 products, and I will talk about two examples of the 13 application.

14 MEMBER BLEY: Jing?

15 DR. XING: Yes.

16 MEMBER BLEY: Okay. This is the last one 17 I really want to jump on.

18 DR. XING: Oh, just jump on it.

19 MEMBER BLEY: You set me up here. The 20 FLEX HRA expert elicitation you see as a source of 21 data that the -- the data report and was used in the 22 FLEX ECA. And now I want to complain about the 23 numbers just a little bit. I don't usually like to do 24 that. But some comments you received in public 25 comments complained about the treatment of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

44 1 uncertainty. I got comments from members of the 2 Committee before today that were very negative saying 3 there's no real basis for how we tried to combine 4 these distributions. For everybody else, Appendix D 5 of the FLEX elicitation report summarizes the data 6 from all their experts. And it gives a lower bound, 7 an upper bound and a best estimate. I've seen some 8 arguments back and forth.

9 The best estimate, at least in the report, 10 isn't defined. Some people have said it should be the 11 means. Some people said it has to be the median. And 12 someone said part of the way you combined these works 13 for me but not for median. But psychological 14 literature, the older literature back in the 70s 15 primarily, found that when you ask people for their 16 best estimate, most of the time what they give you is 17 the mode, the value that occurs the most often because 18 that kind of gets anchored in your mind. And 19 depending on the distribution, that might be very 20 close to the median or maybe it's a little higher but 21 that's a detail.

22 Arithmetic and geometric means on the 23 upper, the lower and the best estimate is kind of hard 24 to justify. Not the cognitive basis document, the 25 report, your white paper report on the elicitation, by NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

45 1 the way I think it's very good and Jing was one of the 2 authors, it gives some descriptions of how you combine 3 distributions. The one most often cited in that 4 report as being the best way is to find a distribution 5 that the experts agree represents the best estimate of 6 the scientific community. And that's one I think you 7 ought to strive for, using arithmetic or, like, 8 additive calculations or geometric means or 9 substitutes, and we give some basis to that.

10 Now the basis that talks about geometric 11 means is really talking about an individual's high and 12 low estimate and using the geometric means to get a 13 mean estimate that does work in many cases quite well.

14 By experience, I'm not sure theoretically. But 15 whichever is right, if you go to the tables, the thing 16 sent me to looking at the details was you make a 17 statement that usually the geometric mean is a little 18 bit less than the arithmetic mean for each of your 19 combinations across the experts. And I said, I 20 thought it would be bigger than that. And when I 21 looked, yes, sometimes it's a lot bigger, sometimes 22 not.

23 But always the geometric mean ought to be 24 lower than the arithmetic mean. I think we've heard 25 that one. So I don't remember the details. So I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

46 1 looked at the tables, and I found a number of cases 2 where that wasn't true. The geometric mean was higher 3 than the arithmetic mean. Those are in Tables D, 4 delta, 5, 10, 11, 13, 16 and 21. So I took out my 5 calculator, and I calculated the geometric means. And 6 in all cases there was an error in the table, 7 substantial. The geometric means are always less than 8 the arithmetic means.

9 I checked a couple of arithmetic means, 10 and most of them were right on, but a few were off, 11 not by too much. But I don't know why they would be 12 off at all. So you've got a bunch of errors in those 13 tables, and you're feeding that in as data for people 14 to use. You need to go back and fix that. That's all 15 on that issue for me for now.

16 DR. XING: Okay. Thanks. I really 17 appreciate the comment. We will come to some of that 18 later. Before that I can quickly say something about 19 that so backward. So for the geometric and arithmetic 20 mean table, the errors were the -- I know where the 21 possible errors we calculate, why analysts send their 22 estimation later also. There are some data entry 23 errors there. So I will definitely go back in the 24 text and fix that.

25 And the tool for using the number, I am NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

47 1 very much with your comment on we didn't estimate 2 before distribution. But in terms of its impact on 3 IDHEAS-ECA, we didn't use the expert elicitation 4 number to inform the HEP, the numbers we need to 5 calculate HEP in IDHEAS-ECA. The reason was because 6 expert elicitation data, because that kind of data is 7 multi-component, the expert estimates the error to HEP 8 of the entire action, which consists of multiple 9 failure modes and the multiple PIF conditions.

10 So we couldn't disassemble that data. So 11 we only used the expert elicitation for verification 12 purposes, like, we got the IDHEAS-ECA, developed it.

13 We tried out the expert elicitation and specification 14 of the scenario. And, we say, okay, the number, you 15 say it is not far from what a data expert got off of 16 that expert got because different expert has different 17 assumptions. So that's the way we -- that's what 18 caused the numbers we used for IDHEAS-ECA.

19 (Simultaneous speaking.)

20 MEMBER BLEY: Would I be able to find that 21 in IDHEAS-ECA? I didn't. That's why I ask.

22 DR. XING: Yes. It's not in the report.

23 But I can give you my scrap of pages on -- I tried to 24 say if IDHEAS can refute the numbers in the FLEX 25 expert elicitation, that was a -- has two sides.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

48 1 MEMBER BLEY: I think if this gives more 2 confidence in ECA, it would be worth including.

3 DR. XING: Yes. And so the reason that we 4 only had expert estimate the upfront and the low level 5 and most like HEP, back to 2018 it was the best 6 experienced expert did not have much experience, of 7 survey experience, in errors of FLEX equipment. So 8 the cognitive expert and our expert clearly indicated 9 they didn't have sufficient analogy to come up for 10 distribution. The best that they can do was the 11 upfront and lower level and the most likely case.

12 MEMBER BLEY: Okay. I think you got to 13 work more with it, but if your claim earlier, or 14 Sean's claim, that IDHEAS-G is human-based and 15 applicable across the board, you wouldn't need 16 specific ECA experience to be able to evaluate the 17 human -- the effects on humans of events that could 18 occur during the FLEX operations. Are your methods 19 general enough to cover that?

20 DR. XING: Yes, that's true.

21 MR. PETERS: But you do want to test it, 22 though. That's a key, Dennis, yes.

23 (Simultaneous speaking.)

24 MEMBER BLEY: Did we actually test it? I 25 don't see that we said how we tested it. I mean, I'm NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

49 1 hearing it now, but I don't remember reading it.

2 MR. PETERS: Jing, would you like to 3 answer that?

4 MEMBER BLEY: You don't have to do it 5 today. That's something to think about. You know, 6 it's almost a good story, but it ought to be written 7 down somewhere.

8 DR. XING: Okay. Yes, so that was the 9 reason we didn't do that before distribution. But the 10 main input --

11 (Simultaneous speaking.)

12 MEMBER DIMITRIJEVIC: Sorry. I was 13 waiting to get to the FLEX part of presentation to 14 discuss this. But I think the one of them -- the 15 level of discussion that you have is very good. There 16 is the measure of elicitation in white paper, which 17 wasn't used in the way -- the integration process as 18 described in this paper was not used in FLEX 19 elicitation. Developing distribution, you know, is 20 essential for this integration process because you 21 cannot just integrate numbers. And those were not the 22 most likely numbers. They represented the middle.

23 There were 50 percent that mean, like, whatever 50/50 24 tends to be higher or lower this number. Those are 25 not the same as the most.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

50 1 And actually we cannot even talk about 50 2 mode and middle without assuming distribution and how 3 integration of the basic FLEX was assuming to talk to 4 the symmetrical distribution where the 50 percent and 5 mean are the same. But the difference between 1 and 6 99 percent, well, anyway, not to put the contents in 7 technicality, I think that proper integration of the 8 data presented in FLEX external elicitation was not 9 done, and it should be done. And in that case, maybe 10 this integration fact would be omitted from the report 11 without discussing distribution. It doesn't make any 12 sense to do integration.

13 (Simultaneous speaking.)

14 MEMBER DIMITRIJEVIC: -- the interval in 15 distribution but that should be also part of 16 preparation. And I think that this data has a value 17 even if it wasn't integrated because obviously you 18 were not using integrated results in the Volume 2. So 19 maybe this moment, this integrated result should not 20 be presented.

21 MEMBER BLEY: Well, Vesna and the 22 Committee and Jing, the real reports, regulatory 23 information letters on this, have been published.

24 Where I'm leaning is to say if or before you ever 25 publish a NUREG based on this, this should be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

51 1 revisited. Anyway, back to Jing.

2 MR. PETERS: Yes, and, Dennis, I'd like to 3 just weigh in just a little bit. The original intent 4 of the FLEX HRA expert elicitation, we did this in 5 2018. This was prior to the development of IDHEAS-ECA 6 and prior to its publication. So the original intent 7 of that work was to give our reviewers over in NRR a 8 concept for the feasibility and capabilities of FLEX 9 because they're trying to encounter this in a day-to-10 day, like, license amendment or, you know, no ed type 11 situation where they wanted to get some type of credit 12 for this work.

13 So what we saw in this was an opportunity 14 to get some data from it. But it's original intent 15 was not to develop data for IDHEAS-ECA, but it was to 16 help the users in NRR. And so from that, obviously, 17 direct, as Jing indicated, direct one to one data 18 capture is not kind of the way we would normally do it 19 in an IDHEAS-ECA program because, you know, IDHEAS-ECA 20 is built from the micro level. And this is kind of 21 macro data that comes in. So I just wanted to give a 22 little more context behind it.

23 MEMBER BLEY: Yes, thanks, and, you know, 24 the horse is out of the barn sort of thing. That 25 makes sense. A few words in the introduction to put NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

52 1 that in context might have helped, but it makes it up.

2 DR. XING: Thanks for this discussion. I 3 really appreciate it. Okay. So let's move forward.

4 So we will move to the next section, the individual 5 products, beginning with IDHEAS-G. So just quickly 6 what is IDHEAS-G? First, it's an methodology for 7 developing applications specification to HRA method, 8 and it's also a platform for generalizing and 9 integrating human error to support HEP estimation.

10 And finally by itself it is the general --

11 it can be used as the general HRA method for human 12 event analysis and human error root causal analysis if 13 you don't want to have to get an HEP number and are 14 only interested in what are the causes and how to 15 prevent the causes. So I didn't see -- it is consists 16 of with the three parts. It's consists of a 17 combination model as the framework for HRA. And it's 18 the implementation in the HRA process that makes it an 19 HRA method and the detail, the guidance for HRA 20 application. So we had all those appendix into a 21 bunch of them for having the guidance for different 22 elements in IDHEAS-G.

23 So the combination model has two parts, a 24 combination basic structure, which is discussed in 25 Chapter 2 that Dr. Bley mentioned earlier and a PIF NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

53 1 structure. We implemented the combination model in 2 both stages of the HRA process. Stage 1 is the 3 scenario analysis. Stage 2 goes into the detailed 4 modeling of important human interactions in the 5 scenario. Stage 3 is for HEP quantification. Stage 6 4 is integrative analysis, which includes uncertainty 7 identification documentation and dependency analysis.

8 So I will talk about each of these 9 elements at a very high level, but feel free to 10 interrupt me. Okay. So the combination basic 11 structure, the human task that is represented takes 12 place inside a micro combination function. We take 13 the information and in expanding the situation, make 14 divisions of plans and executing the plans of the 15 position and the inter-team coordination in the bigger 16 complex working environment.

17 So failure of each of business 18 microcompany function can lead to the failure of the 19 human task. And each macrocompany function is 20 achieved through a processor, which each processor is 21 a key element to how you achieve the function. For 22 example to achieve with the texture. It's not just 23 that you take a quick look at that and stop. It 24 starts with the basic you know what you're going to 25 look at so you have a mental model for what you're NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

54 1 going to look at.

2 And you will attached it to the top of the 3 queue, and you perceive the information but you also 4 need to recognize and classify that information and 5 the unique verifiers of that information. But you 6 process the information you need to verify and 7 properly need to correct it. Then at the end, the 8 last of the process, you need to export what you 9 perceive. You either use it for your own use in your 10 decision-making or you gave that to your teammates or 11 whatever. So those are some processors for obtaining 12 the detection.

13 MEMBER BLEY: Hey, Jing.

14 DR. XING: Yes?

15 MEMBER BLEY: Yes. This is Figure 2-3 out 16 of your report, which I said was a good cartoon. And 17 it does show that a single PIF can affect more than 18 one cognitive mechanism and likewise cognitive 19 mechanisms can affect more than one processor. And 20 that's what disappears in the figures as it continues 21 through the chapters. I just wanted to make that 22 clear to you.

23 DR. XING: Okay. Thank you. Now I see 24 what you mean. Okay. Probably we want -- maybe 25 because we wanted to make the figure look less messy.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

55 1 MEMBER BLEY: It is much less messy, but 2 the words don't match it any longer.

3 DR. XING: Thank you. I'm pleased you 4 said that. Yes. Yes. As you said, the messy part is 5 because they're multiple things not the one to one 6 thing, I think, like the combination mechanism for 7 attention not the company's mechanism for detection is 8 attention. But it also affects other processes, other 9 functions. And with that, so the last part of this 10 slide is most common are the PIFs. So basically if 11 you look from left to right, a PIF can effect a 12 combination, mixed mechanism not exact thing. Then 13 the mechanism leads to increased chance of error in 14 one or more processes and an error in the process 15 would be a failure of the detection.

16 MEMBER BLEY: Jing, you explained why you 17 changed the language to processors. But is that 18 common language in the psychological literature?

19 DR. XING: No. Well, I took it upon 20 myself from my years working on companies on 21 combination with your side. There wasn't a single 22 term to describe all those things.

23 MEMBER BLEY: Okay, That's what I thought.

24 DR. XING: Yes. I tried many, many 25 versions, I tried to come up with a word that makes NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

56 1 sense. Okay. PIF structure. So IDHEAS uses the PIFs 2 to model contacts. Contacts are the combinations that 3 challenge of a facility of human performance in this 4 scenario.

5 So IDHEAS collected by contacts in four 6 categories, in environmental institutions and systems, 7 personnel and organization and tests. For each 8 category there are several PIFs associated. Some 9 parts look at the first category, environmental 10 elicitation. The PIFs are acceptability to the 11 workplace and include the entire path. And the 12 workplace visibility, noise, cold and heat, humidity 13 and the resistance to physical movement.

14 So based on the PIF that's at a high 15 level. And as we said, to give a more explicit 16 description of the PIFs, in others you can assess them 17 more consistently, but each PIF will develop a set of 18 PIF attributes. Every attribute represents one way 19 that the PIF can challenge a combination method and 20 therefore increase the likelihood of failure or human 21 error. So we can look at example of the attributes.

22 So one cause of PIF, human system interface, the 23 detonation into HSI, human system interface, refers to 24 indications such as takes place in indicator amounts 25 and the controls. So the indication is for the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

57 1 technical information controls up for team actions of 2 systems.

3 Some attributes, this is not a full list.

4 Some attributes, like first off was the indication, 5 it's similar to other processes nearby, are the 6 indications that have no work, or come to the control.

7 The labels on the controls do not agree with what you 8 have documented. And in a worst case, controls are 9 not reliable if the person is not aware of the 10 controls are not reliable so based on the more 11 explicit description of what a PIF is.

12 So looking at IDHEAS-G, how IDHEAS-G model 13 human failure event that comes to other HRA process.

14 So Stage 1, scenario analysis, and Stage 2, modeling 15 human action. So starting with the human failure 16 event and we have guidance to identify the human 17 actions and context in the event and the event 18 contexts but that's the scenario for this part.

19 Also scientists will model the task this 20 with a five micro combination function or the more 21 detailed lever, you can model the failure of the task, 22 basically, the failure of those processors, which now 23 we can say the failure of a processor is what we call 24 the proximate cause in 2114 or you can model the 25 failure of a task in very detailed and elaborate using NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

58 1 application specific failure mode developed for the 2 processes. IDHEAS-G provides a set of examples for 3 the detailed failure mode.

4 For modeling the context that we have 5 plenty PIFs and their attributes. And we also have 6 model templates. So after that model, now we come to 7 Stage 3, quantification. For quantification, the HEP 8 of the human study event consists of two parts. PTs 9 attributing to 10 available and 10 requirements. So 10 when you have less time, when you have time available 11 on what you require it to do to complete a task, the 12 failure have a chance of error. So PT is the 13 probability that less time.

14 And the PC is the error probability 15 attributing to the combination failure mode. So you 16 look at the complicated human error event, and we had 17 multiple critical tasks. Each critical task may have 18 one or more failure modes if you look at all the 19 failure modes. That's the probability appropriated 20 back to the total HEP. And IDHEAS-G, this slide shows 21 the three ways to estimate HEP off of the PC part.

22 Ideally, you can do the calculation from the number of 23 errors, you may divide it by the number of occurrences 24 you perform test --

25 MEMBER BLEY: Jing?

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

59 1 DR. XING: Yes.

2 MEMBER BLEY: Just so we don't confuse 3 each other and ourselves, on your last slide, when you 4 talk about critical failure modes, if we go back to 5 what was Figure 2-2 in the IDHEAS-G report, human 6 action and the number of tasks and under that the 7 macrocognitive functions, the critical failure modes 8 correspond to a particular task, right?

9 DR. XING: Yes, the failure mode 10 corresponds to --

11 MEMBER BLEY: A task somebody has to carry 12 out, and they fail it. Yes. Just so they have a 13 relate back to the IDHEAS-G book.

14 DR. XING: Mm-hmm. Yes.

15 (Simultaneous speaking.)

16 DR. XING: Thank you. After this, I would 17 like to mention that I think that very useful in 18 IDHEAS-G. The HRA community has been -- how we break 19 down our human event to the task, where is the level 20 to stop? IDHEAS-G has a guidance on what it means by 21 critical test and how to break -- where you should 22 start breakdown time. And so far the feedback we got 23 that was a very useful thing. The main concept is you 24 break down into critical tests only when you have to, 25 which means a the PIF can't in the HEP so you cannot NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

60 1 apply. Then you have to break them down.

2 Okay. So back to the calculation. Yes.

3 If you perform this -- you have the data that you 4 performed identical tests and the identical context 5 for 10,000 times. And you can certainly get the error 6 probability from that data. But unfortunately the 7 data is really eventful. So we still need enough time 8 to rely on expert judgment. That's another way to 9 estimate HEP. And IDHEAS-G also comes with HEP 10 quantification model that you can use that model to 11 calculate HEP.

12 MEMBER BLEY: We're getting --

13 DR. XING: Yes?

14 MEMBER BLEY: You just brought up the 15 language I wanted to get up eventually. Some of the 16 comments you repeat, some others talked about this.

17 You just presented us, we can use expert judgment or 18 we can have data. The FLEX expert judgment report 19 does the same thing. And as some of your commenters 20 pointed out, it's not either or on this. In fact, 21 rarely is your data so perfectly applicable that you 22 don't need to bring your judgment to the process.

23 And often a good place to start is an 24 elicitation to get what invasion analysis you'd call 25 prior probability distribution, which is a good place NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

61 1 to start. And as you get more data, you update that.

2 I would say the one caveat there is when you generate 3 such distributions, you need to ensure they have at 4 least some tail because anywhere the prior is 5 identically zero, no data was ever affected. So you 6 have to be careful about that, and there's a number of 7 ways to deal as you update things using Bayes' 8 theorem, ways to double-check and make sure you 9 haven't started to acquire what's causing a problem.

10 But you don't talk about that anyway. You 11 kind of make it -- you either use expert judgment or 12 you use data. And I don't find that a reasonable 13 approach. I think you use a combination to the extent 14 it makes sense. And when there's more data to 15 accumulate, you use the data long with what you have 16 previously. And I don't know anywhere you talk about 17 that either in IDHEAS-G or in the FLEX expert 18 elicitation report. And I'm not sure if you talk 19 about it in the white paper. It would make sense if 20 it were in the white paper. It probably is, but I 21 don't remember for sure.

22 DR. XING: Okay. Thanks for the 23 questions. The quick answer is that's the reason we 24 got so far with data in the report by saying you can 25 use one based approach or a combination of this.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

62 1 However, it was our intention, we purposely failed to 2 mention the prior approach you just talked. The 3 reason was we saw from the studies actually in the 4 IDHEAS scope and used that approach. The result shows 5 that the less you set the prior because we don't have 6 a huge, huge amount of data, the final result is 7 putting innovation out of the final outcome is pretty 8 much from the advised prior. I think, as you said, it 9 had to be cautioned with the prior. So --

10 MEMBER BLEY: Yes, but if you don't have 11 much data that's probably better than using data that 12 might be peculiar.

13 DR. XING: Yes. So at this point we don't 14 have good confidence, which use data that might 15 mislead you or not use it. You don't know where to 16 start. So we would rather wait, like, in the future 17 if we are more mature in that area, with more studies 18 and then we introduce that into future versions of the 19 report.

20 MEMBER BLEY: Well, I reiterate, it's not 21 an either or proposition. And the example you cited 22 is just the example I would cite to the other 23 argument. If your data is so sparse that they don't 24 mean anything, you're probably better using the best 25 judgment you can bring together on an issue. We can NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

63 1 talk about this more later. It's not in any of the 2 reports except maybe the white paper. And it ought to 3 be somewhere. So be thinking about it.

4 DR. XING: Okay. Yes, the white paper.

5 I wrote that, like, five years ago. I will read it.

6 But I definitely agree that we should discuss it in 7 the white paper. Okay. So the last stage, 8 integrating analysis, I would like to mention based on 9 ACRS subcommittee's recommendation from the previous 10 meeting, the previous, previous meeting, 2019 meeting, 11 the staff developed IDHEAS dependency model, which 12 this model is different from what has been 13 traditionally used to start model.

14 I wanted to go through the details on this 15 slide how the model works. Basically, it's based on 16 the IDHEAS combination structure. So you identify 17 dependency context and the model dependency context, 18 the IDHEAS failure modes and the PIF then calculate 19 the HEP for that context. That was basic the concept.

20 And that now we have a workgroup going through the 21 analyses for guidance for how to use this dependency 22 model. So when we get that guidance advanced, we will 23 update IDHEAS-ECA and put the guidance in the IDHEAS-24 ECA report and software.

25 So a summary of IDHEAS-G, just as we said NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

64 1 earlier, I think we said, we already talked about the 2 first of three bullets. I just want to emphasize that 3 because it's human centered, the last bullet is 4 applicable to all nuclear applications. So if no more 5 questions on IDHEAS-G, I will move to IDHEAS-ECA.

6 MEMBER BLEY: Let's not move. Let's say 7 we're an hour and a half in, and it's time for break.

8 I don't know how many slides you have, Jing. Is this 9 more than halfway through or about halfway?

10 DR. XING: I think it's one-third.

11 MEMBER BLEY: Only one-third?

12 DR. XING: Yes. IDHEAS-ECA and IDHEAS-13 DATA, because we already had most of discussion 14 upfront.

15 MEMBER BLEY: Okay. So maybe you can skip 16 some of that. Okay. Well, let's take a break. And 17 we have to finish it, I think, 11 o'clock my time and 18 1 o'clock in Washington. Let's take a 20 minute 19 break, about a 20 minute break. Let's come back at 25 20 after 11 Eastern Time, and we'll take this up again.

21 And when we get to places we've already discussed, 22 feel free to go quickly or even skip some slides.

23 DR. XING: Okay.

24 CHAIR SUNSERI: Hey, Dennis. Hey, Dennis.

25 This is Matt. Let's just round it off to 11:30.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

65 1 MEMBER BLEY: Okay. Good idea. Be back 2 at 11:30.

3 CHAIR SUNSERI: Thank you.

4 MEMBER BLEY: We are -- I almost used the 5 wrong word. We are in recess.

6 CHAIR SUNSERI: Thank you.

7 (Whereupon, the above-entitled matter went 8 off the record at 11:07 a.m. and resumed at 11:30 9 a.m.)

10 CHAIR SUNSERI: Okay, members. It's 11 11:30. We will reconvene the ACRS session here. I'll 12 start with the roll call to confirm that we have the 13 quorum returned. Ron Ballinger?

14 MEMBER BALLINGER: Here.

15 CHAIR SUNSERI: Dennis Bley?

16 MEMBER BLEY: Here.

17 CHAIR SUNSERI: Charles Brown?

18 MEMBER BROWN: Here.

19 CHAIR SUNSERI: Vesna Dimitrijevic?

20 (No response.)

21 CHAIR SUNSERI: Walt Kirchner?

22 MEMBER KIRCHNER: Here.

23 CHAIR SUNSERI: Jose March-Leuba?

24 MEMBER MARCH-LEUBA: Here.

25 CHAIR SUNSERI: Dave Petti?

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

66 1 MEMBER PETTI: Here.

2 CHAIR SUNSERI: Joy Rempe?

3 VICE CHAIR REMPE: Here.

4 CHAIR SUNSERI: Pete Riccardella?

5 (No response.)

6 CHAIR SUNSERI: And Vesna Dimitrijevic?

7 MEMBER DIMITRIJEVIC: I'm here. Sorry.

8 (Simultaneous speaking.)

9 CHAIR SUNSERI: And Pete Riccardella?

10 (No response.)

11 CHAIR SUNSERI: Okay. Dennis, we have a 12 quorum. Pete, I'm sure, will join us as soon as he 13 gets back. So I'll turn it over to you for 14 continuation.

15 MEMBER BLEY: Thank you very much, Mr.

16 Chairman. I mentioned to the committee that we've had 17 so many discussions with the staff before this 18 meeting. And our main purpose here is to look very 19 closely at IDHEAS-G and decide what is the primary 20 intent of the SRM.

21 We are also looking at the other reports 22 and we'll address their stage. But the main focus is 23 IDHEAS-G. At this point, I'll turn it back over to 24 Jing and we'll continue with the presentation. And 25 just a reminder, Jing, we do end at 1:00 o'clock your NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

67 1 time. And so jump through the things that we've 2 already discussed. Go ahead.

3 MS. XING: Yes, thanks. So I think I'll 4 go quick, IDHEAS-ECA and IDHEAS-DATA, to leave us time 5 to look at the two examples and talk about the 6 revision. So some IDHEAS lead to IDHEAS-ECA.

7 Basically, we're talking of developing application 8 specific method.

9 So this is an approach we develop specific 10 methods. Of course we have guidance in IDHEAS-ECA, 11 IDHEAS-G. But to make it short, you just try to make 12 your best tradeoff between going through very detailed 13 analysis and meet users' needs.

14 So essentially, every IDHEAS is a specific 15 method. It's a subset of IDHEAS-G, one way or the 16 other. So for IDHEAS-ECA, it was measured from our 17 users in NRR.

18 The scope of the method was to be able to 19 perform event and condition assessment for all NRC 20 risk informed application. Specifically, the method 21 should be applicable for plant HRA. The requirement 22 is easy to use, not overburden HRA analysts.

23 And the resource we had, we had human 24 error data. We had -- we already performed the FLEX 25 expert annotation, of course, IDHEAS-G. I just want NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

68 1 mention, look at the data between IDHEAS-G and IDHEAS-2 ECA.

3 It's important to realize the qualitative 4 analysis guidance for application specific method stay 5 the same as in IDHEAS-G. And for failure mode, we use 6 IDHEAS-ECA, use the high level setting mode. And we 7 talked to experts in --

8 (Simultaneous speaking.)

9 MEMBER BLEY: Jing?

10 MS. XING: One per person was to select a 11 subset of PRA that was specific for FLEX. Then it 12 turned out all the 20 PIFs are important. So IDHEAS-13 ECA preserved all the 20 PIFs.

14 Total, the PRA have a compressed set of 15 PRA attribute, means we combine the sub-attribute.

16 And then the special feature of IDHEAS-ECA is it use 17 the HEP quantification model. The numbers of the 18 prong to in the model came from IDHEAS-DATA.

19 We're not going to talk this. This is a 20 quantification model. We're not going to talk this 21 again. But mainly, I would like to --

22 (Simultaneous speaking.)

23 MR. PETERS: Hey, Jing.

24 MS. XING: Yeah?

25 MR. PETERS: Dennis is trying to break in.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

69 1 MEMBER BLEY: Thanks. Can you go back one 2 slide?

3 MS. XING: Okay.

4 MEMBER BLEY: You started up at the 5 qualitative analysis guidance being the same as 6 IDHEAS-G and that really deserves some emphasis. The 7 empirical studies that Sean discussed really flag that 8 as a key area where a lot of HRA analysis has gone bad 9 in the past and that people didn't really thoroughly 10 look at defining what they were analyzing and what was 11 important. And that's the real reason why you're 12 requiring that qualitative analysis to be thorough in 13 every application. Go ahead.

14 MS. XING: Okay. So mainly I want to say 15 too, in order to use the quantification model, we need 16 to get all the numbers, the failure HEPs for every 17 failure mode and every PRA attribute. So properly, I 18 never calculate it. But probably we need around 19 somewhere from three to five hundred numbers. All 20 those numbers came from IDHEAS-DATA.

21 And I just would like to emphasis. So 22 here is the IDHEAS-ECA process of diagram. So that 23 diagram we had in our report.

24 What you want to look at this line, the 25 three colored boxes are the ones in our software NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

70 1 that's for calculation for calculating HEP. And all 2 the rest are qualitative analysis. And also at the 3 end is certainty documentation is not in the software 4 yet.

5 So we're really emphasizing you need to go 6 through this whole qualitative analysis before you use 7 the software to calculate HEP. That's what we 8 emphasize, but no guarantee analyst will do that. So 9 we have the ECA report and --

10 MEMBER KIRCHNER: Jing?

11 MS. XING: Yeah?

12 MEMBER KIRCHNER: This is Walt Kirchner.

13 Can you go back one slide?

14 MS. XING: Yeah.

15 MEMBER KIRCHNER: Your slide makes the 16 point that I had asked earlier. It shows the PRA as 17 kind of the input for the scenario narrative 18 development. How close coupled are those? Do you 19 take the event trees to develop your scenario 20 narratives and timelines? Or what's the relationship 21 there in a typical application of the PRA model as 22 input to the IDHEAS-ECA process?

23 MS. XING: Okay. We just in our 24 dependency workgroup so far five -- so far four people 25 presented how they take from PRA model go to the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

71 1 scenario narrative. And when they only have a PRA 2 model, they do the event tree. They take the event 3 tree and go to the developer's scenario narrative.

4 Then in that process, you have to make 5 many assumptions. And there's other examples we had 6 in dependency workshop were from HEP. There you have 7 more detailed information. Pretty much, you already 8 have a scenario there. You just need to organize the 9 information to IDHEAS qualitative analysis for 10 guidance format.

11 MEMBER KIRCHNER: Since the PRA often 12 involves expert elicitation or PERTs and so on, does 13 that then replace the need for that expert elicitation 14 to develop these scenario narratives and timelines?

15 Or is that -- see where I'm going with that? Is that 16 a redundant thing, or does the PRA suffice to kind of 17 give you a consistent framework to develop the 18 narrative and timelines?

19 MEMBER BLEY: Jing, can I help on that 20 because I know --

21 MS. XING: Yeah.

22 MEMBER BLEY: -- you don't do PRA. Walt, 23 if you don't mind, I'll --

24 (Simultaneous speaking.)

25 MEMBER BLEY: -- because in a well done NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

72 1 PRA that's integrated with this HRA. You might do all 2 of that together, or you would at least do a lot of 3 it, as you suggest. And then when you refine the HRA, 4 you might and do additional investigation. But -- and 5 you start with the PRA, but that's why -- and I don't 6 remember if IDHEAS-S really recommends this, but it 7 ought to. The team doing the HRA ought to include the 8 people who are really knowledgeable about the PRA and 9 the engineering of the plant.

10 MEMBER KIRCHNER: Yeah, that's where I was 11 going with that, Dennis. Thank you.

12 MS. XING: Thank you, Dennis. Okay. So 13 we talk in the process to enforce the course guidance.

14 We have a set work center making the analyst do step 15 by step, follow that process. And we also develop 16 preliminary training materials. I think that's an 17 area we really need improvement to develop better 18 training materials.

19 And the good news is we have IDHEAS-ECA 20 software. So for everyone, it's a computer interface 21 and placement in the ECA for HEP calculation. We 22 start out to recommend you need to first analyze the 23 event documents that result in the work space, then 24 enter that information in calculating HEP.

25 However, because the software is so NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

73 1 lovely, we couldn't prevent people from jumping to 2 play with the software. That's something we need to 3 think about in the future. This is the PDF snapshots 4 of what the software look like. So any question on 5 IDHEAS-ECA?

6 (No response.)

7 MS. XING: Okay. So I will move to 8 IDHEAS-DATA. We always say HEP is a function of PIFs.

9 It's easy to say that, but not so easy to solving that 10 function with a lot of data.

11 The good news, we know that lots human 12 error data exist from various domains. And the bad 13 news is it varies in format, basically the content, 14 the number of details. We talk about that has been a 15 hurdle of using data.

16 So the IDHEAS approach is simple. The way 17 of performing IDHEAS HRA for every data source, you 18 take the data source thinking about this analogy to 19 the PRA event. And then you identify the task in the 20 data source, the context. Then model the task with 21 IDHEAS combination failure mode and model the context 22 with PRA. Therefore, at the end, we will get not 23 human error probability, most likely a human error 24 rate of a failure mode for the human PIF.

25 MEMBER BLEY: Jing?

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

74 1 MS. XING: Yeah?

2 MEMBER BLEY: Sorry. I keep interrupting 3 and telling you to go faster. We are awaiting your 4 peer review work that's been done on the data report.

5 But I will make one comment for me. I've read it and 6 I have a lot of trouble understanding what I find in 7 the appendices.

8 And I don't know if you're still working 9 on that or not or if you think that's all complete.

10 But it's not transparent to me yet. I'll keep working 11 on it. Go ahead.

12 MS. XING: Yes, actually, I fully 13 understand when you say is not transparency to see how 14 we come from the data sources. We bound them to a 30-15 page report to one line in the appendix table. So 16 yeah, I agree that documentation part, we intend to 17 publish that documentation.

18 MEMBER BLEY: Good.

19 MS. XING: Okay. Any question?

20 (No response.)

21 MS. XING: Okay. So I will keep going.

22 So we do the same to another data source. So it put 23 the entire data application, but it will still end up, 24 it will end with the error rate of some failure mode 25 and some other PIFs. So that's what we call it data NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

75 1 generalization, generalize them into the same format.

2 So the process with the first evaluation 3 is assess the data source, locate the context, the 4 variable measurement to use the uncertainty in the 5 data source and user generalization by representing 6 them with the CFM and PIF, and finally, integrated the 7 analyzed data for our purpose. And I would like to 8 talk about the data sources. I would say over the 9 last decade since the beginning we collecting human 10 error data, we probably would do several thousands of 11 research papers that had numbers on human error rate 12 of related measures.

13 So we have collect them and based them in 14 five categories. The first category are nuclear 15 simulated data and operation data such as the NRC's 16 SACADA database collecting operator simulator 17 performance. And the HuREX is a similar database that 18 carry the current nuclear power plant operators and 19 the German's nuclear power plant maintenance database.

20 So those are the sources for this category.

21 And the second category is operation 22 performance data from other domains that are delegated 23 to some action in nuclear power plants such as 24 transportation, offshore oil, manufacturing. And a 25 lot of the data come from experiment studies in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

76 1 literature when the combinated (phonetic) behavior 2 sites the human factors in the neuroscience 3 literature. That we have thousands and thousands of 4 the literature.

5 And we also collect data from expert 6 judgment. Also we don't use them to calculate HEP.

7 We use them for verification.

8 MEMBER BLEY: Jing?

9 MS. XING: Yeah?

10 MEMBER BLEY: I'm going to go back to C.

11 Oh, first, we urged you to do this a long time. It's 12 a massive job and that you've taken it on is really 13 wonderful. Experimental studies in the literature, 14 many of them I've read are kind of graduate school 15 projects that people are doing. And so for those kind 16 of studies, have you found ways to adapt what you see 17 there to somehow account for the fact of the relevance 18 of expertise to the tasks they're doing in some way 19 that relate to what people would do in nuclear power 20 plants or in other professional fields?

21 MS. XING: Well, it's a talking detail.

22 I can talk about that all the way to five o'clock 23 today.

24 MEMBER BLEY: Is there anywhere you've 25 documented that? I don't know that I saw it in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

77 1 data report, but maybe you have documented that in 2 there.

3 MS. XING: It's not documented, but that's 4 a very good point. So we should document it.

5 MEMBER BLEY: It's really key for some of 6 this, at least for some readers who say, I don't think 7 that's relevant at all. And if you can explain the 8 relevance, and you can't write books on each of these.

9 I know that. But somehow clearing up the relevance to 10 the particular element of human performance that 11 you're addressing, I think it would add a lot of 12 confidence to users of the data and ECA, of course.

13 MS. XING: Okay. I think I was at least 14 probably add appendix just on the selection of data 15 sources. And to make --

16 MEMBER BLEY: I know this keeps 17 ballooning. But it's --

18 MS. XING: Yeah.

19 MEMBER BLEY: -- so much work already that 20 to not make it clearer would be a disservice.

21 MS. XING: Yeah. So maybe the next slide 22 can sort of answer your question, not specific on this 23 category. So this slide is about data source 24 evaluation. And especially for those experiment 25 literature. We look at the participants, like they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

78 1 are last -- at the last meeting, some are asking why 2 we use the normal thought --

3 There's a typical word used in experiment 4 literature means -- for example, one study can study 5 the people with a color vision deficiency. So that's 6 not -- we're not going to use that data. And for the 7 participants and there are plentiful from the past, 8 and there's a good sample size in the study.

9 And we look at the measurements used. Of 10 course, human error is preferred. Sometimes the test 11 of performing measures relating to human error rate, 12 we also take that. And the specificity, if the 13 experiment of the original data source give a clear 14 description of the task and the context, therefore we 15 can say what are the CFMs, what are the PIFs. You can 16 identify those.

17 And that's what I say. I can talk about, 18 find out how to evaluate it. It varies, 19 uncertainties, in the data source. And we look at 20 uncertainty in the data source.

21 Ideally, uncertainties are controlled.

22 They made a specific control on the uncertainty. And 23 basically they talk about what are the uncertainties, 24 how they would affect the results. That's what we 25 need this for.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

79 1 And also the breadth of representation, 2 that's related to Dr. Bley's question. I can give a 3 rough estimation. We probably select less than one 4 percent of the experiment literature we reviewed for 5 our -- codified for our data sources. So in regard to 6 them not codified because of more reasons than this 7 five elements.

8 MEMBER BLEY: So that's really 9 interesting. And I think having that appendix you 10 talked about would really enhance this report.

11 MS. XING: Okay, thanks. And we 12 generalize basically for each piece of data source we 13 selected. We likely performed that and applied 14 IDHEAS-G and then take the generalized data, document 15 it in IDHEAS-DATA structure.

16 I guess they had 27 tables, one table for 17 each performance influencing factors. That's Table 1 18 to 20. And we have seven other additional tables to 19 capture other information we need in HRA. Like, where 20 was the HEP of failure mode which means though 21 apparent PIF, there still can be some hidden of 22 uncertainty there. You're still getting a low HEP 23 rate.

24 And how the PIF interaction document in 25 the data on the effect of more than one PIF come to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

80 1 play, how they get combined. The attribution of time 2 needed to complete a task and largely how the various 3 factors modify the time needed. There's the data that 4 you perform this task in daytime or in the dark, for 5 example.

6 And we also collect -- I wouldn't say 7 those -- probably we shouldn't call that data. But we 8 call that empirical evidence on dependency, on 9 recovery of human action, and the main drivers to 10 human events. The last three tables are still 11 preliminary. We just started.

12 So a quick summary where we are in IDHEAS-13 DATA. By 2020, we documented and generalized the data 14 in the nuclear operation simulator data in SACADA, 15 HuREX, and some human studies, not all. So far, we 16 generalized somewhere between three to four hundred 17 paper literatures of data sources.

18 Another 200-plus were selected for 19 generalization. We hadn't got time to do it. And the 20 generalized data were independently verified and 21 revealed by PNNL. So it would be a NUREG report on 22 their evaluation which will fill in the data in 23 IDHEAS-DATA how you going down from 100 pages to two 24 lines.

25 So in the future, we do need human error NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

81 1 data for teamwork and organization factors. They're 2 not related to these errors, but they don't talk about 3 the human errors. The performance used cannot somehow 4 relate to human error.

5 Overall, data generalization is an ongoing 6 continuous effort. And the data integration should be 7 periodically updated. That's the -- I think Sean will 8 talk about that at the end. Okay. Any questions on 9 IDHEAS-DATA?

10 (No response.)

11 MS. XING: Okay. So I will move to the 12 two examples of using IDHEAS, the 2018 FLEX expert 13 annotation and the 2019 FLEX evaluation. So the 14 objective for 2018 FLEX expert annotation, as Sean 15 talked earlier, one, because NRR at that time had the 16 (audio interference) reviewed the PRA applications 17 related to crediting FLEX equipment. And we -- sorry, 18 yeah. And I did at that time did not directly 19 generate HEP numbers.

20 So we intend to develop an application-21 specific method from IDHEAS-G for that purpose. So we 22 go back to the message we want to first, a better 23 sense to quantify some HEPs of representative FLEX 24 action to give us some benchmark where the HEPs are 25 likely to be. And we do that for using FLEX equipment NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

82 1 to do the FLEX design scenario and for added defense-2 in-depth during the non-FLEX design application.

3 That's what our NRR users need.

4 And another important purpose is to 5 evaluate the performance shaping facts -- performance 6 enhancing factors in IDHEAS-G. And with the 7 information we hope that maybe we can select a subset 8 of PIFs that are unique for use of FLEX equipment.

9 And also, we --

10 (Simultaneous speaking.)

11 MEMBER BLEY: Well, are you going to say 12 more about that?

13 MS. XING: Yes, for short, we can say we 14 select the object. We were not able to select the 15 subset. That's why IDHEAS-ECA had all 20. So yeah --

16 MEMBER BLEY: Okay. That doesn't surprise 17 me. In your report on Item 1 here, you point out that 18 you kind of try to generalize these to be appropriate 19 at most places but not all. And then in the ECA 20 report, they were very specific to plan. So the 21 details of plan specifics and scenario specifics I 22 assume is what makes number 2 not quite work the way 23 you had hoped.

24 MS. XING: Yeah, like, especially when you 25 come to the FLEX design scenarios, everything can NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

83 1 happen. All those PIFs can play a role. And the 2 first objective as we also said, the intention was to 3 have the experts to quantify the contribution of the 4 PIFs, the HEP.

5 And that's when we ask them to think about 6 the macrocognitive functions, how this PIF would 7 change the error rate, the probability of failure 8 detection and failure decision making. And again, I 9 would say we didn't achieve the objective. But we got 10 a very useful message from our expert panel.

11 The experts were given the human error 12 data of (inaudible) by different PIF. And they said, 13 you already have this data. You should use this data 14 to develop a method, not from our judgment. So that 15 was a good message we got.

16 So the expert elicitation process, it was 17 sponsored by the NRC. And we use the white paper 18 guidance in place that, yes, we already talked. And 19 the author of the white paper guidance, we didn't 20 quite follow the process in term of coming -- have an 21 expert come up with a probability distribution.

22 So what we did, we had an extensive data 23 set on HEP, gave those to the expert. And we had five 24 meetings and one face-to-face workshop. The expert 25 panel consists of three NRC staff and three industry NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

84 1 experts who are knowledgeable in PRA, HRA, and 2 implementation of office of FLEX strategies and the 3 maintenance practices at a nuclear power plant.

4 So we had a former shift supervisor 5 operator, and we have an expert from PWR Owners Group.

6 And we also have an NRC expert approve text (audio 7 interference). And the expert presented two scenarios 8 and a FLEX design scenario and which means in the 9 scenario, basically IDHEAS-G is followed by an SBO.

10 So when IDHEAS-G is done, you will prepare 11 to think about use the FLEX equipment as a backup.

12 And the FLEX design scenario and which is the SBO 13 caused by a super severe -- by a severe external event 14 with super strong winds and flooding. And we evaluate 15 five FLEX actions of practice use of proper generator, 16 proper pumps within a water storage tank, ELAP 17 restoration and disabled shed.

18 And one thing we did use, this is an 19 answer to Dr. Bley's earlier question. One thing we 20 did offer the expert to come up with a scenarios. We 21 characterize them with IDHEAS-G performance utilizing 22 factors.

23 And so this was an iteration process.

24 Expert will come up with the performance shaping 25 factor. We were able to ask more detailed information NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

85 1 to experts. So experts will fill in more official 2 information.

3 And for example, this is the environment 4 context for the non-FLEX design scenario. It was a 5 design that the environment has no impact. It had no 6 weather. But it was a normal day. And there may be 7 some water in the plant due to lots of ups and downs.

8 They bring down some debris which can cause difficulty 9 to transport FLEX equipment.

10 It was -- we purposely said it was cold.

11 But the experts, it's cold, but not to the extent 12 making error or unable to work. So in the FLEX design 13 scenario, so we have this data change. You've got 14 poor lighting for some parts of the work. Darkness, 15 fog, smoke, and dust, all this could happen.

16 And there are some places the water's 17 accumulated so the cold can have difficulty to the 18 travel path. And the physical resistance faces a 19 lapse in environmental PIF. So you got strong winds 20 that would focus the debris to intake structure.

21 Therefore, you have difficulty. Experts have 22 difficulty access the covered path, and it's very 23 cold.

24 So I -- well, already we talked probably 25 it doesn't make sense to average -- do an arithmetic NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

86 1 average of expert's judgment. But I will just show 2 the -- illustrate some insight we learned from expert 3 elicitation. So this would result from declare ELAP.

4 The action is declare ELAP by 60 minutes 5 if power is not back within four hours. So you need 6 this judgment there. And so expert actually come up 7 for high, high, high and stays here on the average.

8 Pretty much, every expert gave a higher HEP for the 9 non-FLEX scenario compared to the FLEX design 10 scenario.

11 So we got a pages and pages of experts' 12 insight. What are the challenges to perform this 13 action? What are the ways to facilitate this action?

14 And what are the uncertainties we don't know?

15 That's those insights to our project team 16 are more valuable than the HEP number. For some 17 example insight, we had expert talk. The information 18 was incomplete with uncertainty. I don't have any 19 information yet right this minute because the 20 diagnosis of the work done, these are generated, takes 21 longer than one hour. So I need more information to 22 make a decision.

23 And more importantly, why they got a high 24 HEP for this because at that time the -- let's see --

25 FLEX support guidance, FSG, were not integrated with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

87 1 the OPs. In the non-FLEX scenario, the expert 2 consider the OPs is there, if they're not. And the OP 3 direct them, restoring the power. Then first the 4 supply team restoring the power instead of going to 5 ELAP, and that was a good lesson learned.

6 So compared to the non-FLEX scenario, the 7 FLEX design scenario has zero uncertainties. So even 8 you don't know when that these are generator, if I 9 open a door, you see the way it is outside. You know 10 you should be declaring most likely. You should 11 declare ELAP.

12 So both are considered real valuable 13 insight we gain from this expert. So this is the 14 example I promised earlier. It talk about the HRA 15 morbidity due to uncertainties in the scenario.

16 Let's take a look at the example, the 17 action, DC load shed. We specified it's open 18 18 breakers in two locations. We didn't specify in the 19 first place the expert come to the questions when 20 going through the PIFs and what is specified.

21 During the uncertainty, some uncertainties 22 in the scenario, that is the lay ups and the labels 23 of the breakers. Some plants have FLEX specific 24 labels that are shiny so you can see it. Some plants 25 don't. And who does the work? And the travel path to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

88 1 the breakers, do you need an elevator to exit the 2 room? Will the elevator work when you have these 3 single strategies and the like affect the stresses?

4 So these factors can change from plant to 5 plant. Even the same plant can change from scenario 6 to scenario. Those are the uncertainties. I mean, 7 we, of course, in the expert presentation, we can make 8 assumption of this. But we appropriately left that 9 open to the expert.

10 So let the expert use their guess of their 11 assumption of this uncertainty. But we ask them to 12 document what their assumptions are. So let's look at 13 the two expert.

14 Expert A got a very relatively high HEP, 15 0.2, as he consider this. I mean, no matter even if 16 it's in a non-FLEX scenario, whatever get you into 17 using FLEX equipment, it got to be something terribly 18 wrong. So operators would be in high stress. And 19 there's variations in the economic interface. And 20 they are entering with the breakers that they need to 21 open for FLEX -- using for FLEX.

22 On the other hand, if we look at Expert C 23 got a much lower HEP. The justification was these are 24 the similar actions to what operators' performances 25 made on a routine basis, just to open and close the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

89 1 breaker. And schedules should have no impact. So 2 this, we can say how the uncertainties in the scenario 3 carried us into the ASME HEP. So any question, 4 comment on this example?

5 MEMBER KIRCHNER: Jing Xing?

6 MS. XING: Yeah?

7 MEMBER KIRCHNER: This is Walt Kirchner.

8 I think someone else has their mic on. Let me step 9 back. Let me try again. What does the 1 percent, 99 10 percentile mean?

11 MS. LUI: Jing, you need to unmute your 12 mic.

13 (Pause.)

14 MS. LUI: Jing, your mic is muted.

15 MEMBER BLEY: And whoever has a phone 16 number ending in 03 is not muted.

17 MEMBER KIRCHNER: Yeah, there's a guest on 18 the line that's causing feedback.

19 MS. XING: That was an automation. I 20 didn't unmute it.

21 MS. LUI: So the 03 number is the public 22 bridge line.

23 MEMBER KIRCHNER: Someone on the public 24 bridge line has their mic open then.

25 MS. XING: Oh, can you hear me now?

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

90 1 MR. PETERS: Yes, we can hear you, Jing.

2 MEMBER KIRCHNER: Yes.

3 MS. XING: Okay. Thank you. Yes, the 4 question was, what happen at 1 percentile and 99 5 percentile? And the other expert made a long list of 6 their justification and the potential uncertainty that 7 they see. The 1 percentile represent for the 8 uncertainties in their consideration, everything is 9 ideal.

10 Like in baseline, I may not -- like I 11 said, may not match what's in the report of what 12 baseline you would consider. Okay? It's the 13 experience the operators did the work and that there's 14 no problem on the travel path. You have a clear label 15 on the breakers that go to the 1 percentile. So 16 several experts actually said if the breakers that had 17 specific FLEX label, that would make an order of 18 magnitude difference in the HEP.

19 MEMBER KIRCHNER: I should've been more 20 explicit, Jing. What is the basis? Is this the 21 numbers that are using the IDHEAS GE software? These 22 are the numbers that are generated?

23 MS. XING: No, these are the numbers the 24 expert estimated.

25 MEMBER KIRCHNER: Wow. That's a precision NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

91 1 that I find, well, questionable. Let me say it like 2 that. Not to judge the experts. These kind of 3 numbers with these significant decimal points is what 4 you get when you run critical heat flux experimental 5 tests. And you run several hundred data or more data 6 points to get that kind of precision. So just given 7 that we're dealing with humans, I just question the --

8 certainly down at the 1 percentile.

9 MR. PETERS: We're looking at it -- if we 10 think about it from the reliability perspective, when 11 you look at a 0.2, it's not really a very precise 12 number. What we're talking about is 2 out of 10 13 people fail. If you look at the 0.01, 1 out of 100.

14 So it's really a rough estimate that we 15 put down into a really refined decimal. Now if we 16 had, like, 123 out of 1,000, okay, well, that's very 17 precise, right? But 1 out of a 1,000 is not -- it's 18 just a rough guess from the experts.

19 MEMBER KIRCHNER: But Sean, then -- and 20 Jing, how do you actually use those numbers? I mean, 21 I know you're trying to deal with uncertainty. But I 22 just -- I'm scratching my head figuratively here 23 thinking that kind of precision is what you see from 24 experimental data with many, many tests. I get the 25 50th percentile, but just I'm questioning the tales, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

92 1 I guess, with such precise numbers.

2 MS. XING: Yes, our expert also had 3 mention it in the training for coming up the 4 probability. And they also expressed basically coming 5 up to the 1 percentile and the 99 percentile. So the 6 training we gave them for calibration, just think in 7 the worst case everything -- for the all uncertainties 8 you have in your mind, everything goes bad. And how 9 many of your crew would fail this action out of 100 10 times that they try this?

11 MEMBER BLEY: So if I could help a little.

12 Walt, I think your envisioning a precision that's not 13 intended. On that first slide for Expert A, a 0.5 14 would mean it's a toss of a coin. It could go either 15 way. And they're saying, well, it's not quite that 16 bad but it's almost that bad. And then you expect the 17 D, the difference between the 50th and the 1st 18 percentile is saying about 1 in 10 which is a really 19 high failure rate. You don't see that in most things 20 people do in the plant.

21 MEMBER KIRCHNER: Yeah.

22 MEMBER BLEY: And this is going to be a 23 lot higher than we usually see in the plant. It might 24 be a factor of 10 less than that. It might be a 25 factor of 3 higher. But you can't get much higher NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

93 1 than a 0.3 or a 0.5, even if everything is going 2 against you, it's hard to get to a guaranteed failure 3 unless it's an --

4 MEMBER KIRCHNER: Yeah.

5 MEMBER BLEY: -- impossible situation.

6 There isn't -- I don't think there's the precision 7 you're seeing in what the estimates mean.

8 MEMBER KIRCHNER: Yeah, yeah. Thanks, 9 Dennis. Yeah, that's what I was thinking too. The 10 first one, for example, in my simple-minded approach 11 to it is to say, well, there's 0.2, 1 out 5 chance 12 that it's not going to work. And then yeah, probably 13 things are really compounded. It's twice as bad. So 14 you come up with 0.4.

15 If I -- thank you. If I look at it in 16 that sense, fine. But it's just the visuals for me 17 just kind of misled how much precision really is --

18 MEMBER BLEY: I feel what you're saying.

19 MEMBER KIRCHNER: -- achievable.

20 MEMBER BLEY: But you got to get used to 21 these a little. A 0.5 is about as bad as it gets, the 22 toss of a coin kind of thing where it goes. Unless 23 it's just locked in, there's nothing you can do. Then 24 there isn't much question about human performance.

25 You just can't do it.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

94 1 MEMBER KIRCHNER: Yeah, you just can't do 2 it if it's very cold and everything freezes. You're 3 just -- that particular function impacted by that 4 environmental factor, yeah, the chance of it going, 5 it's just not going to work.

6 MEMBER BLEY: Yeah, the 0.5 is probably 7 getting close to what you're talking about, and that's 8 probably somebody who didn't want to say 1 in 100 and 9 wanted to say maybe it's not quite that bad.

10 MEMBER KIRCHNER: Yeah, okay.

11 MS. XING: I didn't mention the definition 12 for failure in this action. It's very restricted.

13 It's failure of any of the 18 breakers is considered 14 a failed action. That's another factor contributing 15 to consider.

16 MEMBER KIRCHNER: Okay. I read this 17 differently. I thought there were 18 breakers and two 18 locations to deal with so that you -- to be 19 successful, you had to open all of them. Okay. Thank 20 you.

21 MEMBER DIMITRIJEVIC: That was one of my 22 questions for you, Jing. The definition of the --

23 there is no clear definition of the failure here. Is 24 it to open all breakers and in what time frame? So I 25 mean, it's not really -- I mean, but I have -- when I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

95 1 read your -- the report, it seems like you provided 2 the same qualitative information to all the experts.

3 So they have the same understanding of the scenario, 4 right?

5 MS. XING: On a large scale, yes. But on 6 the microscale, people may still have some, what Dr.

7 Bley said, are the sub-scenarios. Like, some people 8 would think, oh, I will always access there. Well, I 9 have to go through elevator. Elevator was down, no 10 other ways. The staircase was blocked. So there are 11 always sub-scenarios that they don't -- even if we ask 12 them to document everything, all their assumptions, 13 there could be assumptions they just took it for 14 granted.

15 MEMBER DIMITRIJEVIC: Just to give to 16 Walt some additional information. Those scenarios 17 were not supported by PRA because PRA was not 18 developed for the FLEX scenarios in this time. So the 19 success criteria and timing and everything is based on 20 the different FLEX procedures.

21 MEMBER KIRCHNER: Right. Thank you, 22 Vesna.

23 CHAIR SUNSERI: Hey, this is Matt. I have 24 a question, just a time check. It's 20 after the 25 hour2.893519e-4 days <br />0.00694 hours <br />4.133598e-5 weeks <br />9.5125e-6 months <br />. We're scheduled to go to 1:00. We know we have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

96 1 some stakeholder input that we're going to receive at 2 the -- public stakeholder input at the end of the 3 meeting. So I just want to be mindful of the 4 schedule.

5 MS. XING: Okay. Thanks. Appreciate 6 that.

7 MEMBER BLEY: Yeah, Jing. I think you can 8 put a little light on details in the examples. We 9 have all the reports if we want to delve into them in 10 more detail.

11 MS. XING: Okay. Thanks. So mainly the 12 insight we got from 2018 expert elicitation. So we 13 kept this particular technical community's knowledge 14 about the uncertainties, challenges, and opportunities 15 in FLEX in this FLEX action. And when estimate HEP 16 are valid only for the very specific assumptions and 17 specifications we made for the scenarios and the 18 action in the study so that do not recommend people 19 use this HEP for their PRA application.

20 And yeah, this is what we said earlier.

21 We find we have -- we need to use all of the PIF for 22 in FLEX. And the expert recommend we should use human 23 error data to inform the HRA being measured. That's 24 what we did in ECA.

25 Okay. So I'll quickly go on the 2019 FLEX NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

97 1 HRA evaluation. So this was received by NRC and EPRI.

2 So the main purpose was to evaluate several 3 represented FLEX action and using IDHEAS-ECA and also 4 provide feedback for improve IDHEAS-ECA. And in this 5 study, teams of FLEX and the HRA experts modeled 6 several FLEX action.

7 Both teams had two plant data to better 8 understand FLEX strategies associated with equipment 9 to operate the action. And the FLEX expert create a 10 set of realistic scenarios and the HFEs for using FLEX 11 equipment. Then the HRA experts start to modify the 12 scenarios and quantify the HEPs using IDHEAS-ECA 13 software.

14 The expert had a three-day workshop to 15 finalize their -- to discuss their analysis and 16 finalize their quantification. The three scenarios 17 were -- one was beyond the design basis seismic event.

18 It's a PWR that result in SBO and the loss of the 19 water. So you need to deploy the FLEX pump. And the 20 SBO is pre-stage FLEX diesel generator.

21 Okay. So I'll probably spend some time to 22 say another source of HRA variability which is analyst 23 practice as an example. So this scenario is beyond 24 the design basis, a seismic event. So it's obvious 25 that power cannot be restored quickly.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

98 1 So for that, we had three cases. The base 2 case is easy. By 60 minutes power not back, you're 3 going to declare ELAP. By the second case, Case 2 is 4 like what we had in 2018 expert elicitation. If AC 5 power cannot be restored within four hours, you need 6 to declare ELAP within one hour of no AC power.

7 Case 3 is pretty much the same. You have 8 same instruction as Case 2, but the situation is less 9 obvious that the power cannot be restored. So there's 10 big uncertainties there.

11 So we look at -- let's take a look at Case 12 2. In this instance, here are the five analysts' 13 estimation. Look at A and B. They both chose the 14 same failure mode decision making. And they both 15 chose the same PIFs. Information is unreliable or 16 uncertain. That PIF has a human error entered in.

17 They both chose the number 2.

18 And some justification prevails and 19 actually they think will be very dependent on the 20 details of what the procedure guidance would say. And 21 the level ranges from 3 to 5 given some examples 22 presented to the team. So that's the uncertainty in 23 that event.

24 But now if we look at the other three 25 experts, they select the same failure mode. But they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

99 1 ultimately know the PIF has no impact. So why is the 2 information uncertainty has no impact?

3 Their justification was uncertainty to the 4 plant. But the operators, they talked and told them.

5 So by our time frame, it's set in stone. A decision 6 have to be made. Therefore, even the information is 7 uncertain, they would declare ELAP.

8 So this is the analyst's interpretation or 9 belief. So in this situation, we wouldn't -- these 10 two group of expert gave no other -- HEP had one order 11 of magnitude difference because of that information, 12 uncertain information, uncertainties. We wouldn't say 13 which one is right and which one is wrong. I think 14 both group capture something important about declaring 15 ELAP. The potential pitfall in the first group and 16 the way that you can mitigate that pitfall by 17 improving the procedure of the instruction.

18 MEMBER BLEY: Jing, a quick question.

19 MS. XING: Yes?

20 MEMBER BLEY: Were they given a chance to 21 talk to each other and understand why they decided 22 differently on this? It could be they come from 23 different plants and they actually work differently.

24 Or it could be they don't have as much experience and 25 they think people will do what they're told without NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

100 1 ever being confused. There's a lot of reasons that 2 could be driving it.

3 MS. XING: Yeah, all of the experts 4 visited both plants. So in the, you probably saw the 5 information, it should be equal. However, and I think 6 Susan Cooper got the lead of this project. She also 7 told me that every operator they talked to told them 8 they would declare ELAP. And that's where I think 9 they got the information that they got delayed that 10 one hour frame is set in stone. But in --

11 MR. PETERS: Jing, Susan wanted to jump in 12 with an answer real quick to Dennis' --

13 MS. XING: Okay.

14 MR. PETERS: -- question.

15 MS. XING: Susan, go ahead.

16 MS. COOPER: Thank you. Susan Cooper, 17 Office of Research. Just a couple clarifications.

18 Not all of the HRA analysts went on all of the plant 19 trips. But all of them had participated in 20 discussions to get agreement and a common 21 understanding of the scenarios, the associated HFEs in 22 context, and plant site visit notes.

23 I'm not remembering the specifics of who 24 is who here. But I think Jing is correct that even 25 though every -- all of the analysts had access to the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

101 1 same information and one of my objections was to make 2 sure they had the same understanding as best I could.

3 There still are indications in the results that 4 individual analysts may interpret or that 5 understanding in a different way. Or there may be 6 other information, again, based on their experience.

7 All of the analysts -- I mean, Jing did 8 mention. So we had three analysts from industry, all 9 of whom were quite experienced. And then we had one 10 analyst, an analyst of the Office of Research, and 11 then we had one inspector, and SRA.

12 So most -- all these people were very 13 experienced and had some operations background. But 14 they obviously had different background. And some had 15 more experience with FLEX and multiple plant sites 16 than others.

17 MEMBER KIRCHNER: I would just make a 18 quick observation. This is Walt Kirchner. What 19 you're seeing there in those highlighted 20 justifications is culture. I'll let you think about 21 that.

22 MS. COOPER: You could be right.

23 MR. PETERS: And you're right that with 24 people's backgrounds, they filter differently.

25 Absolutely.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

102 1 MEMBER KIRCHNER: Yeah, organizational 2 background is reflected there.

3 MS. XING: Yes, and also it was quite 4 interesting. This was very different from what we 5 heard in 2018 expert elicitation. Back then, the 6 expert has more -- they had information that people 7 are more hesitant in using declare ELAP. So maybe 8 that plant make them improve the guidance or 9 instructions on how to use the FLEX equipment. That 10 could reflect that change.

11 MEMBER BLEY: Yeah, that was the period of 12 time we were getting briefs from the industry on this.

13 And the approaches were changing pretty rapidly over 14 that time. So it's a good point, Jing. But you got 15 to hustle along. There's a lot to cover.

16 MS. XING: Yeah. Okay. So anymore 17 questions?

18 (No response.)

19 MS. XING: I'll move on. So some insights 20 we learned from 2019 FLEX evaluation, we see the 21 analyst variability generally is between an order of 22 magnitude for most human action. But this still 23 remained a concern.

24 Even you use IDHEAS-ECA, but we saw in the 25 PRA there's still uncertainties in the scenario NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

103 1 variation in HR analysts' practice. We think we found 2 the measured probability and support a better 3 understanding where the variability comes from.

4 Therefore, make it easy to recall finding the 5 variability.

6 MR. PETERS: Hi, Jing. This is Sean. I'd 7 like to make one real quick interruption. What we've 8 found with this one order of magnitude is in the U.S.

9 and the international empirical studies, we tended to 10 average around the order of three orders of magnitude 11 variability back then. So I don't know if this means 12 we can prove variability. But it's definitely (audio 13 interference) --

14 MS. XING: Sean, you're breaking up.

15 MR. PETERS: -- we were seeing in those 16 other reports. Sorry. I'm done. I was just saying 17 that our variability was lower than what we saw in 18 those other reports.

19 MS. XING: Thank you, Sean. And important 20 insight we learned from this exercise was, as Dr. Bley 21 mentioned earlier, the procedures for using FLEX 22 equipment in our FLEX scenarios are important for 23 predicting FLEX in PRA. And that plan has been made 24 lots of improvement on that since FLEX was initially 25 invented. Okay. Anymore questions on this part?

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

104 1 MEMBER DIMITRIJEVIC: I have a question, 2 a general question in this quantification -- because 3 this was from quantification, right? Even if it's not 4 obvious, it looks like, you know, it's from this 5 station because you know, there's two PFIs.

6 How did time component contribute? Was 7 the time component evaluated? Because your 8 probability of HEP has two components, right? One is 9 time related that we get from those time 10 distributions, another one which is the PFI related.

11 So was the time component part of this quantification?

12 MS. XING: I think the answer is -- Dr.

13 Cooper can correct me when I supply this. I think in 14 the study to make it easier, the assumption is 15 throughout the human action, they have adequate time 16 to perform the action. That was actually a 17 requirement in FLEX audit. Susan, do you have 18 anything?

19 MEMBER DIMITRIJEVIC: So you just assume 20 that they have the time required, the time available 21 where that was not contributing? Is that what you're 22 saying?

23 MS. COOPER: No.

24 MEMBER DIMITRIJEVIC: Okay.

25 MS. COOPER: So I think specifically for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

105 1 declaring ELAP, I think that was the HFE. We did look 2 at a timing contribution. There is a feature in 3 IDHEAS-ECA that allows you to do that. No one had 4 completed that -- tried that before the workshop. We 5 walked through it ourselves, identified some 6 difficulties, made a calculation, and it was 7 negligible.

8 The timing -- it could've been that 9 declared ELAP or it could've been FLEX DC motion. I 10 can't remember which. But it was one of those in the 11 FLEX scenario.

12 So -- and then the other ones had more 13 time available. So mostly due to lack of time, we 14 didn't pursue that. And then to the non-FLEX 15 scenarios, because we didn't have any timing 16 calculations or estimates that were based on 17 engineering, we made assumptions. We didn't evaluate 18 it at all for the non-FLEX scenarios.

19 MEMBER DIMITRIJEVIC: Thanks. And my 20 other question was you never really -- what I noticed 21 within one or two there is some connection discussed 22 in the text. The expert elicitation was used for 23 benchmarking. But actually, you never really 24 connected those two volumes in any way, right?

25 There is not any -- you did not really --

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

106 1 for the base HEPs, you didn't look in that 2 presentation. And there is a big difference also in 3 the HEP values because Volume 1 and Volume 2. And I 4 notice some limited discussion-wise that one was done 5 in 1980 and one was done in 1990. So obviously, it 6 seems like you learn much more about FLEX scenarios 7 between the one we have. But basically, those two 8 volumes don't talk with each other.

9 MS. COOPER: If you don't mind, Jing, this 10 is Susan, I'll answer real quickly. So the FLEX 11 expert elicitation effort was used as a lessons 12 learned for approaching the using IDHEAS-ECA FLEX HRA 13 effort. There was some overlap of personnel, 14 especially among industry members, those that helped 15 us develop scenarios and so forth.

16 But so far as the HEPs, no, there wasn't 17 anything done there. The only thing again is a 18 lessons learned. From the expert elicitation effort, 19 my job to my mind was to make sure the scenarios that 20 were selected and described were as realistic and as 21 detailed as possible and that the HRA analyst had as 22 close as this identical understanding of those 23 scenarios in that context as possible.

24 MEMBER DIMITRIJEVIC: And because when the 25 measurement comes to variability, variability is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

107 1 discussed within this volume. But if you look at the 2 variability between two volumes, then you can see, 3 like, this three order of magnitude differences, like, 4 declaring ELAP, going from 0.5 to 1.0 to the -3. I 5 mean, so there is a big variability in the -- if you 6 compare the HEPs from the different volumes. That's 7 what I want to say.

8 MS. COOPER: Yeah, I think some of that 9 was due to changes in the industry. I think also 10 having specific details. I mean, although we made 11 some departures from all plant-specific details, it 12 was for a particular plant. So could this be a 13 factor?

14 Or I remember at my plant, this is a 15 factor. So I'm going to apply that. I try to excise 16 that kind of stuff from the analyst. So some of it I 17 think could be improvements and others could just be 18 that we did a better job of constraining the problem.

19 MEMBER BLEY: I think we got down to the 20 last 20 minutes. John has some important stuff to get 21 to and we have comments from one member of the public 22 too.

23 MR. PETERS: And what I would say is 24 maybe, Jing, we should skip the modifications since 25 the last versions of the report. I'd like to propose NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

108 1 that. That would save us a few slides and we can just 2 go to a summation. The ACRS members can read those 3 slides and see if they have any questions associated 4 with them.

5 MEMBER BLEY: I think that's a good idea.

6 I mean, you guys on those slides point out that you 7 had many reviews and that you tried to respond to as 8 many of them as you could. And I think Jing makes a 9 point on slide seven tear downs and rewrites. This 10 report started very rough and has gotten much more 11 coherence. And part of that's due to your very hard 12 work and part due to what you got from commenters.

13 MR. PETERS: And Jing, if you'd just like 14 to mention briefly the people that did do the 15 comments, that would be great.

16 MS. XING: Yes, so this is the -- as we 17 said, we have been through many reviews. I just want 18 to take this as an opportunity to express our team's 19 appreciation for all the review input. The review 20 input is not just for improvement. It's actually part 21 of the development of this product.

22 So for example, one early ACRS 23 recommendation point out this important commodity time 24 effect. That lead to our development of the time a 25 certain model as part of quantification. And a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

109 1 reviewer, Dr. Emily Rothman, had gave us a very useful 2 comment on teamwork. So that solved us a big puzzle 3 of how to model teamwork and lead to the development 4 of the fifth macrocognitive functions, interteam 5 coordination.

6 And Dr. Mason (phonetic) and Ken, of 7 course, our own NRC staff, they had a very 8 comprehensive comments to say, our 2016 version. That 9 version, I think, had 18 chapters or something like 10 that. They said you need to come up with a cohesive 11 methodology. And that lead us to develop eight steps, 12 IDHEAS-G process and the standalone method for human 13 event analysis.

14 And at this time, we keep this kind of 15 list in our files. It goes up to several hundred 16 pages. So this is the opportunity want to express our 17 thank you.

18 And I will not talk about the details of 19 our revision since 2019. But I'd like -- again, I'd 20 like to take this opportunity to thank our -- the 21 comments we received from Mr. John Stetkar. He gave 22 us very thorough, thoughtful -- not just a comment but 23 constructive suggestions and recommendations.

24 We addressed most of the comments. For 25 the ones we couldn't address at this moment, those are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

110 1 very useful directions for our future improvements.

2 So having said that, I think I can end my 3 presentation.

4 MR. PETERS: Thanks, Jing. I'd like to 5 share my slides real quick. Oh, wow. I'm on the 6 wrong slide. Sorry, guys. I'll go real quick through 7 here.

8 So just wanted to go a little bit and talk 9 about our path forward from this point. And the 10 future work that we're going to be doing for IDHEAS is 11 we're going to be finalizing a publication of Rev. 0 12 of IDHEAS-G. This was already signed out in the 13 December time frame through our offices.

14 We are also refining and rolling out 15 IDHEAS-ECA. We're including the dependency model.

16 We're looking at improved guidance for recovery. And 17 we are going to be integrating that with the SAPHIRE 18 and SPAR models.

19 And we're looking at publishing a revision 20 after our public comment period that I mentioned 21 earlier in the April time frame. And of course, we're 22 still completing the IDHEAS-DATA project. You've seen 23 the draft IDHEAS-DATA report.

24 We're going to be taking the revisions 25 that are from our internal reviewers. And we're going NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

111 1 to be sharing this with the public. And we're going 2 to be taking public comments and incorporating 3 revisions based upon the comments plus the new data 4 that we have coming forth.

5 And other HRA work that you will be seeing 6 coming out also, we have a side project we're working 7 on, on understanding better what environmental and 8 human error probabilities are. We have a lot of work 9 we're working with our international counterparts.

10 We're trying to get more international partners to 11 supply data and exchange data so that we can even put 12 this out in the public domain so we can get work from 13 the rest of the world helping to analyze human error 14 data.

15 And of course, the last three that we 16 always have, there are big things that are being 17 worked right now. I've seen people starting to work 18 on organizational factors out in academia and in 19 industry. We would love to continue the work that 20 we've been doing on the Commission.

21 And the Golden Cybersecurity PRAs are 22 becoming of very high interest to the industry as a 23 whole. And the path forward, I think this is the time 24 to get a discussion even after the public comment 25 period here or the public comments we have at the end NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

112 1 of our meeting on whether or not we think we can close 2 SRM-M061020 because from the NRC's perspective, from 3 a research perspective, we developed a complete and 4 practice HRA method. We've made significant 5 improvements to the current state of practice and 6 developed some state-of-the-art techniques into the 7 methodology, especially based on timing was one of 8 those that was state-of-the-art that was recommended 9 by the ACRS.

10 It's human-centered scientific database, 11 so it can be modified to future applications. I think 12 it's a little easier than some of our older 13 methodologies. And we have a program for periodic 14 updates. So that is all, and I'd like to complete the 15 presentation. And thanks to the committee for this 16 time.

17 MEMBER BLEY: Thanks, Sean. The only 18 thing we didn't talk about today is SACADA, the data.

19 At one time, you were actively seeking additional 20 participants to provide data into SACADA. Is that 21 happening, or are we --

22 MR. PETERS: We still are. Yeah, we still 23 are. It's always through fits and starts. So it 24 usually takes about a half year to a year to get 25 initiated through a concurrence process and then NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

113 1 another half year or so to train the teams up. And 2 there's been a lot of turnover in the industry.

3 So we will get some industry members that 4 say, yeah, we'd like to participate and then, like, 5 there's a management turnover or a change in 6 operations. And they've decided not to invest money 7 into doing that. But the other side of the business 8 is we have a lot of international interest in it.

9 So we're currently working through our 10 internal governmental processes to try to get 11 international partners on this also. So more to be 12 heard on that. I'd like to come and present what we 13 have here in another year or so because we're just at 14 the early stages of building of that international 15 relationship.

16 MEMBER BLEY: Okay. That sounds good.

17 But we'll also have to hear how we adapt international 18 data because there are some practices that are 19 different as you go around the world. But thanks very 20 much, Jing and Sean. I want to get to the public 21 comments. But first, are there any members who want 22 to ask a question or two before we do that?

23 (No response.)

24 MEMBER BLEY: I take it this time we'll 25 ask former Chairman of the ACRS and now member of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

114 1 public, Mr. John Stetkar, to make comments if he's 2 available. And I would ask Chris if we've received 3 Mr. Stetkar's comments in writing too.

4 MS. LUI: So while Tom is opening up the 5 bridge line, I do not have Mr. Stetkar's upcoming oral 6 statements. But Mr. Stetkar's comments are all 7 available from the ADAMS -- publicly available from 8 the ADAMS.

9 MEMBER BLEY: Thank you. John, are you 10 there?

11 (Simultaneous speaking.)

12 MR. STETKAR: I hope I'm here. Can you 13 hear me okay?

14 MEMBER BLEY: We can. Please go ahead.

15 MR. STETKAR: Excellent. For the record, 16 my name is John Stetkar. I'm a former member of the 17 ACRS as Dr. Bley mentioned. And I am speaking today 18 as a member of the public.

19 First of all, I'd like to add my 20 congratulations to Chris Lui for her new role in 21 research. I have very fond memories of working with 22 her when I was an ACRS member. And I hope, Chris, 23 that you'll have just tons of fun over there in RES.

24 I'd like to comment briefly on the IDHEAS 25 general methodology in NUREG 2198. I know that the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

115 1 committee has received a copy of all of my detailed 2 comments on the previous version of the report. In my 3 opinion, the current version of the report has 4 addressed a vast majority of those comments, and I 5 very much appreciate the staff's stamina and their 6 extensive efforts to consider my comments and make all 7 of those changes.

8 I do have a few remaining high level 9 concerns about the methodology in NUREG 2198.

10 Considering the brief time available here, I'll just 11 highlight a couple of those most important issues.

12 First, I'm still concerned about the lack of technical 13 justification for the quantification model that's 14 represented by the equations in that report -- Section 15 4 of the report.

16 In particular, I still don't understand 17 why the primary influence on human performance is 18 determined by 3 specific base performance influencing 19 factors while the other 17 factors are cumulative 20 modifiers. I also don't understand why the assumed 21 linear summation of those performance influencing 22 factor weights is justified. I've reviewed the 23 IDHEAS-DATA report, and I understand the committee has 24 received a copy of my comments on that report.

25 I could not find any compelling NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

116 1 justification in that report for the format of the 2 quantification model. Furthermore, Section 6.3 in 3 Appendix D in the current version of NUREG 2198 have 4 removed examples which were originally intended to 5 support the conclusion that linear addition of the 6 performance influencing factor weights provides the 7 best method to account for the composite effects. The 8 current version of the report relies primarily on only 9 qualitative assertions that the model is justified 10 based on reviews of other studies without any 11 quantitative examples that clearly support that 12 justification.

13 Second, I think that the examples in 14 Appendix M of NUREG 2198 are very important for 15 prospective analysts to understand how the methodology 16 is applied in practice. Again, because of the time 17 available, I only have a couple of comments on those 18 examples. My first comment is related to 19 documentation of the analysts' decisions that are made 20 in each example.

21 As been mentioned today, one of the most 22 important objectives of the IDHEAS methodology is to 23 reduce sources of analyst variability in their 24 analyses. As Jing noted on Slide 6 of her 25 presentation, a critical task to achieve that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

117 1 objective is clear documentation of the rationale and 2 justification for a large number of analyst decisions 3 that are made throughout the process. For example, 4 the examples in Appendix M only document those 5 decisions very partially.

6 For example, they simply list specific 7 cognitive failure modes and performance influencing 8 factors that the analysts have left for each 9 evaluation. They do not document why other cognitive 10 failure modes for performance influencing factors were 11 excluded as being not relevant. It's very important 12 to document the rationale for those exclusions because 13 that rationale can often reveal key sources of 14 disagreement between different analysts and the 15 reasons for that disagreement.

16 In practice, I found that the 17 documentation process by itself also often prompts 18 each analyst to more carefully consider the basis for 19 their own judgment and selections. The same comment 20 applies to lack of documented rationale for selection 21 of a particular form of the uncertainty distributions 22 for the time estimates and the assignments of specific 23 parameters in those distributions, for example, the 24 5th and 95th percentile values. So in summary, I 25 think to provide instructive examples that demonstrate NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

118 1 the expectations of how the IDHEAS methodology should 2 be implemented, I really think that the examples in 3 Appendix M should better document the rationale -- the 4 analysts' rationales for their decision, including why 5 they excluded particular performance influencing 6 factors and cognitive failure modes as being not 7 relevant.

8 My second comment on Appendix M is that 9 the quantification example in the current version of 10 Section M.2.6 has been revised, but it's still not 11 correct. If you look at the combined uncertainty 12 distribution in that example, it evaluates the time at 13 which power is restored from a FLEX generator. That 14 distribution shows that there's a small but not zero 15 probability that power is restored before the crew 16 begins to extend -- use the extended loss of AC power 17 guidance.

18 In other words, there's some probability 19 that power is restored before one hour in that 20 scenario. That's not realistic, and the example 21 should be corrected. Perhaps staff will need to use 22 other forms of the uncertainty distributions to 23 provide a more realistic estimate that's consistent 24 with the physical constraints of the scenario.

25 Now my comment is not intended to be an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

119 1 obsessively detailed critique of probabilistic 2 arithmetic. I think it's conceptually important for 3 the methodology because the examples should clearly 4 demonstrate the perspective of analysts. But the 5 uncertainty quantification is not an abstract 6 mathematical afterthought.

7 The uncertainty should be an integral part 8 of the analyses. They should account for realistic 9 engineering and operational assessment of personnel 10 performance within the physical and functional 11 constraints of the event scenario. If the uncertainty 12 analysis results provide non-physical conclusions, for 13 example, that you can restore power before you start, 14 then something is drastically wrong with those 15 analyses.

16 And finally, if the committee will indulge 17 me, I'd like to simply emphasize the fact that the 18 FLEX application example in RIL-202013 is a product of 19 several key methods and reports that deserve really 20 careful attention before conclusions are made about 21 that report. The presentations today discuss those.

22 The relationships among the IDHEAS general methodology 23 and NUREG 2198, the IDHEAS-DATA report, and the 24 IDHEAS-ECA application in RIL-202002 are shown on 25 Jing's Slide 7 and 8.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

120 1 Those relationships are pretty complex and 2 somewhat convoluted. So I think until the committee 3 has an opportunity to carefully examine the IDHEAS-4 DATA and ECA reports, it may be premature to make 5 specific conclusions or recommendations about RIL-6 202013. And with that, I think you still have three 7 minutes to go. So I will thank you very much for the 8 opportunity to make my comments and I will go on mute.

9 MEMBER BLEY: Thank you. We appreciate 10 your comments. Are there any other members of the 11 public or other people sitting in on the meeting who 12 would like to make a comment? If so, please state 13 your name and affiliation and make your comment.

14 MR. JULIUS: Yeah. Hi, Dennis. This is 15 Jeff Julius with Jensen Hughes.

16 MEMBER BLEY: Yeah. Hi, Jeff.

17 MR. JULIUS: And I echo John Stetkar's 18 recommendation or concern that, yeah, we should get 19 some careful look and feedback on the RIL-2020-13 20 because it is integrating a large amount of data. And 21 so some additional discussions or workshop or 22 opportunities for comment would be useful.

23 MEMBER BLEY: Thank you, Jeff. I 24 appreciate that. Anyone else like to make a comment?

25 MR. PETERS: Just to respond to that, this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

121 1 is Sean Peters. Yeah, we will be discussing that in 2 our April workshop with the -- or our April public 3 comment period.

4 MEMBER BLEY: Thanks, Sean. Last chance 5 for comments if anyone wants to make them. We can 6 close the public line now, Thomas. And I have a 7 question for Sean.

8 I've been really pulling together a lot of 9 strings on this letter. I won't -- we won't do our 10 deliberations at this point. The committee will get 11 together and talk about this. There's a lot of 12 information provided to us today that goes beyond what 13 I've read.

14 And the question for you, Sean, is we have 15 on our calendar to write the letter this week. But 16 unless you have a really pressing need, I wonder if it 17 would be acceptable if we did it at the March meeting 18 which would give me a chance to look through the 19 transcript and adjust and chase the information and 20 chase some of the threads. But if you need it at this 21 point, I think we can certainly do that. But what are 22 your thoughts on it?

23 MR. PETERS: Yeah, I don't think that 24 there's any time crunch from our perspective. As you 25 guys have seen, we have so much work that's going on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

122 1 and the limited resources we have. We're still just 2 trying to process all the comments and the changes and 3 incorporate all the date that we've been gathering and 4 develop the finalized reports that are out on this.

5 But you guys know that we also still plan 6 to get some new updates, both IDHEAS-G and IDHEAS-ECA, 7 as we get more and more information, more data, and 8 more feedback from the users and more feedback from 9 the advisory committee. So my thoughts are from a 10 timing perspective, I am not aware of any timing 11 restrictions on the SRM. And the SRM is more directed 12 to the ACRS. So from our perspective, the ACRS should 13 have the final say on timing of closing out that SRM.

14 MEMBER BLEY: Okay. Well the committee 15 will deliberate on this, this week. I wasn't thinking 16 that we'd invite you back in March. We would just 17 include that as part of our letter writing if we skip 18 it until then.

19 But thank you. We appreciate that. And 20 the revisions you're talking about, are you 21 anticipating that the RIL documents will turn into 22 NUREGs at some point?

23 MR. PETERS: We haven't made that 24 determination. One of the challenges with the RILs is 25 they are really snapshots in time, right? So the 2018 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

123 1 was a hypothetical plan as hypothetically practiced 2 from what we understood the industry practices were.

3 The 2019 report was a little more 4 realistic because we were able to actually go to a 5 couple plants and model their behaviors. But of 6 course, those behaviors continue to change. So I 7 think depending on how much feedback and how much 8 industry and public interest is in it, we can take 9 that interest and revise those reports and put it out 10 as a NUREG.

11 It really just depends on what level of 12 interest there is in those reports because in our 13 perspective, the effects expert elicitation was for a 14 very particular purpose of helping NRR in that interim 15 period before we had a method developed. The second 16 report was to show some of the evolutions in at least 17 how FLEX had been practiced up to that point and to 18 show that we could actually quantify it using our 19 IDHEAS-ECA method.

20 So to us, it was more of a we're not 21 saying this is the end all, be all of FLEX because 22 FLEX is very dependent on your very particular 23 situation at your site and what type of scenarios are 24 thrown at you. So I don't know -- in the back of my 25 mind, I don't know the full utility of putting that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

124 1 out as a revised NUREG. But that's something I'm 2 willing to be open to.

3 MEMBER BLEY: Okay. I think it's fair --

4 I can't tell you what will be in our letter. The 5 committee has to deliberate on that. But I think from 6 what you've heard, it's clear that with a few 7 exceptions, IDHEAS-G report looks pretty good.

8 The RILs, however, you heard a number of 9 questions raised about them. And the data report, as 10 you said, it's not complete yet. But at least in my 11 opinion, it's got a fair way to go to be not just the 12 data source you want it to be but clearly justified in 13 what it has to say.

14 I think the last question for Jing is Mr.

15 Stetkar brought up something that slipped my mind.

16 And when I read the new revision of IDHEAS-G, I didn't 17 remember that the statement about the preponderant 18 effect of three PIFs is still included in IDHEAS-G.

19 Is it in there?

20 MS. XING: Somewhere in the report.

21 MEMBER BLEY: Okay. That's --

22 (Simultaneous speaking.)

23 MS. XING: -- that would be in IDHEAS-DATA 24 report.

25 MEMBER BLEY: I know, and I've been NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

125 1 through the IDHEAS-DATA report. And I don't -- it's 2 not there yet.

3 MS. XING: But it's --

4 MEMBER BLEY: At least I didn't find it.

5 MS. XING: -- there. But it's embedded in 6 the -- in some lines. So we think of what is 7 important. We probably make a separate appendix to 8 extract all the information.

9 MEMBER BLEY: If you're going to really 10 claim that's the truth, I think it needs to be really 11 clear and justified because it isn't -- again, it's 12 not transparent yet. So I would like to thank Jing 13 and Sean for all their presentations and the 14 commenters from the public for their comments. We 15 appreciate them. And at this point, four minutes 16 late, I turn it back to the chairman.

17 CHAIR SUNSERI: Thank you, Dennis, and 18 thank you, staff, for the thorough presentation today.

19 All right. Members, it is a little after 1:00. We 20 are going to recess for lunch until 2:00. At that 21 time, we will resume with a presentation on Advanced 22 Reactor Code, Volumes 4 and 5. So any comments before 23 we recess?

24 (No response.)

25 CHAIR SUNSERI: Okay. We are recessed NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

126 1 until 2:00. Thank you.

2 (Whereupon, the above-entitled matter went 3 off the record at 1:05 p.m. and resumed at 2:00 p.m.)

4 CHAIR SUNSERI: We'll start with a roll 5 call. Ron Ballinger?

6 MEMBER BALLINGER: Here.

7 CHAIR SUNSERI: Dennis Bley?

8 MEMBER BLEY: Here.

9 CHAIR SUNSERI: Charles Brown?

10 MEMBER BROWN: Here.

11 CHAIR SUNSERI: Vesna Dimitrijevic?

12 (No response.)

13 CHAIR SUNSERI: Walt Kirchner?

14 MEMBER KIRCHNER: Here.

15 CHAIR SUNSERI: Jose March-Leuba?

16 MEMBER MARCH-LEUBA: Here.

17 CHAIR SUNSERI: Dave Petti?

18 MEMBER PETTI: Here.

19 CHAIR SUNSERI: Joy Rempe?

20 VICE CHAIR REMPE: Here.

21 CHAIR SUNSERI: Pete Riccardella?

22 MEMBER RICCARDELLA: I'm here.

23 CHAIR SUNSERI: Vesna Dimitrijevic?

24 (No response.)

25 CHAIR SUNSERI: I'll look at the list.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

127 1 Okay. Well we'll just go ahead. We have a quorum.

2 We will proceed. I'm sure she will join here shortly.

3 So our next topic is Advanced Reactor Computer Codes 4 Volume 4 and 5. Dennis Bley is the subcommittee 5 chair. At this point in time, I will turn it over to 6 Dennis.

7 MEMBER BLEY: Thank you, Mr. Chairman. I 8 just turned mute off. I take it that didn't affect me.

9 CHAIR SUNSERI: No, you're okay. Yeah, we 10 hear you.

11 MEMBER BLEY: Okay. Well we're continuing 12 on our last two volumes of the strategy to 13 implementation action plan, Volumes 4 and 5. And I 14 want to mention, Kim, that I wasn't here for the 15 meeting on Volume 4, but I have had a chance to review 16 the transcripts. And we're looking forward to hearing 17 your presentation. I'll turn it over now to Kim 18 Webber of Division of Systems Analysis in RES. Kim?

19 MS. WEBBER: Great. Thank you so much.

20 Good afternoon. And I really appreciate your time 21 reviewing our latest volume, Volume 4, called 22 Licensing and Siting Dose Assessment Codes, and Volume 23 5 which have plans for our Radionuclide 24 Characterization, Criticality, Shielding, and 25 Transport for the Nuclear Fuel Cycle.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

128 1 I am Kim Webber, and I'm the Director of 2 the Division of Systems Analysis in the Office of 3 Nuclear Regulatory Research. And I'm glad to be with 4 you today talking about our Volumes 4 and 5, which as 5 Dennis said appropriately, these are the last volumes 6 at least that we have planned at this time. Let's go 7 to the next slide please.

8 Okay. So with me today are John Tomon --

9 he's the Chief of the Radiation Protection Branch --

10 and also Don Algama, a senior reactor systems 11 engineer. And both of them are in the Division of 12 Systems Analysis in the Office of Research. Drew 13 Barto is also on the panel, and he's a senior nuclear 14 engineer from NMSS.

15 As you know, we've been working really 16 hard with staff in NRR and NMSS over the last year to 17 develop these strategies in Volumes 4 and 5. And we 18 believe they represent the most resource effective 19 approach for our codes and code development 20 activities. I'll provide a brief overview of the 21 status of the non-light water reactor code development 22 project which encompasses an introduction and five 23 volumes, and then I'll turn the presentation over to 24 John, Don, and Drew. So can we go to the next slide?

25 Many of you have seen this slide several NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

129 1 times. And really, I keep using this slide because it 2 has the important key message which is still the same, 3 and that is that we're really trying to do the best we 4 can to enable the regulatory offices to be ready to 5 perform their oversight responsibilities such as 6 licensing in a time efficient but safety focused 7 matter. To improve mission value, we're working hard 8 to deliver the tools, the expertise, and the 9 information in a cost effective and efficient manner 10 so that licensing can be completed on time and within 11 the allotted resources.

12 A key element of our strategy is 13 developing the codes and analytical tools and the 14 approaches to use those codes like the ones you see on 15 this slide and to have those codes ready to go for 16 potential use in safety analysis. Through code 17 development activities in our collaborations with many 18 organizations you see here on this slide, our staff 19 has been acquiring new knowledge about advanced 20 reactor design and phenomena important to safety, thus 21 growing staff expertise and analytical capabilities.

22 Additionally, they've been working really hard to 23 capture knowledge about these reactor designs and the 24 phenomena in the codes and in the code manuals that go 25 along with code development activities. Next slide NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

130 1 please.

2 To facilitate the agency's readiness, 3 NRC's near term implementation action plan was 4 developed in the summer of 2017. And the IAP is the 5 vehicle to execute the NRC's vision to safety achieve 6 effective and efficient non-light water reactor 7 mission readiness. As you know, the IAP includes six 8 strategies, and Strategy 2 focuses on computer codes 9 and knowledge to perform regulatory reviews which is 10 the focus of today's presentation. If we go to the 11 next slide please.

12 So last year, we completed the 13 introduction in Volume 3. That is the Revision 1 to 14 those. The introduction and the volumes were 15 completed. And they focused on systems analysis, fuel 16 performance, neutronic source term, severe accident 17 progression, and accident consequence codes.

18 We really appreciate the time you spent 19 reviewing those documents and engaging with us over 20 the course of several subcommittee meetings in 21 addition to full committee meetings. We also 22 appreciate your insights and the feedback that you 23 provided in the letter for those volumes.

24 (Simultaneous speaking.)

25 MEMBER BLEY: Kim?

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

131 1 MS. WEBBER: Yes?

2 MEMBER BLEY: I just wanted emphasize for 3 all the members that the introduction in Volumes 1 4 through 3 were you reissued -- you revised them after 5 our meeting --

6 (Simultaneous speaking.)

7 MS. WEBBER: Correct, yes.

8 MEMBER BLEY: And at least the 9 introduction is a substantial revision. I've looked 10 at the others and I'm not sure how substantial they 11 are. Maybe you can say something about that.

12 MS. WEBBER: Well so in general what we 13 presented at the subcommittee meeting during actually 14 2019, those were all draft documents that we shared 15 with you. And then through the subcommittee meeting, 16 we obtained substantial feedback. And that feedback 17 was considered and incorporated into the January 2020.

18 I think the date is January 20th, 2020.

19 So those Rev. 1 revisions account for the 20 feedback that we receive through the committee 21 meetings along the way. And then for Volumes 4 and 5, 22 we had a future plant design subcommittee meeting a 23 few months ago where we presented information on our 24 code development plans in the licensing and siting 25 dose assessment area in addition to criticality NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

132 1 shielding and accident analysis for the fuel cycle.

2 So each of the five volumes -- and when I refer to the 3 volumes, I'm really referring to numerically Volume 1 4 through 2, 3, 4, and 5 and not necessarily the 5 introduction.

6 But the introduction provides an approach 7 for the code development activities that are 8 represented in Volumes 1 through 5. And each of those 9 volumes identifies the computer codes that we plan to 10 use for our independent safety analysis that 11 identifies the gaps in code development capabilities 12 and data. It also has information about verification 13 and validation needs along with specific code 14 development tasks and methods.

15 And as we did with Volumes -- the 16 introduction in Volumes 1 through 3, we look forward 17 to the interaction today with you. And we also look 18 forward to the letter that we receive from the 19 committee. We'll use that information that we get 20 today and through the letter to revise Volumes 4 21 through 5. And our intention is to complete those 22 volumes in the springtime. I think in the April time 23 frame is what we're targeting. Next slide please.

24 So I just wanted to let you know that 25 these activities are really critically important for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

133 1 the readiness, and so to support NRR and their needs.

2 And we also try to communicate what our interim 3 activities are and the completion of those activities.

4 And so if you're interested in, excuse me, having more 5 specific information about our next steps coming up 6 and what we completed in 2020, that information is 7 available on the website.

8 A kind of high level summary, we talked 9 about the use of reference plant models. And the 10 notion behind the reference plant models is that we 11 build plant models based on publicly available 12 information. With those plant models and the publicly 13 -- excuse me, publicly available information, we're 14 then able to test the codes, validate the codes, get 15 those codes ready so that when we're being called upon 16 -- whenever we're being called upon to perform 17 confirmatory analysis, we hope to gain a time savings 18 when we have to use more design-specific plant DECs.

19 So that's the whole notion behind this 20 reference plant model approach. And so on the 21 external public website, it does identify several 22 reference plant models that have been completed 23 already. And this year coming -- this year in FY 24 2021, we'll complete a substantial number of those 25 reference plant models in addition to completing some NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

134 1 source term demonstration activities.

2 MEMBER BLEY: Kim, excuse me.

3 MS. WEBBER: Mm-hmm.

4 MEMBER BLEY: Just for the members, if you 5 haven't been up on that part of the public website, 6 it's worth a look. And that schedule is very handy to 7 have. But Kim, I take it an X in a darker color means 8 you're complete.

9 MS. WEBBER: Yes, so if you look towards 10 the far left and you see the green boxes with the X, 11 that designates that they've been completed. And then 12 the other colored boxes to the right indicate the 13 planned completions. That red line represents 14 nominally the date at which this particular chart was 15 produced. And so you can see as a function of time 16 our plans to complete the various activities that are 17 listed in that first column.

18 MEMBER BLEY: So if I look at Strategy 2, 19 I see there are three reports you expected to finish 20 at the end of last year but maybe you're not quite 21 done with.

22 MS. WEBBER: Well actually, no. I think 23 that schedule is pretty up to date.

24 MEMBER BLEY: Okay. I didn't see an X in 25 those ones, like the reference plant model for the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

135 1 fluoride salt.

2 MS. WEBBER: Yeah, I think our plans are 3 to complete that this year. So I think you're right.

4 I just can't see what's on the --

5 (Simultaneous speaking.)

6 MEMBER BLEY: I got a big copy. I'm 7 sorry. Go ahead though.

8 MS. WEBBER: I'm working from my laptop 9 and I can't really see where the boxes are right now.

10 But like I said, I think that schedule may be a little 11 bit out of date but not that far actually. I'm not 12 aware of any slips in or schedule. We're really doing 13 really great work and on time work. So I'm really 14 proud of the staff actually.

15 MEMBER BLEY: Perfect. Go ahead.

16 MS. WEBBER: Okay. And then if there 17 aren't any more questions, I'd really like to turn the 18 presentation over to John because I know he and Drew 19 and Don have a lot of material to cover.

20 MR. TOMON: So can everybody see my screen 21 and the slides?

22 MEMBER BLEY: Yeah, they're up. Hey, 23 John?

24 MR. TOMON: Yes?

25 (Simultaneous speaking.)

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

136 1 MEMBER BLEY: I missed that meeting in 2 September when you talked about this. If you can as 3 you go through, can you highlight anything that's 4 changed since that point in time?

5 MR. TOMON: Sure. There's only like one 6 or two -- I only added like one or two new slides, and 7 I've actually taken some out to pare it down because 8 I have less time this time around. So --

9 MEMBER BLEY: That's good. Go ahead.

10 MR. TOMON: Okay. So as Kim said, good 11 afternoon. My name is John Tomon. I'm the Chief of 12 the Radiation Protection Branch in the Office of 13 Research. And this afternoon, I'm going to discuss 14 Volume 4, the License and Siting Dose Assessment Code 15 Plan that my staff developed in collaboration with the 16 program offices and several of our code contractors 17 and developers.

18 This report describes the vision and 19 strategy to achieve readiness for non-light water 20 reactor designs for the license and siting dose 21 assessment codes. It provides an overview of the 22 technical issues related to the license and siting 23 dose assessment codes and the various non-light water 24 reactor technology and fuel designs that are being 25 considered. The staff and code contractors identified NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

137 1 several issues within the current suite of licensing 2 and siting dose assessment codes, which should be 3 addressed for all non-light water technologies while 4 at the same time continuing the code's applicability 5 to the current light water reactor fleet, issues such 6 as the number of licensing and siting dose assessment 7 codes, code function and capability overlaps, and 8 inconsistent and independent code development 9 throughout the years.

10 Working with our individual dose 11 assessment code developers and the radiation 12 protection computer code analysis and maintenance 13 program, the RAMP contractor, Pacific Northwest 14 National Laboratory, the staff developed the five 15 tasks listed on this slide to prepare the licensing 16 and siting dose assessment codes for non-light water 17 reactor readiness. These tasks included looking at 18 code consolidation and modernization, improved 19 characterization of source terms, improved atmospheric 20 transport and dispersion modeling, updates to dose 21 coefficient values, and updates to the environmental 22 pathway modeling used in some of the codes, and where 23 necessary, include additional radionuclides specific 24 for the non-light water reactor technologies.

25 Licensing and siting dose assessment codes, as shown NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

138 1 in this image from Volume 4, we're looking at the 2 possibility of having to make ready approximately 10 3 licensing and siting dose assessment codes for the 4 various non-light water applications.

5 Included in these codes are the 6 radionuclide transport removal and dose estimation 7 code or sometimes referred to as RADTRAD, the control 8 room habitability code or HABIT, the atmospheric 9 relative concentrations in support of control room 10 habitability code, ARCON, the ground level relative 11 air concentrations for accidental release code, PAVAN, 12 the gaseous and liquid effluent code, GALE, the normal 13 effluent dose assessment and siting code NRC dose 14 which includes the liquid pathway modeling dose 15 assessment code, LADTAP, and the gaseous atmospheric 16 pathway modeling dose assessment code, GASPAR, the 17 normal relative air concentration and relative 18 disposition factors code, XOQDOQ, the radioactive 19 material transport dose assessment code, RADTRAN, the 20 radiological assessment system for consequence 21 analysis code, RASCAL, the decontamination and 22 decommissioning code, D&D, the residual -- and 23 finally, the residual radioactivity code, RESRAD. In 24 Volume 4, we also included discussions on other 25 computer codes that either the non-light water reactor NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

139 1 designers are considered using in their applications 2 code such as the GENII code or GENII code or codes 3 which have inputs to the codes on this slide such as 4 the dose coefficient package code, DCSPAC, the SCALE 5 code, and the MELCOR code.

6 MEMBER KIRCHNER: John, this is Walt 7 Kirchner.

8 MR. TOMON: Yes?

9 MEMBER KIRCHNER: Is it fair to say that 10 the slide you just covered, essentially all those are 11 in current use for LWR applications?

12 MR. TOMON: Yes, sir. Yeah, they're all 13 used in one form or another and they're kind of 14 stovepiped. And that's going to get to the first task 15 we came up with, the code consolidation.

16 MEMBER KIRCHNER: Okay. All right. Thank 17 you.

18 MR. TOMON: One of the -- so the first 19 task we came up with and it was listed on my previous 20 slide, code consolidation and modernization. This was 21 one of the first tasks that became obvious to the 22 staff and it was based upon the number of licensing 23 and siting dose assessment codes and the number of 24 different types of non-light water reactor designs and 25 fuel types being considered. The staff decided that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

140 1 code consolidation where possible was an efficient 2 means of maintaining and writing the codes with the 3 resources available.

4 Code consolidation and modernization was 5 viewed as a means to help reduce the functional 6 redundancy between the codes, outdated science and 7 technology associated with the design, and the 8 development of those codes. The limited ability of 9 the current codes to assess advanced reactor designs, 10 a history of changing ownership and associated loss of 11 the code development knowledge over time, and the 12 inefficiency of having to maintain multiple codes.

13 Additionally, we were looking to implement this task 14 in phases depending on several factors such as the 15 timing of the non-light water reactor submittals and 16 the availability of resources, both staff and 17 contracting funds.

18 MEMBER BLEY: John?

19 MR. TOMON: Yes?

20 MEMBER BLEY: This is Dennis again and 21 then Joy has something. Have you gotten far enough 22 into this to have some idea of the difficulty of some?

23 From what I read, some of these codes are really 24 dated. You probably don't have much information on 25 them. You almost have to go through it and figure out NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

141 1 what it's doing to see what you have to change. It 2 might be easier starting from scratch on some of them.

3 Where do you stand?

4 MR. TOMON: Well actually some of the 5 codes have been undergoing some kind of maintenance 6 work throughout the years with our developers, our 7 code developers and our contractors. So they're 8 getting -- with trying to keep them up to date and 9 keep them up to date to the various operating system 10 platforms as computers change because some of them 11 were still working in 32-bit systems, they've had to 12 kind of go in and pull them apart a little bit for the 13 light water reactor fleet. So they know kind of where 14 all the skeletons, the faults, the traps, the huge 15 sections of commented code are.

16 So they have a pretty good idea of what's 17 in there, especially like in the atmospheric codes.

18 So combining them and then making the best use of 19 what's in there to the regulations. We think we have 20 a pretty good idea of that and that we should be able 21 to do it with the existing -- taking the existing 22 codes as a framework to build these consolidated 23 modules.

24 MEMBER BLEY: Okay, good. Because I read 25 more into what you had written. And so that's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

142 1 encouraging. Joy?

2 VICE CHAIR REMPE: Yeah, during our 3 subcommittee meeting -- first of all, I want to say 4 I'm really glad to see the consolidation effort going 5 on. And I understand that with the funding 6 uncertainty and timing uncertainty, the incremental 7 approach that you're taking is consistent with what's 8 talked about in the introductory report that Kim 9 mentioned earlier, that you've got a lot of 10 constraints as you make decisions. But in the report, 11 it referred to as I harped on during the subcommittee 12 meeting that it's a common misconception that you 13 couldn't go and take the MACCS code and have it 14 simplified and do what's done with RASCAL.

15 And before it was over with when I went 16 and reviewed the transcript, I think you said I guess 17 I'm not saying it's impossible. But because of what 18 we wanted to do, this seemed to be -- and I'm 19 paraphrasing -- a better way to go at this time. And 20 I think we're in agreement on that.

21 But are you planning any -- do we need to 22 put it in a letter to say, are you going to kind of 23 fix the words so it's more accurate? I'm more into a 24 factual correction here, or what's your thought on 25 what you're going to do about the comments that were NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

143 1 made during the meeting about this?

2 MR. TOMON: Well thanks for bringing that 3 up, Joy. Yes, I've spoken with the staff and I've 4 spoken with Kim. And as Kim alluded, we're going to 5 take input from that. And we're going to take inputs 6 from the subcommittee and specifically that section, 7 2.1.

8 You're absolutely right. The words that 9 we chose probably weren't the best in there to use 10 that. I mean really what we were thinking about it is 11 what -- and you paraphrased exactly how we're going to 12 change the words to say that we were looking at the 13 limited resources that we have and then looking at the 14 work that's being done on MACCS and being done for 15 codes like MELCOR and SCALE.

16 And we're going to put what we can from 17 there into RASCAL because basically the user community 18 for RASCAL looks at the code and wants it to act in a 19 certain way, both in time internal to the NRC and 20 external. So we are going to revise that section of 21 -- I think it's Section 2.21 and to take out probably 22 the -- now that I think about it, more and more, the 23 word misconception was probably a bad choice of words.

24 And I think we used it twice in paragraphs following 25 each other.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

144 1 But yes, exactly what you said. We're 2 going to build on what we -- what other code 3 developers and researchers do in research for the 4 MACCS code, MELCOR, and SCALE and put them into 5 RASCAL. So we're not going to try to go out and do 6 that separately because of the limited resources and 7 because of the uncertainty when certain things are 8 going to come when the certain applications are going 9 to come in. Does that help?

10 VICE CHAIR REMPE: Very much. Thank you 11 very much.

12 (Simultaneous speaking.)

13 MEMBER KIRCHNER: Yes. Joy, also I'd like 14 to observe that if and when they do this that we'll 15 address the recommendations of the last two biannual 16 research review reports on this particular area.

17 VICE CHAIR REMPE: Right. That's one of 18 the reasons that I wanted to make sure that this was 19 corrected. But I don't think it really merits a 20 paragraph in our letter. It's just a word choice and 21 I think we're all in agreement on that now.

22 MR. TOMON: Yes, ma'am.

23 DR. CORRADINI: This is Corradini. I'm a 24 consultant to the committee. Just a quick question.

25 The users of this are not just within the NRC.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

145 1 They're within the licensee applicant community. Have 2 you gotten input from them on how a consolidation 3 should be done or what they would recommend in a 4 consolidation?

5 MR. TOMON: We have not, not specific 6 recommendations. But we've had some general 7 recommendations. As part of the RAMP program, some of 8 these code designers and developers have joined the 9 RAMP, the Radiation Protection Code Analysis and 10 Maintenance Program, as a user and they use the code.

11 And they have given us feedback during the 12 meetings when we talk about what would be best to see.

13 And that's where we got a lot of feedback from the 14 developers specifically with regards to our three 15 atmospheric transport codes and that some of the 16 options in one code they've like to use. But it's not 17 available because -- just because the different -- the 18 way the codes were built individually and kind of 19 siloed. So we are using some of their feedback in our 20 code design, and that's how we went through our 21 thinking process.

22 DR. CORRADINI: Okay. Thank you. I was 23 going -- I didn't know the name of your user 24 community. So it's RAMP. Thank you.

25 MR. TOMON: RAMP, yes. Okay. So Slide 5 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

146 1 shows our code consolidation approach. With the 2 assistance of our RAMP contractor, Pacific Northwest 3 National Laboratory, we developed a three-pillared 4 approach to code consolidation, including first, 5 create consolidated engines. This is a set of 6 functional modules or engines that would be developed 7 to perform the regulatory calculations as those 8 performed by the current suite of licensing and siting 9 dose assessment codes.

10 The functional engine approach improves 11 development flexibility by allowing for future 12 modification and efficient data transfer.

13 Furthermore, separating these capabilities into 14 standalone engines eliminates some of the current code 15 redundancies and inefficiencies. The second was to 16 develop a standard data transfer schema.

17 Using a standardized data transfer schema 18 such as an extensive markup language for encoding the 19 data for each engine would make data input universal 20 and adaptable while making it easy to pass the output 21 data between the different functional engines. And 22 finally the last pillar was to build a single user 23 interface. The single user interface would be 24 developed separate from the functional engines that 25 would interact with the users and communicate with the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

147 1 functional engines to execute the user defined 2 commands.

3 The user interface will be designed with 4 the thought that to effortlessly guide the users 5 through the relevant code engine input screens 6 primarily through a series of questions about the 7 desired outputs. The next slide slows the conceptual 8 models for the consolidated code. This figure on this 9 slide is a conceptual diagram of the proposed 10 consolidated code paradigm, showing how the models 11 from the existing siting and licensing codes could be 12 integrated into this new consolidated code. Modules 13 within the consolidated code would be grouped or 14 characterized within this general dose assessment 15 approach.

16 In addition, the modules will be further 17 broken down into scientific disciplines to account for 18 the unique differences of these fields. The proposed 19 consolidated code would have several modules or 20 components, each of which will contain like 21 phenomenological models from the existing light water 22 reactor licensing and siting dose assessment codes.

23 The eight modules of consolidated codes include the 24 source term, including core inventories, release 25 fractions and timing sequences, the atmospheric NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

148 1 transport and dispersion modeling including near, mid, 2 and far field models, the aquatic pathway model 3 including ocean, river, and lake dispersion, 4 environmental accumulation, human biota consequence 5 modeling, non-human biota consequence modeling, dose 6 coefficients and health risk factors, and integrated 7 dose module.

8 MEMBER BLEY: John?

9 MR. TOMON: Yes?

10 MEMBER BLEY: Has your work thus far 11 stayed at the planning stage? Or have you actually 12 begun work on some of these modules?

13 MR. TOMON: We've actually -- we have just 14 gotten through the source requirement documents from 15 our contractor on the atmospheric -- consolidated 16 atmospheric transport module. So we should start 17 moving ahead now with the consolidated model. We 18 actually -- we took the report and we made sure that 19 NRR -- the meteorologists in NRR, they read through 20 it.

21 They commented. We addressed the 22 comments. And now we have a path moving forward for 23 that. The source term module, we just started 24 planning. And we expect to get a similar source 25 requirements document from them in the next month or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

149 1 so.

2 MEMBER BLEY: That's excellent. I'm glad 3 to hear it. Go ahead.

4 MR. TOMON: My next slide, this goes to 5 the second task, the source term task. The second 6 task is to identify source term input such as fuel 7 inventories, reactor cooling inventories, plant design 8 and operational data, where available, for each of the 9 non-light water reactor fuel designs and fuel types.

10 For normal operations, the radionuclides of interest 11 in the source term include fission products, CAPTCHA 12 products, and activation products produced during 13 normal operation in the reactor cooling system.

14 For accidents, both severe accidents, 15 beyond design basis accident, and design basis 16 accident work, the primary source term information 17 will be from the work on the MELCOR and SCALE codes as 18 described in Volumes 3 and 5. For transportation 19 source term, the various non-light water fuel types 20 vary significantly from the current light water fuel 21 configurations. In addition, some of the non-light 22 water reactor designs may adopt a mobile reactor type 23 approach where it is likely that the entire core 24 containing spent fuel will be transported in a single 25 shipment. Therefore, the transportation source term NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

150 1 module will need to take these issues into account.

2 MEMBER KIRCHNER: John?

3 MR. TOMON: Yes?

4 MEMBER KIRCHNER: This is Walt Kirchner 5 again. One complication, I'm sure you're thinking 6 about it, and I don't know at what step you start 7 integrating it. But when we get to some of the --

8 particularly the liquid fuel designs and also some of 9 the salt designs, you've got significant -- I'm trying 10 to choose my words carefully -- chemical toxicity 11 issues that are -- how should I say it -- code 12 travelers with any release of the radionuclide 13 inventory to the -- within the actual design, within 14 the plant, outside the plant, and to the environment.

15 MR. TOMON: So right now, none of the 16 codes that we currently have address chemical toxicity 17 issues, per se, because of the light water reactor.

18 So that is one of the things we are thinking about for 19 the future with some of those particular designs you 20 mentioned. But right now, none of the -- and we have 21 to figure out how to work that in.

22 That hasn't -- we really haven't gotten 23 very far on that portion of it because none of that is 24 -- we've had to deal with that with the light water 25 reactor fleet. And it wasn't something we were NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

151 1 thinking about initially. But it's something we have 2 to probably add to the capabilities of maybe a RADTRAN 3 code or something like that.

4 MEMBER KIRCHNER: Something that's 5 analogous from the LWR code development, particularly 6 the systems codes, first, the emphasis was on being 7 able to handle two-phase flow. But then it was 8 realized that carrying non-condensable gases or for 9 that matter when you get into a severe accident 10 situation, the hydrogen and oxygen became a 11 consideration for the system codes and MELCOR. So I 12 don't know to what extent the kind of physics methods 13 that were used there would fit into your current code 14 suite. But it's something to be thinking about going 15 forward.

16 MR. TOMON: We are. It's just that we 17 haven't gotten very far on that. And we plan to try 18 to use, like you said, leveraged research activities 19 from Volumes 3 and 5 as much as we can. Let them do 20 a lot of the work and then when we get into our codes, 21 not spend money twice as it were.

22 MEMBER KIRCHNER: No, no. That's smart to 23 do that.

24 MS. WEBBER: Thanks for the comment, Walt.

25 I think that's a good comment. I appreciate it.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

152 1 MR. TOMON: So some of the source term 2 considerations, like I said, we plan to work with the 3 other code development branches and research to 4 leverage activities from Volumes 3 and 5. We're also 5 working with our RAMP contractor, Pacific Northwest 6 National Laboratory, to leverage their source term 7 work and activities with the National Reactor 8 Innovation Center, including the National Reactor 9 Innovation Center fission product modeling approach 10 using publicly available information on the various 11 non-light water reactor designs and fuel types to 12 create categories of general reactor types.

13 And finally as an aside note, some of the 14 current licensing and siting codes we have now are 15 flexible in their current configuration to accept 16 source terms outside light water reactor designs. For 17 example, as currently constructed RADTRAD under SNAP, 18 the code will allow users to enter user defined source 19 term release fractions and timing sequences. However, 20 this is a little bit more involved process than just 21 selecting from the current hardwired pressurized water 22 and boiling water reactor options already in the code.

23 Additionally, the NRC dose code can allow 24 for the import of user defined normal or routine 25 source terms. However, there is currently no code NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

153 1 that will perform normal affluent reactor coolant 2 source terms for non-light water reactor technologies 3 like the GALE code does for light water reactors. The 4 next slide is a new slide you asked me to point, and 5 this is just recently from some work that we've done 6 with our contractor, Pacific Northwest National Lab.

7 And this slide depicts kind of the 8 methodology and the general concepts and strategies 9 that our contractors have mapped out for developing a 10 normal or routine source term for the various non-11 light water reactor and fuel designs. The proposed 12 methodology for the normal or routine source term will 13 draw again on that National Reactor Innovation Center 14 fission product modeling approach and will be similar 15 in concept to how the GALE code calculates normal 16 source terms for light water reactors. The 17 methodology will use built-in source -- built-in 18 origin source term data for each non-light water 19 technology and fuel design coupled with code features 20 to determine the fuel isotope concentrations, 21 calculate fission product release fractions to the 22 primary coolant based on the ASME 18.16 nuclide 23 classes or more if we need to develop more based upon 24 the actual coolant, the fission products in the 25 coolant.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

154 1 Determine activity concentrations in the 2 primary coolant, both from fission and activation 3 products and also in the secondary coolant if that 4 happens to be applicable to the design. And then 5 last, determine the liquid and gaseous waste streams 6 for each reactor design to include rates, activity, 7 and waste stream cleanup mechanisms such as hold up.

8 Additionally, the normal source term methodology, when 9 we initial develop it and then I imagine it'll stay 10 throughout, we plan on it to being very flexible to 11 allow for user defined parameters wherever possible.

12 The third task in Volume 4 for non-light 13 water reactor readiness involves the atmospheric 14 transport and dispersion modeling. Most of the 15 license and siting dose assessment codes have use or 16 have atmospheric transport dispersion models which are 17 typical Gaussian plume models. For example, ARCON 18 PAVAN, and the XOQDOQ code uses straight line Gaussian 19 models with different correction factors such as 20 building wake effects, wind direction, wind speed, 21 atmospheric stability class, location of release 22 point, stack down wash, and plume rise to adjust for 23 the code use.

24 The staff is looking to consolidate ARCON 25 PAVAN codes into a single atmosphere engine in a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

155 1 phased approached. Phase one would be to integrate 2 the atmospheric engine that would have the capability 3 of performing near field, mid field, and far field 4 calculations, thereby the user could perform the 5 regulatory calculations relevant to any of these three 6 distances. Phase two would involve adding 7 capabilities to the consolidated atmospheric engine to 8 support non-light water reactor technology siting such 9 as in remote areas with different atmosphere stability 10 class diffusion and dispersion characteristics.

11 What we -- this next slide is kind of the 12 general outline that we've got for our atmospheric 13 engine prototype. It shows examples of user inputs 14 and features that will be incorporated to the 15 atmospheric engine prototype developed during phase 16 one along with the data flow between the interface and 17 the engine. After selecting a dispersion distance 18 model, i.e. near, mid, or far field, the user could 19 provide source receptor inputs such as distance, 20 intake height, direction, using 2D and 3D graphical 21 displays.

22 The user would then choose a 23 meteorological file and visualize the wind 24 distribution. Train data could be imported from a 25 public database such as the National Elevation NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

156 1 Dataset. Also train heights for source receptors 2 could be extracted from elevation data and provide the 3 input to the atmospheric model.

4 Once the inputs are set up, the data shall 5 be transferred to an extensive -- using extensive 6 markup language schema to the atmospheric engine where 7 calculations are performed. And then after the 8 calculations are complete, the atmospheric dispersion 9 engine will allow the users to select various output 10 options for both reporting and plotting the results.

11 The fourth task involves the dose coefficient module, 12 developing a dosimetry module that has the flexibility 13 to use different integers, dosimetric models, and dose 14 coefficient values, and examining the dose coefficient 15 models with respect to aerosol particle size since 16 non-light water reactor technologies could release 17 particle size smaller than the one to ten micron range 18 that are typically considered in the current code.

19 Currently, the dose coefficients and 20 dosimetry models are hardwired into most of the codes.

21 And the user has few options to edit or change these 22 values. An important element of this task is to 23 design the module with the understanding of different 24 dose endpoints of siting and licensing of non-light 25 water reactor designs. The other dose coefficient NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

157 1 consideration is that this module will be flexible 2 that it would allow the user to select from Federal 3 Guidance Report 11 and 12 dose coefficients which are 4 used by the current regulations, as well as future 5 federal guidance report dose coefficients such as 6 those in FGR 15 and possibly 16 when that is released.

7 Another part of this module will consider 8 options to allow the user to select aerosol particle 9 sizes for the radionuclide which could directly impact 10 the calculated dose. And as I said, many of the 11 existing dose assessment codes with the exception of 12 the RADTRAD code do not possess for the user select 13 user defined dose coefficient values. We have a 14 couple codes like NRC dose and RASCAL which does allow 15 the user to choose between different dose coefficient 16 values from different hardwire dosimetric models such 17 as ICRP 26, ICRP 2630, ICRP 2, and ICRP 6072.

18 MEMBER KIRCHNER: John?

19 MR. TOMON: Yes?

20 MEMBER KIRCHNER: At this point, how much 21 of that you just covered is already in MACCS?

22 (Simultaneous speaking.)

23 MR. TOMON: I'm not sure. I mean I don't 24 know how much of MACCS actually uses DCFPAK. We're 25 planning to -- I'm sure they do. But I don't know to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

158 1 what extent and to what vintage that MACCS actually 2 has. I wouldn't be able to answer that. I'm not 3 really sure. I'm not a very fluent MACCS user.

4 MEMBER KIRCHNER: Well MACCS went through 5 a lot of effort to incorporate HYSPLIT, the 6 atmospheric model developed by NOAA. Maybe I 7 misunderstood, but I thought most of the engine you're 8 describing already existed in MACCS and that you would 9 extract that and --

10 MR. TOMON: Well --

11 MEMBER KIRCHNER: -- make it compatible 12 with your data and user interfaces. It seems to me 13 that you have put a tremendous amount of effort into 14 MACCS. What am I missing here?

15 MR. TOMON: Well that's -- I was talking 16 about the dose coefficients. And previously, I was --

17 the previous task was talking about the atm, which was 18 you're referring to. But you're right. I mean they 19 are also exploring in MACCS, I think, in changing the 20 near field modeling.

21 But a lot of what -- I think they're 22 planning to use some of the models that are in ARCON 23 for the near field modeling. Those models obviously 24 are in ARCON already, and they are also because ARCON 25 was developed in the same -- at the same contracting NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

159 1 place, Pacific Northwest National Lab, as RASCAL was.

2 A lot of those are already in RASCAL itself.

3 But we will -- if there are significant 4 changes to the far field modeling that are different 5 than what we use in a code like MACCS or further out 6 like XOQDOQ, we would definitely look to employ them 7 into the atmospheric consolidated engine. And that's 8 one of the thing we hope is that using the data schema 9 -- the XML data schema, it'll be more flexible and 10 we'll be able to do drop in and replace models as work 11 is done. And we find that model more efficient than 12 the current model we're using.

13 MEMBER KIRCHNER: Well maybe this is a 14 question for Kim and can be deferred. But in our 15 review of the RES efforts in the systems analysis 16 area, we spent a fair amount of time going over MELCOR 17 and MACCS and their attempts to use the same kind of 18 approach you're using, John, in terms of separating 19 the development of the physical models from the 20 development of the solvers and such so that you had 21 the equivalent of your engine packages. I'm just a 22 little concerned right now because our recommendations 23 in this area were to take the best estimate practices 24 and tools that were developing for MACCS and see if 25 you could extract those engines out of that particular NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

160 1 code and use them to backfill all of those specific 2 user codes that you have that you mentioned 3 previously. And I guess I'm missing something here if 4 MACCS -- is MACCS going to continue as a standalone 5 code?

6 MS. WEBBER: Yes, it is. So MACCS is 7 going to continue as a standalone code. The real 8 benefits of MACCS are to support a lot of precise kind 9 of detailed analysis that's needed to evaluate 10 consequences. And so the application of the radiation 11 protection codes and the codes that John is talking 12 about really go to ease of use of these codes.

13 Not to say that MACCS isn't easy to use, 14 but it does require a level of understanding of a lot 15 of input parameters in order to get some analysis that 16 kind of makes sense. So there's definitely a 17 substantial amount of skill that's needed to run the 18 MACCS code. And it is used external to the agency as 19 well.

20 So it definitely has a place not only in 21 doing research studies such as SOARCA and other 22 studies. But it also has a place with some of our 23 external vendors or developers as well. But these 24 codes, John's codes, these RPB codes, they also have 25 their place. And a lot of people choose to use those NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

161 1 because of the ease of use.

2 And so I can't really answer the detailed 3 question, Walt. We'll have to take that one back.

4 But I think your message really is where we have 5 capabilities in other codes regardless of the code, we 6 should leverage those capabilities to our benefit.

7 MEMBER KIRCHNER: Well yes, and I think 8 that was the theme of our recommendation over the past 9 two biannual research reviews in this particular area.

10 So --

11 MS. WEBBER: Yeah, it's a good one.

12 MEMBER KIRCHNER: Yeah, it's just a 13 concern on my part because your MACCS code is evolving 14 to be the state of the art, so to speak --

15 MS. WEBBER: Yeah, yeah.

16 MEMBER KIRCHNER: -- for atmospheric 17 dispersion and consequence analysis. So it would seem 18 to me if it's modularized, then you could take the 19 engine from that effort and plug it into some of your 20 legacy codes. And as far as the user is concerned, 21 the user doesn't have to know the --

22 MS. WEBBER: Correct.

23 MEMBER KIRCHNER: -- detailed physics of 24 the solver and for MACCS. Okay. I've made my point, 25 I guess.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

162 1 MS. WEBBER: Yes.

2 MEMBER KIRCHNER: Just I got a little 3 confused here because it seems like the efforts are 4 redundant with what's already going on for the MELCOR 5 MACCS modernization efforts.

6 MS. WEBBER: Yes. So we definitely will 7 take a look at that. I appreciate you raising it.

8 (Simultaneous speaking.)

9 VICE CHAIR REMPE: Go ahead. I'm sorry.

10 I thought you were done, Walt. Walt?

11 MEMBER KIRCHNER: I'm done. No, I think 12 --

13 VICE CHAIR REMPE: Well during the 14 subcommittee meeting, we spent a lot of time 15 discussing it, and it's a different approach. One 16 could have a simple user interface with MACCS and try 17 to take that on. But there are these other codes that 18 are simpler and consolidating them so you don't have 19 10 of them and go down to a fewer number. My 20 impression --

21 (Simultaneous speaking.)

22 MEMBER KIRCHNER: No, I agree with all of 23 that, Joy. My point was if indeed the internal 24 engines, to use John's terminology, of MACCS are state 25 of the art, it would seem to me -- and they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

163 1 functionally equivalent of the -- and I'm not going to 2 get all the acronyms right, ARCON, PAVAN, XOQDOQ, et 3 cetera, with the HYSPLIT addition from NOAA. Now 4 maybe that's too much to put in these other legacy 5 codes. Maybe it would not function. But it's just a 6 concern.

7 VICE CHAIR REMPE: Yeah. Well again, I 8 think two different paths could've been taken. Again, 9 as long as you get there and there is -- what can you 10 do with a little bit of money if you aren't sure it's 11 going to continue? Those kind of factors seem to come 12 into play too.

13 And so yeah, maybe there is some 14 redundancy. But as long as they are making progress 15 is why I kind of -- I wanted to make sure you didn't 16 say it was impossible to do the other path because I 17 don't think it is. It's just --

18 MEMBER KIRCHNER: No, no.

19 (Simultaneous speaking.)

20 VICE CHAIR REMPE: -- a lot of money, do 21 it for a long period of time, I would be pushing 22 harder to say let's go with MACCS and just have a 23 simpler user interface because ultimately you might 24 save money in the long run.

25 MEMBER KIRCHNER: No, I understand and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

164 1 appreciate fully. You've got a user community out 2 there. You've got immediate needs. And this path 3 that John is outlining may be -- especially to get the 4 flexibility you need for the non-LWR concepts may be 5 the best approach with all the boundary conditions 6 applied. But it just strikes me a little bit 7 different than the conversation we had when we did the 8 research reviews.

9 VICE CHAIR REMPE: Yeah, I agree. And 10 it's not totally clear to me, but I get what they're 11 saying.

12 MR. TOMON: Just to throw one other little 13 wrinkle in it is that ARCON, PAVAN, and XOQDOQ really 14 derived a lot of their equations and how they'll do it 15 from several reg guides. And the reg guides haven't 16 been updated in years, and that's one of the functions 17 that we want to do with consolidating the code. We 18 didn't put it on here, but work with the meteorologist 19 too because their fallback is the reg guides tell the 20 licensee how to do these calculations for. And that's 21 what PAVAN, ARCON, and XOQDOQ are set up to do.

22 But they can be revised since they haven't 23 been revised in a while. So going forward having this 24 one consolidated engine may be in the revisions of the 25 reg guides. We don't want to get too far in front of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

165 1 the reg guides. We want to make them come at the same 2 time.

3 But it might be pulling in the better or 4 different models from MACCS. They're more state of 5 the art and they can do the same functions. And 6 that's agreed upon with the meteorologists and their 7 licensing activities.

8 MEMBER BLEY: Are the same people going to 9 be working on the reg guides as they're working on the 10 codes then?

11 MR. TOMON: I'm going to -- well the plan 12 right now is talking to then -- in talking with NRR 13 is, yes, to have the contractor that's PNNL that's 14 working on the consolidated ATD code as once they get 15 close to having the consolidated code is to work on 16 doing the updates to the reg guides with, but making 17 sure that NRR is actively involved in any updates and 18 revisions we do to those codes, so yes.

19 MEMBER BLEY: Okay. Thanks.

20 MR. TOMON: So I only have I think two 21 more slides left. So hopefully, I'll get through them 22 quickly here. So my next slide, Slide 14, is it shows 23 the last task, the environmental pathways.

24 And this is going to be a task that we're 25 going to accomplish in the future down the road. It's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

166 1 not really dependent upon the non-light water reactor 2 designs and fuel types. So we're looking at a time 3 frame of greater than five years to actually look and 4 explore changing some of the models and the 5 environmental pathway.

6 This task will also look to explore 7 feasibility of radionuclide particle size as we 8 discussed before. And the task will also leverage 9 models from the GENII code and decommissioning codes 10 like RESRAD. My final slide shows some of the -- and 11 this is a new slide to answer that question -- shows 12 some of our near term planning and milestones for 13 license and siting dose assessment code readiness.

14 The staff determined that the dose 15 assessment codes would have to be updated for the 16 various non-light water reactor designs and phases 17 based upon the availability of resources and the time 18 frames of application. In our phased approach, we are 19 considering near term to be now through the next three 20 years, an intermediate phase three to five years, and 21 longer term greater than five years, with the ultimate 22 long term goal of reducing the 10-plus siting 23 licensing codes down to two or three by that five-year 24 point that accomplish the same regulatory functions as 25 the existing suite of codes.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

167 1 MEMBER BLEY: Hey, John?

2 MR. TOMON: As a matter of fact -- yes?

3 MEMBER BLEY: Could you go back a slide?

4 MR. TOMON: Sure.

5 MEMBER BLEY: You mentioned particle size 6 and behavior. Where do we stand on knowledge base for 7 those kind of chemical-related effects for some of the 8 new designs?

9 MR. TOMON: I really haven't seen an 10 environmental source term for any of the new designs 11 yet. So I'm not really sure. It just could be I 12 haven't seen it yet. But I know that -- I know 13 they've come up with generic kind of inventories for 14 the National Reactor Innovation Center. But I don't 15 know if they've come up with environmental inventories 16 that get at -- whether it's a normal or an accident 17 and what those -- what the radionuclides are in that 18 release and then basically what the particle size. I 19 haven't seen that as of yet, so I'm not really --

20 (Simultaneous speaking.)

21 MR. TOMON: -- sure where that it.

22 MEMBER BLEY: It strikes me that we really 23 need some experiments on this. Maybe Dave or somebody 24 has some thoughts on that.

25 MEMBER PETTI: Yes, so Dennis, I'm not too NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

168 1 worried down to half a micron. The equations are all 2 there, whether or not there's corrections as you get 3 smaller. So they're not -- so things could just be 4 not very difficult to add has to do with the size of 5 the particle relative to (audio interference) things 6 like that.

7 In terms of what is actually emitted from 8 the reactor design, that's going to be partly due to 9 -- on the shoulders of the applicant where there is 10 some data, for instance, on dust in a pebble bed 11 reactor. And I think it's probably reasonably 12 adequate they went and they got a distribution and 13 mean size, all the stuff you expect. I'm not so sure 14 about some of the other -- this stuff being done.

15 MEMBER BLEY: Yeah, I know the ACGRs have 16 looked at that a lot. But yeah, the other areas are 17 suspect.

18 MEMBER PETTI: Well, in the sodium, if 19 there's a higher -- we know what particle size it is.

20 There were tests done years ago. The only one is 21 salt. That's where things are a little bit less 22 clear.

23 I just had a question, John. My sense 24 here that there's a huge value in the consolidation of 25 the existing fleet. And let's call it the existing NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

169 1 regulatory infrastructure, all the stuff that happened 2 by people, whether it's the applicant, the NRC, even 3 I'd assume EPA and some of these other organizations 4 that we would need that. And so although a lot of 5 this has to do with the advance of the non-LWR 6 designs, this should be leaning into the 21st century, 7 if you will, for the LWRs as well. Is that fair?

8 MR. TOMON: Yes, that's the way we're 9 looking at it as well. Because yeah, it goes back to 10 what we said before, limited resources and everything 11 being built in kind of silos before and now trying to 12 get them all into the 21st century. Making this 13 consolidated code would make things a lot easier for 14 the existing fleet as well as -- and making it 15 flexible to add things for the non-light water 16 reactors as well.

17 And then the only other thing I wanted to 18 point out on this slide is that we are planning for 19 the development and piloting of the consolidated 20 atmospheric transport engine by the end of the second 21 quarter of fiscal year '22. And we will continue to 22 meet and collaborate with our code developers, our 23 contractors, and the counterparts in the program 24 offices in developing and coordinating implementation 25 plans for code consolidation. And that's my last NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

170 1 slide. And if you have any additional questions, I'll 2 be happy to answer them.

3 MEMBER BLEY: Thanks, John. I think we 4 can go ahead with the next talk.

5 MR. ALGAMA: Howdy. My name is Don 6 Algama. Oh, sorry, Ken. Let me know when you're 7 ready.

8 MR. ARMSTRONG: I'm ready, sir. Can you 9 see it?

10 MR. ALGAMA: Oh, yes. Thank you very 11 much, Ken.

12 MR. ARMSTRONG: Perfect.

13 MR. ALGAMA: Hello. My name is Don 14 Algama, and I'm here with Drew Barto, NMSS. I'd like 15 to present the staff's draft approach to develop 16 insights into fuel cycles from non-LWRs. It's 17 actually for non-LWRs and understand computer code 18 performance. And I want to make sure, as everyone 19 understands, is as we learn more, we might update the 20 process, particularly from information from the DOE 21 and vendors. Next slide, please.

22 I'd like to acknowledge a lot of great 23 people who helped produce this document from the 24 program offices, NRO, NMSS, and Research. And within 25 Research is Oak Ridge, in particular, Dr. David Luxat NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

171 1 and Dr. Will Wieselquist who were crucial in this 2 report. Next slide, please.

3 I won't spend too much time. This is a 4 summary of what Kim spoke about earlier today. Next 5 slide, please. The goal of the plan is to apply and 6 understand the performance of existing NRC tools to 7 support fuel cycle evaluations. And the hope is in 8 doing this we'll demonstrate our computer code 9 performance and readiness, and also staff will gain an 10 understand of what to expect or some insights into the 11 non-fuel cycle processes.

12 And the plan is intended to be updated as 13 we learn more from DOE and the industry of both the 14 designs and what we might expect from the actual 15 operations of these designs. The plan takes on a 16 delta approach using the existing LWR fuel cycle as a 17 reference. It's basically an incremental approach 18 comparing the candidate non-LWR design against 19 existing fuel cycle capabilities and developing on the 20 same spectrum hazards and accident scenarios.

21 As we are taking LWR approach, this means 22 in practice we will also be coordinating closely with 23 internal partner groups when the scenarios demonstrate 24 the need such as those in Volume 3 and Volume 4 and 25 NRR and NMSS teams. As in Volume 3, the input decks NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

172 1 and analysis notes will be made available to the 2 public upon completion. Next slide, please.

3 As I mentioned, the objectives -- the 4 results -- the reports ultimately demonstrate computer 5 code readiness and to understand how they perform. To 6 achieve this, we will have to look at developing 7 scenarios to identify potential hazards to assess 8 against. For this, we will look at both existing 9 guidance as well as anything available from the DOE 10 and design information that's public.

11 We intend to be flexible in our approach 12 and the plan will evolve as we implement as mentioned 13 before. The following list are some documents that we 14 intend to leverage to support criticality safety, 15 radionuclide inventory characterization, decay heat 16 estimation, radiation shielding, and radionuclide and 17 non-radionuclide hazard evaluations. As noted 18 earlier, anytime we have to deal with consequence 19 assessment areas, et cetera, they'll be raised to the 20 appropriate team at NMSS, NRO, and within Volume 3 and 21 4 as they occur. So we'll make sure to coordinate.

22 With 6410, the focus is on understanding 23 characteristics of releases of radionuclide material 24 and/or hazardous chemicals from non-reactor 25 facilities. Particularly, it includes a description NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

173 1 of process hazard analysis and other structured event 2 scenario development approaches that can be used to 3 support integrated safety analysis. That handbook 4 covers actual scenarios, including criticality events, 5 release of materials, the in-facility transport 6 depletion processes, and leak path factor estimation.

7 Table 2-2 of that report provides a range 8 of scenarios that we can evaluate. NUREG-1520 9 complements 6410 and covers 10 CFR 20 and 70 10 applications. The focus of this SRP is to ensure 11 quality and uniformity of reviews, and it provides us 12 with further insights.

13 With NUREG-2215, we move from facilities 14 to storage, Part 72. This focuses on certificate of 15 compliance for dry storage systems for use at a 16 general license facility, and two, a specific license 17 for a dry storage facility that is either an 18 independent spent fuel storage installation, ISFSI, or 19 a monitored retrievable storage installation, an MRS.

20 This SRP provides us with insights into what to look 21 for in storage conditions such as margins to 22 subcriticality, how it would prevent releases, et 23 cetera.

24 In NUREG-1567, this complements 2215 as it 25 applies to wet storage. With NUREG-2216, we move to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

174 1 transportation. That covers shielding, criticality.

2 And Table 1-2 provides an example scenario to 3 demonstrate subcriticality. And Attachment 2A of that 4 report provides staff expectations for computer codes.

5 Complementary DOE documents, one is an 6 example, it may be useful for developing hazards as 7 presented. But there are other DOE documents such as 8 DOE Standard 102792 which is for hazard evaluation 9 techniques; DOE Standard 30072007 which covers SER, 10 non-path facilities; and DOE Handbook 30101-94 11 provides airborne release fractions for non-power 12 facilities. These will all be reviewed in the 13 implementation phase to understand how best to 14 evaluate this scenario. Next slide, please.

15 This slide captures the basics of Volume 16 3 approach for non-LWR analysis and an illustration of 17 how we plan to apply each for Volume 5. As before, we 18 start with fundamental data processed by SCALE and 19 then passed as input to the severe accident source 20 term code MELCOR. Anything to do with consequence 21 analysis will be handled by either Volume 3 or Volume 22 4 as appropriate. Next slide, please.

23 With a scenario and understanding of what 24 is interesting, we can investigate existing code 25 performance in the areas of criticality safety, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

175 1 radionuclide inventory characterization, decay heat 2 estimation, radiation shielding, and radionuclide, 3 non-radio, and hazard evaluations. Areas such as 4 mining, milling, long term storage and disposal, 5 consequence, radiation protection, chemical toxicities 6 will be covered elsewhere. As with Volume 3, we 7 expect to reasonably apply comprehensive and 8 methodological approach from starting with scenario 9 definition, identification of items of interest, 10 identification of dominant phenomena we need to 11 capture through to V&V and appropriate documentation.

12 The design selected in Volume 3 will be 13 used as a driver for Volume 5 analysis approach. And 14 here we're looking at HPR, HTGR, FHR, SFAR, and MSR 15 classes. Next slide, please. Thank you. Right now, 16 we anticipate ten reports, as before we're flexible on 17 this. They're not shown in order and we are flexible.

18 As we learn more from the DOE and its 19 partners, we will adapt. This includes how we 20 prioritize work. The ten reports are broken down into 21 five reports looking at non-LWR specific fuel cycles 22 and five reports that cover common fuel cycle 23 activities.

24 The reason for this is to try to make use 25 of efficiencies and commonalities. For example, look NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

176 1 at HTGR and FHR, we anticipate that Reports 6, 7, and 2 10 will be common. So once we do it for one, it'll be 3 ready for the other one. Next slide, please.

4 So this is how we plan to begin our 5 approach. As mentioned, the LWR fuel cycle will be 6 used as a reference to understand the anticipated non-7 LWR fuel cycle. To make the task tractable, we broke 8 down each of the steps into six major steps and 9 several sub-steps.

10 These are labeled with the first letter of 11 the stage and a number with a sub-step. So 12 fabrication, for example, can be broken down to two 13 sub-steps, identified as F1, fabrication of UO2 14 facilities for example, and F2, fabrication of fresh 15 fuel assemblies. This work right now will not look at 16 scenarios of interest in the T3 and S1 steps due to 17 less information available on what DOE and industry 18 intentions are. We will revise as we learn more.

19 Next slide, please.

20 This is an example of what we anticipate 21 the stages will look like for an HTGR fuel cycle 22 report. The HTGR concept has been developed for 23 decades with experimental reactors starting from 24 DRAGON from 1963 to 1969, AVR from 1967 to 1998 to 25 HTGR which is 1998 and still operating, HTGR-10 from NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

177 1 2003 to the present, and through to commercial 2 operations from Peach Bottom 1 which operated from 3 1967 to '74, the THTR, Thorium High Temperature 4 Reactor, I like saying that, from 1986 to 1989 and 5 even Fort St. Vrain from 1976 to 1989.

6 For this work, as based in Volume 3, we 7 chose the PVR-400 to drive this class of analyses.

8 And there are two basic types of this reactor which is 9 pebble bed and prismatic. And they're using a pebble 10 bed type reactor.

11 We can expect there are hundreds of 12 thousands of pebbles in the reactor core, tens of 13 thousands of TRISO particles and online refueling and 14 helium as the coolant. So those are the basic 15 characteristics of this design. We've learned a lot 16 from the NGNP base.

17 So with our approach for this fuel cycle 18 that covers this reactor design, it'll be in a series 19 of products. So for example, in Report 6, we will 20 tackle the generic E1 and T1 activities which cover 21 UO6 enrichment in transport. We anticipate maybe some 22 primary hazards there with the enrichment facility 23 with chemical hazards through the release of UO6 and 24 the criticality hazards associated with enriched 25 uranium.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

178 1 In a trial phase of UO6, we will look at 2 different configurations of canisters and overpacks to 3 understand how the system responds. And Report 7 4 which is the F1 step, we will look at -- we will cover 5 TRISO fabrication. We will look at the various 6 fabrication steps and look for criticality and 7 radionuclide and non-radionuclide hazards that may 8 arise.

9 In Report 10 which is the F2 step and T2 10 step, we will look at fabrication of pebbles and their 11 transport to the utilization facility. In Report 4, 12 this report is expected to cover the utilization step.

13 Sub-steps U1, U2, and U4 stage, the U2 step will also 14 be coordinated to the Volume 3 report as it leverages 15 the work performed there. The U3 -- oh, excuse me.

16 The U3 step which is spent fuel pool shuffling as seen 17 in the reference cycle is not covered here. In the U4 18 step, we will look at onsite discharge of pebble 19 storage -- for pebble storage. Next slide, please.

20 So in conclusion, we believe that we have 21 a reasonable approach in the reference to delta 22 strategy. With the LWR fuel cycle used as a reference 23 to understand the non-LWR needs, we believe that the 24 development and assessment work being performed under 25 Volume 3 will cover the needs we expect to have in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

179 1 Volume 5. We believe that sufficient experience in 2 the application of SCALE and MELCOR to non-reactor 3 facilities exist to start the process.

4 This experience will be developed and 5 refined as we get more experience and information with 6 DOE and its partners. To develop experience in the 7 future we are also developing reasonable scenarios to 8 apply the codes. And we will leverage other NRC teams 9 including those in Volume 3 and Volume 4 as the 10 scenario dictates. Thank you. That's the end of my 11 presentation.

12 VICE CHAIR REMPE: Don, this --

13 MR. ALGAMA: Yes, ma'am.

14 VICE CHAIR REMPE: -- is Joy. During the 15 subcommittee meeting, I know I emphasized a lot about 16 there's a lot of real world examples that weren't 17 covered by the sources that you indicated even on this 18 slide earlier in your presentation. And that was 19 similar to what I saw in the report, as I recall. And 20 after the meeting and maybe even during the meeting, 21 Amy Cubbage mentioned, well, we actually have started 22 to ask the laboratories to do some research to give us 23 some real world examples.

24 And after the meeting, we were sent some 25 reports. I note the level of depth varied NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

180 1 considerably for the different reactor types. But I 2 just wanted to acknowledge that that's the kind of 3 detail I was looking for because the DOE report on 4 your slide was very outdated and old and had limited 5 information. And so I was glad to see that you're 6 starting to do that because -- and you might want to 7 think about updating this report before it's issued to 8 acknowledge those references and that effort that's 9 ongoing with the advanced reactor folks.

10 MR. ALGAMA: Yes, ma'am. Thank you for 11 the pointer. I appreciate that.

12 MEMBER BLEY: Don, I don't know if this is 13 for you or for Kim or someone else. In the 14 conclusions of your report, you list the ten reports 15 that are expected in the future. And they were 16 numbered that way, one way in the report and another 17 way on the slide to the last time I saw them.

18 MR. ALGAMA: Yes.

19 MEMBER BLEY: But that isn't my point.

20 MR. ALGAMA: Oh, okay.

21 MEMBER BLEY: In the report, I think at 22 the time of the subcommittee, we were told the 23 enrichment -- and you have swift handling up to 20 24 percent was available now. The TRISO fuel form of 25 fabrication was available as well as the table TRISO NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

181 1 fabrication. Do you have a schedule for the other 2 seven reports?

3 MR. ALGAMA: Right now, apologies if I 4 misspoke at the subcommittee meeting. Those -- we 5 don't have those reports available yet. We haven't 6 done the work. There are some complementary work that 7 we can leverage, for example with the UO6 8 transportation packages we're looking at for the LWR 9 ACS HLU work. That -- we have a report available for 10 that. But that only covers part of what we anticipate 11 we'll need to cover non-LWR activities.

12 MEMBER BLEY: Okay. So it'd be my 13 misunderstanding.

14 MR. ALGAMA: I apologize if I misspoke.

15 MEMBER BLEY: Do you have a schedule for 16 them, and are you planning to bring them to our 17 Committee?

18 MR. ALGAMA: We can. Yes, sir. We don't 19 have a schedule yet, but we can bring it in once we 20 do.

21 MEMBER BLEY: Okay.

22 MR. ALGAMA: The first task is to try to 23 figure out what Joy was saying, bring all the 24 available knowledge together to understand what we 25 should look at and how to develop the scenarios to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

182 1 apply the codes.

2 (Simultaneous speaking.)

3 MEMBER BLEY: I'm sorry. Go ahead.

4 MR. ALGAMA: No, no. I apologize. Go 5 ahead.

6 MEMBER BLEY: Have you thought about the 7 priority of which ones you want to get done first and 8 why?

9 MR. ALGAMA: Yes, sir. We're using Volume 10 3's approach is which is coordinate with NRR to 11 dictate which designs we would pick first. So the 12 first one will be the HPR, the HTGR, and the FHR.

13 Those are the first three we're going to look at.

14 MEMBER BLEY: Okay, great. Thank you.

15 And I think you would find a lot of interest on the 16 Committee to get a look at that.

17 MR. ALGAMA: Yes, sir.

18 MEMBER BLEY: Any other members have a 19 question?

20 MS. WEBBER: Can I just comment on that?

21 MEMBER BLEY: Oh, please.

22 MS. WEBBER: That's a good question, 23 Dennis. I mean, so what Don and the staff and Drew 24 and the staff have really done is outlined an 25 approach. And so maybe -- this is a suggestion.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

183 1 Maybe in terms of writing the letter, you write a 2 letter based on the overall approach and some other 3 things that you noted, Joy, in terms of references and 4 so forth. And then maybe we finish that report, make 5 it a Rev 1 or whatever like we did with the other ones 6 and then come back to talk about some of the details 7 because I think there is quite a lot of important 8 technical detail that would be included in those 9 subsequent ten reports. So I think we need to just 10 think about it.

11 MEMBER BLEY: I got a little confused from 12 what Don told me. And looking at this last slide, it 13 does say that in this numbering system, 6, 7, and 10 14 are already available. And that's true?

15 MR. ALGAMA: No, it's not.

16 MEMBER BLEY: That's not true either?

17 Okay.

18 MR. ALGAMA: No, I'm afraid not. We 19 haven't actually started the work. So none of these 20 --

21 (Simultaneous speaking.)

22 MEMBER BLEY: That's what I thought you 23 said. Okay.

24 MR. ALGAMA: Yes, sir. Yeah, I may not be 25 helping with this in a --

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

184 1 MEMBER PETTI: I doesn't mean you have 2 enough information that you could start. That might 3 be what you meant to say.

4 MR. ALGAMA: Yes, we believe we have 5 enough information to get something going to provide 6 some kind of scenarios to assess our codes against.

7 We believe there's enough knowledge out there.,

8 MEMBER PETTI: So in terms of the first 9 five volumes, just in general the priority is going to 10 be what month are we asking because things are fairly 11 volatile. But DOE issued some big money to a pebble 12 bed design and a sodium fast reactor design. So 13 actually number 3 there might move up in priority 14 whereas the fuel cycle analysis for the FHR may not be 15 all that different than a pebble bed in terms of the 16 fuel cycle per se. The reactor analysis will be 17 different. So there's some things that could flip 18 this stuff around.

19 MS. WEBBER: And to your point, Dave, 20 thanks for raising the question of priority. I mean, 21 all along since we've been developing these volumes, 22 there's been so many changes with the non-light water 23 reactor community. And notably, DOE's funding of 24 these two advanced reactor demonstration projects and 25 at a substantial level, really does help with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

185 1 prioritization of activities.

2 And so as best we can, we do try to pivot, 3 if you will, towards the priority du jour. And it 4 does seem that, again, with last year's awards -- DOE 5 awards, there seems to be some substantial interest in 6 those particular reactor types. So it's a good 7 comment.

8 MEMBER KIRCHNER: Don, this is Walt 9 Kirchner. Following the kind of fuel cycle from 10 mining to disposal approach, of your codes, if we put 11 aside for the moment the utilization aspect, pretty 12 much I would think that your workhorse tools are 13 fairly flexible and available to analyze certainly all 14 those parts of the fuel cycle with the exception of 15 the utilization. Is that a good summary? I mean, 16 I'll pick on something. I'm assuming --

17 MR. ALGAMA: Yes.

18 MEMBER KIRCHNER: -- you're going to use 19 the Monte Carlo code for MCNP for criticality --

20 (Simultaneous speaking.)

21 MR. ALGAMA: It'll be SCALE. So this will 22 be --

23 MEMBER KIRCHNER: Or SCALE? Okay.

24 MR. ALGAMA: Yes, sir.

25 MEMBER KIRCHNER: I misspoke. But pretty NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

186 1 much aren't the tools -- I think the tools as I know 2 them, and my experience now is quite dated, have the 3 flexibility to address pretty much on both sides of 4 the utilization.

5 MR. ALGAMA: Yes, sir. And we're using it 6 in some parts of even U2 power production phase to 7 provide in Volume 3, so where it makes sense for 8 containment analyses or --

9 (Simultaneous speaking.)

10 MEMBER KIRCHNER: I'm thinking transport 11 -- the codes available for structural analysis and 12 thermal analysis of transport, for example, probably 13 are flexible enough to accommodate these different 14 designs. Is that a fair statement?

15 MR. ALGAMA: I believe so, sir. That's 16 our starting-off position is that we have the 17 capability.

18 MEMBER KIRCHNER: Yeah, okay. So then the 19 real rub comes into the utilization area with some of 20 these advanced concepts.

21 (Simultaneous speaking.)

22 MEMBER BLEY: Any other members of the 23 Committee have anything you'd like to raise?

24 (No response.)

25 MEMBER BLEY: Thomas, can we get the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

187 1 public line open? We'll be opening the public line.

2 Any member of the public who would like to make a 3 comment in a moment, we'll ask you to state your name, 4 affiliation, and give us your comment for our 5 consideration. Thomas, are we open yet?

6 MR. DASHIELL: Lisa, coming up live.

7 Going live. Public line is open --

8 MEMBER BLEY: Thank you, Thomas.

9 MR. DASHIELL: -- for public comments.

10 MEMBER BLEY: Anyone out there on the 11 public line who would like to make a comment, please 12 state your name and give us your comment.

13 (No response.)

14 MEMBER BLEY: I guess we have no one.

15 Thomas, we can close the public line. And Mr.

16 Chairman, I can turn it back to you.

17 CHAIR SUNSERI: Thank you, Dennis, and 18 thank you, staff, for that presentation. It's 3:22 19 right now. Our next presentation is scheduled to 20 start at 3:30. It's an informational briefing on 21 Post-Halden plans.

22 Since that is an informational briefing, 23 what I'm going to do is I'm going to call for an 24 eight-minute break here. We will resume at 3:30 25 because we need to respect the research group's time NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

188 1 on this. They're expecting to start, and we should 2 abide that.

3 However, since this is an informational 4 briefing, members will be -- I'll ask that members can 5 come and go as you determine your needs are being met.

6 So we will take a recess till 3:30. We'll reconvene 7 at 3:30, and members are free at their will to take 8 their break however they want. Thank you.

9 (Whereupon, the above-entitled matter went 10 off the record at 3:23 p.m. and resumed at 3:33 p.m.)

11 CHAIR SUNSERI: We will proceed with the 12 meeting. This is a session on post-Halden plans, 13 informational briefing by RES. I'll turn it over to 14 Subcommittee Chair Joy Rempe, who's also Vice Chair of 15 the full Committee.

16 VICE CHAIR REMPE: Thank you, Mr.

17 Chairman. Colleagues, this briefing pertains to our 18 efforts to address the 1997 SRM to the Commission to 19 examine the need, scope, and balance of the Reactor 20 Safety Research Program and follow the Commission's 21 directive that ACRS consider how the Office of Nuclear 22 Regulatory Research, or RES, anticipates research 23 needs and positions for the changing environment.

24 In our 2020 review and evaluation of the 25 NRC Safety Research Program, we observed there were NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

189 1 several research topics of special interest for 2 additional briefings. In subsequent interactions with 3 Ray Furstenau, the Director of RES, he indicated that 4 the Committee's review on such topics would be of more 5 value to RES than the quality reviews of selected 6 projects that we previously performed for RES.

7 In September 2020, we were briefed about 8 the future-focused research project efforts by RES, 9 and today we're going to hear about a second topic of 10 interest, RES activities to address the gap associated 11 with the closure of the Halden boiling water reactor 12 that was used to obtain data to assess the irradiation 13 performance of fuels and material.

14 So at this point I'd like to ask Ray 15 Furstenau, the Director of RES, to begin today's 16 meeting.

17 MR. FURSTENAU: Thank you, Vice Chair and 18 Chairman, for having us come in and provide this 19 information briefing. It really is my pleasure to 20 introduce this part of your agenda today.

21 I know it's been a long day, but I think 22 you'll find this pretty interesting. I know it is for 23 me, and I do appreciate Matt, both you and Joy in 24 working with us to come up with topic areas. I really 25 think this will be very beneficial to the NRC's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

190 1 research program to provide these briefings and get 2 feedback from you. So thank you for that. Next 3 slide, please.

4 Here's the agenda. I'll provide some 5 brief opening remarks, but I really want to have time 6 for, you know, some really top quality researchers 7 that we have on our staff to talk about the fuels 8 activities and strategies and then the material, 9 structural material strategy. Looking ahead, a vision 10 of the future, our Engineering Director, Louse Lund, 11 will talk about that, and then I'll have some brief 12 closing remarks. Next slide, please.

13 And Dr. Rempe already talked about our 14 motivation for doing these information briefings, so 15 I won't spend a lot of time on that. This post-Halden 16 planned discussion will only be for the fuels and 17 materials research. There was a separate activity as 18 part of the Halden project called the man-technology 19 organization interface, and that's still going on and 20 we are participating in some of those activities.

21 Next slide, please.

22 This is kind of a hard one to see, but 23 this just kind of shows you the timeline of recent 24 Halden events and the shutdown, the decision announced 25 permanent closure of Halden and some of the early NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

191 1 activities that have went on since that occurred.

2 And I'd like to point out we've been 3 participating in the INL workshops and the NEA 4 activities related to post-Halden activities as well.

5 And we'll be getting into much more detail on these in 6 some subsequent presentations. The next slide, 7 please.

8 And you'll hear also, there's been a lot 9 that's been happening. You know, we've -- a lot, you 10 know, Halden was a nice total package, and I think in 11 some respects we were fortunate to have it as long as 12 we did. It was a very good deal for the NRC and the 13 rest of the international community, but once it shut 14 down we -- you have to move on. And we have to fill 15 those gaps the best we can.

16 And I'm very optimistic that we'll be able 17 to do that. Halden did a lot of good things, it was 18 a bargain to us. But we -- we're going to find ways 19 to fill those gaps. And you'll hear much about that 20 today again.

21 And our NRC staff have been in a 22 leadership role in many of the initiatives, both 23 domestic and international. We've been kind of 24 spearheading the effort, along with DOE, on the FIDES 25 framework, that's the framework for irradiation NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

192 1 experiments, which is kind of the life after Halden on 2 the international end for the NEA.

3 And we've been working on that framework.

4 That agreement is almost in place. I think probably 5 within a month we'll be ready to go and there'll be 6 several JEEP activities that we'll be able to get 7 started. And those'll be talked about later as well.

8 So these are the types of things we'll 9 talk about. Archiving of the Halden data, which we 10 think's important. The new research projects we're 11 going to be doing, both with participants domestically 12 and internationally. Retrieving of some of the Halden 13 materials so it doesn't get disposed of as they're 14 decommissioning that facility. And the new domestic 15 capability -- new experimental capabilities.

16 And we do have some -- we're going to be 17 talking about some of the DOE programs. And so if 18 there are any questions regarding the DOE 19 partnerships, we do have some people that can help 20 answer questions on that. So next slide, please.

21 And the benefits and the challenges, I 22 think you're aware of those. The benefit of having a 23 diverse network of research facilities can be positive 24 so we're not vulnerable to facility closure like what 25 happened with -- single facility closures like what NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

193 1 happened with Halden.

2 And I think the investment in US 3 facilities, DOE's really stepping up in investments in 4 their research reactors, hot cells. The NRIC program, 5 which is the Nuclear Reactor and Innovation Center 6 program that kicked off in the last couple years.

7 They're building capabilities, like a 8 molten salt thermophysical examination capability in 9 Idaho, the LODIS facility and DOME (phonetic) facility 10 in Idaho to really be able to do demo activities that 11 we can take advantage of the work that's being done 12 there as well.

13 So, but the challenge obviously is these 14 multiple research activities will probably be a higher 15 cost than the Halden project was. But we have been 16 working with DOE. DOE's picking up more of the cost 17 of the FIDES framework and the projects coming out of 18 FIDES. We are able to join as participating members of 19 JEEP projects when they're domestically located in the 20 US at no cost to the NRC.

21 So we're trying to look at a lot of 22 different avenues of how we can best utilize the 23 funding we have and participate in research activities 24 for fuels and materials. Next slide, please.

25 VICE CHAIR REMPE: Could I stop you for a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

194 1 second?

2 MR. FURSTENAU: Sure.

3 VICE CHAIR REMPE: The situation about the 4 Halden gap was brought up during our December meeting 5 with the Commission. And after we discussed it, one 6 commissioner said as you -- eventually we were going 7 to have this briefing and we were going to be 8 evaluating it. And they mentioned that they were 9 concerned about having diverse test facilities, that 10 it might cause some differences in the data that might 11 be more difficult to explain.

12 And it's my opinion, and I didn't have the 13 opportunity the way the comment was offered, that if 14 you have adequate instrumentation, you can overcome 15 such difficulties. And so your point about having 16 multiple facilities might be good because we won't be 17 so dependent on one facility.

18 But on the other hand, this challenge 19 you've brought up about the cost of multiple research 20 initiatives is going to be most likely higher. And 21 one of the nice things about Halden was they had 22 standardized test rigs.

23 They knew what they were doing by always 24 having the same kind of test rig go in for a 25 particular type of test, so they didn't have to expend NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

195 1 resources for designing a test rig, which I know from 2 your background at ATR you understand that that is 3 quite expensive. But we don't worry about cost, we 4 worry about safety --

5 MR. FURSTENAU: Yes.

6 VICE CHAIR REMPE: If resources are still 7 an issue for ACRS, because it prohibited us from 8 getting -- from advising how the staff can get what 9 they need to address safety issues.

10 So one, I'm interested in your thoughts 11 about the instrumentation overcoming differences in 12 facilities. Then two, as you interact with these 13 various facilities, is there a push to say hey, let's 14 try and have some standardized tests and everybody 15 realized that this is what we're going to do? Because 16 that was a nice benefit of Halden.

17 MR. FURSTENAU: Yeah, okay, Joy, I'll 18 comment on that and I'll try to cover. If I miss 19 something, please let me know. As far as standardized 20 tests, I think, at least my understanding of like 21 what's being done with ATR, for example, if they took 22 a I-Position loop and some of the treat testing that's 23 -- has the capabilities that are being developed and 24 standardized tests at Holgers (phonetic) and 25 instrumentation allows for more economical testing, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

196 1 and maybe quicker testing as well.

2 And the instrumentation challenge, I, you 3 know, I think the national labs get -- trying to learn 4 the benefits, the lessons learned from like Halden and 5 from international test reactors on how do you 6 instrument these things, whether it's material or 7 fuel. And how do you know what you're measuring. And 8 that's always important in any of the tests.

9 And then how do you somehow make sure you 10 get consistent measurements like using, I don't know, 11 using the same type of thermocouples. Where do you 12 put them. And same way with pressure caps and things 13 like that.

14 So I think those are things we have to 15 stay aware of in the integrity of the data and how we 16 pedigree that data is very important. And, but I 17 think that happens whether you're at one facility or 18 multiple facilities, Joy. I think it's just harder 19 when you do it at multiple facilities. But I think it 20 can be done.

21 VICE CHAIR REMPE: I agree with you. You 22 need to be cognizant of how the facilities are run, 23 because Halden is, for example, a type C thermocouple, 24 which might be difficult to -- which transmutes, by 25 the way, due to the radiation and you can't quite use NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

197 1 that at ATR because, you know, at Halden, they always 2 brought it up to the same power level.

3 MR. FURSTENAU: Yeah.

4 (Simultaneous speaking.)

5 VICE CHAIR REMPE: -- controlled runs that 6 you move because of other tests. And so you're going 7 to have to be cognizant of some of the specifics of 8 the facility. But I do think it's a challenge that 9 can be addressed.

10 MR. FURSTENAU: Yeah, it certainly is. I 11 would agree with you, Joy, yeah. Anything else?

12 VICE CHAIR REMPE: Nope, thank you very 13 much.

14 MR. FURSTENAU: Okay, next slide, please.

15 And I won't spend a lot of time on this because we've 16 really already talked about it, about these things.

17 I did want to point our university, integrative 18 university program offers another mechanism where we 19 can maybe get some more research done by the 20 universities, especially in the more future-focused 21 type research.

22 Some of you may or may not be aware, in 23 Fiscal Year '20 funding we receive from Congress for 24 the Integrative University Program, we normally fund 25 fellowships, scholarships, and faculty development NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

198 1 grants. But in FY '20 we broadened the scope of that 2 a bit and we created two funding opportunity 3 announcements, one for the traditional faculty 4 development scholarships and fellowships.

5 But we did a separate FOA that really 6 asked the universities in particular areas to propose 7 multi-year research projects up to three years and up 8 to $500,000 per project. And the response was really 9 overwhelming. We have, oh, in the neighborhood of 10 140, more than 140 qualified applications. And we 11 chose 15 out of that, and maybe we can brief you on 12 those sometime.

13 But it was really -- and so we really want 14 to take advantage of the university program for good 15 ideas on how to help with our emerging research needs.

16 So it's really a kind of exciting time in research, I 17 think.

18 So with that, I would like to turn it over 19 to Michelle Bales and Lucas Kyriazidis and James 20 Corson to talk about our activities in the fuels area.

21 So Michelle, you've got it. Oh -- I'm 22 sorry.

23 VICE CHAIR REMPE: I have a question.

24 Just at a high level, when I think about gap, are 25 there any ongoing activities for the staff where NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

199 1 because right now you've got limited locations to 2 conduct irradiations where something might be at risk.

3 Is it like the Accident-Tolerant Fuel 4 Program if someone came in, or is it the high assay 5 enriched uranium, is there -- do you think you've got 6 a couple of years and adequate time to get data that 7 you need for regulatory decisionmaking? Maybe it's, 8 you know, subsequent license renewal. But what's the 9 one with the least margin that might be a close 10 concern?

11 MR. FURSTENAU: Well, I think I'd ask 12 maybe the subsequent presenters here. And if they 13 don't answer the question, Joy, we'll answer it in the 14 end.

15 I think some of the programs that are 16 going to be subsequently talked about, they'll -- like 17 accident-tolerant fuel, for example, I think Michelle, 18 I'll put you on the spot that you can kind of talk 19 what you see are maybe some of the experimental gaps.

20 And then Lucas, James and then subsequently Matt, if 21 you could kind of express where you see the nearest 22 term.

23 Because I think we're, in my opinion, I 24 think we've got a good path forward, Joy. I don't see 25 anything where we're -- we just can't proceed forward NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

200 1 without it. But I really like our researchers that 2 are going to present subsequently that give their 3 opinion on it too.

4 VICE CHAIR REMPE: That sounds great.

5 Thank you, I just would like to have that emphasis, is 6 there anything with the least margin.

7 MR. FURSTENAU: Yes, good question, Joy.

8 So we'll try to cover that with subsequent talks here, 9 so. Anything else?

10 VICE CHAIR REMPE: I'm good. Colleagues 11 or consultants, do you have any questions.

12 CHAIR SUNSERI: I don't have any questions 13 or comment. I will just, I note that there are some 14 comments in the chat line that are kind of supportive 15 of the meeting. So I just request that people do not 16 use the chat line for meeting content. It's outside 17 the scope of the transcript and the public disclosure.

18 So only use the chat for coordination or Teams 19 communication problems. Thank you.

20 MR. FURSTENAU: Michelle.

21 MS. BALES: Okay, thank you. So Joy, just 22 to answer your question since I don't want to forget 23 to address specifically with ATF. I think what we're 24 seeing is that there's a lot of restrategizing how to 25 get the same data that might have been obtained at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

201 1 Halden through other means.

2 And so I don't see a bottleneck of data 3 that was going to be coming from Halden just not being 4 able to be obtained through any other means. I think 5 we're going to see a lot more reliance on information 6 from LTA, lead test assembly examinations.

7 And while that might not be as detailed or 8 as independent as Halden might have been, I think for 9 the near term decisions that we're going to be faced 10 with at NRC, we're going to have the information that 11 we need. So --

12 VICE CHAIR REMPE: For something like 13 thermal conductivity degradation also?

14 MS. BALES: Yes, I mean, I think that 15 there's going to be technical issues where, when the 16 staff is presented with an application, they are going 17 to, you know, be used to seeing certain things that 18 might have been at Halden and they might ask about 19 those. But so thermal connectivity might be an 20 example of where an online measurement really made a 21 big difference, rather than a end-of-life post-22 irradiation examination kind of situation.

23 And I think eventually those kind of 24 capabilities, one of -- as I go through the 25 presentation and I share some of the work that DOE has NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

202 1 done, I think we'll see that that type of capability 2 to make measurements of thermal conductivity is going 3 to be brought online in the near future. And I think 4 it's just, you know, how the vendors choose to deal 5 with the near-term applications.

6 VICE CHAIR REMPE: These things don't 7 happen overnight, as you know. And so I'm pushing, 8 trying to push the envelope that are we thinking and 9 making sure everybody knows that, again, you're just 10 the regulator, but if they come in and they say, okay, 11 we'd like to do it. And maybe it's not ATF, maybe 12 it's something else.

13 But I'm just kind of bringing it up 14 because that one comes to mind. But is there some 15 place where we're conveying to them that you better 16 have data and, you know, for regulatory decisionmaking 17 we're going to need that data or you won't be able to 18 do something.

19 And that's where I'm going to be pushing 20 during this conversation to make sure that we've 21 identified not just the gaps for today, but the gaps 22 for a couple years, three years from now, or whatever 23 it is, to get the data we need. Okay?

24 MS. BALES: Yeah, no, that's a great 25 point. And I think, like I said, specifically with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

203 1 ATF we have -- the staff is really working with 2 vendors in real time and ahead of their submittals to 3 express our data needs and to talk about what kind of 4 expectations we have. That's part of the ATF project 5 plan where we talked about this new paradigm where we 6 need to get data early and often.

7 And I think it's working pretty well, but 8 you know, that's -- the applications aren't in yet.

9 So I think that there'll be opportunities to prove 10 that that really works.

11 VICE CHAIR REMPE: Okay, thank you.

12 DR. CORRADINI: This is Corradini, can I 13 ask Michelle just one last way of asking the question.

14 To the extent that you're always in communication with 15 the DOE and the potential applicants, I assume that 16 you early on point out things that you're going to 17 need so that they figure it into the budgetary 18 considerations. Because in terms of expenditures of 19 funds, they're the largest expenditures of funds with 20 some of this fuels and materials testing.

21 MS. BALES: Yeah, you know, I mean, I'm 22 going to try to speak to this at a high level. But 23 really the questions that you're both getting at are 24 complicated because when we talk about what data is 25 needed for licensing, and I'm going to very high level NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

204 1 reiterate some conversations that are happening 2 amongst the staff and the Licensing Office in 3 Research, so much of the data needs are defined by the 4 licensing strategy.

5 What the vendors are going to seek to take 6 credit for, or how they're going to impose limitations 7 and conditions when they don't have certain types of 8 data. How they're going to deal with uncertainty.

9 And we just don't know those full licensing strategies 10 yet.

11 And so I think in the -- in NRR, they are 12 thinking about data needs in a very particular way.

13 And it's a little bit different than how Research, 14 when we're trying to develop our fuel performance code 15 and want to get a best estimate solution for fuel 16 performance behavior. You know, their thoughts are 17 going to be a little different than ours.

18 And so I think in the context of research 19 and in the context of Halden, we have a certain way of 20 thinking about the data that was produced and how we 21 used it. This is a gross oversimplification, but NRR 22 is just thinking about it differently. And I think it 23 largely is because the licensing strategy has so many 24 options for dealing with varying amounts of data at 25 different times.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

205 1 VICE CHAIR REMPE: But even earlier this 2 week we heard about the accelerated whatever licensing 3 strategy. But it finally comes down to got to have an 4 integral test.

5 MS. BALES: Yeah. Yeah, and I'm excited 6 to tell you some of the slides that I'm going to have 7 here because I think we really -- there's really 8 exciting developments going on to ensure that we still 9 have those integral test capabilities.

10 So maybe with that I'll get started, and 11 then as there's more questions about this I can 12 certainly -- we can certainly return to it if we're 13 not giving you the right information.

14 MR. FURSTENAU: Joy, could I jump in with 15 something real quick?

16 (Simultaneous speaking.)

17 VICE CHAIR REMPE: Go ahead, Ray.

18 MR. FURSTENAU: Yeah, regarding integral 19 test, I think is very important. But I think what 20 we're trying to do with the collaborations and stuff, 21 as we all know, integral tests are expensive. But you 22 know, we've come a long ways with modeling and 23 simulation capabilities. And so we can really maybe 24 zero in better on what those integral tests ought to 25 look like so we don't have to maybe do as many, you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

206 1 know.

2 And I think getting some data to help 3 validate our codes in areas that they haven't been 4 used before is I think some of the data that we'd be 5 looking for as well. So I think targeted integral 6 tests is what we really want to work with DOE on as 7 well.

8 VICE CHAIR REMPE: Okay, thank you. And 9 I apologize for dragging this out so much with our 10 questions, but I do want to kind of keep that focus on 11 not just where you're at but where we're going. So go 12 ahead. I'll try to contain myself. Michelle.

13 MS. BALES: No, it's great, I appreciate 14 the questions. So I'm going to start my presentation, 15 which is going to focus on the fuels area specifically 16 by reiterating what Ray said, that in the two years 17 since Halden announced that they were officially 18 shutting down, a lot has happened.

19 And so in the slides that you'll hear from 20 myself and from my colleagues, we'll speak 21 specifically to the fuels research, and you'll hear 22 after us what some of the developments in the 23 materials world.

24 And I'm going to start my presentation 25 with some significant investments that have been made NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

207 1 by DOE to address the gaps left by the Halden reactor 2 shutdown. I'm going to then turn it over to my 3 colleague James Corson, who will speak to NEA's 4 progress in launching the FIDES framework, which is 5 designed not only to connect a network of experimental 6 reactors, but also reestablish the valuable technical 7 community that was formed under the Halden reactor 8 joint project.

9 And you know, listening to your questions, 10 Joy, I think this is an important point, because the 11 Halden reactor project wasn't just a way for multiple 12 countries to leverage their resources dollars-wise.

13 But also, the Halden reactor project formed a peer 14 review body with a really immense knowledge base that 15 would scrutinize the Halden results and ask tough 16 questions as data was coming out. They weren't just 17 customers, in other words, they were really part of 18 the fabric of the program.

19 And so I think with the FIDES program what 20 we're going to see is that reestablishing that 21 technical community I think might get to some of, you 22 know, what you were pointing to about ensuring that 23 this diverse network has the same pedigree and that 24 maybe Halden would have had. But there's a lot of 25 smart people who will be at the meetings and poke NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

208 1 holes when they see something that doesn't seem right 2 or have the knowledge to point out some of the 3 different ways that you can mess up an experiment.

4 So I think it's important to just point 5 that out, that it's not just the leveraging of dollars 6 but the leveraging of expertise.

7 And then finally my colleague Lucas 8 Kyriazidis will explain Studsvik nuclear proposal to 9 rescue unique and valuable irradiated material samples 10 from Halden so that they can continue to provide value 11 to the nuclear community in the future.

12 And I'll say this throughout, but I just 13 want to say here at the onset that in each of these 14 areas, NRC staff and management have played a pretty 15 active role to influence the initiative. So even 16 though they're not directed by us, you know, we've 17 been very engaged.

18 So as I said, I'm going to start with 19 information on DOE's investments to address the 20 capability gap left by Halden. But I want to explain 21 that DOE didn't do this in a vacuum. DOE brought 22 together experts from around the world to inform their 23 strategy.

24 They brought other research reactors to 25 the table, from MITR, BR2. They had a number of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

209 1 regulatory bodies giving feedback through their 2 workshops over the summer following the announcement.

3 Industry representatives including fuel vendors and 4 utilities were there to help talking about what kind 5 of capabilities they needed to support the design of 6 new fuel materials.

7 And NRC was very engaged in these 8 workshops that were held. You know, on Ray's slide 9 you could see the timeframe. Sometimes months after 10 significant news on Halden, DOE was having a workshop 11 to collect the opinions of these key stakeholders.

12 So the report picture on the left is the 13 culmination of a month-long conversation that DOE led 14 with key stakeholders. And it was published just six 15 months after the reactor officially announced its 16 shutdown.

17 And the report declares that the primary 18 recommendations for addressing the capability gaps 19 left by Halden and that were really necessary to 20 sustain the US nuclear fleet were, first, to transfer 21 unique technologies and knowledge for testing, 22 refabrication and instrumentation from Halden to 23 relevant facilities.

24 I think, Joy, this gets also to your point 25 about Halden's pedigree. And I think there was a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

210 1 great recognition that what they had developed 2 shouldn't go unused, that we should really try to 3 capture that in transferring knowledge and 4 technologies to other facilities.

5 Another conclusion in the report is the 6 need to expand LWR irradiation capacity in test 7 reactors, both steady state and transient testing 8 capabilities. A third conclusion was the need to 9 establish fuel rod refabrication and re-10 instrumentation capabilities at DOE facilities. This 11 is something that Halden was particularly strong on.

12 And so ensuring that that can be replicated is a key 13 recommendation.

14 Complementing in-pile testing capabilities 15 with reliable in-pile instrumentation, similar to what 16 was available at Halden was another finding. And then 17 finally to establish a domestic center of excellence 18 consolidating irradiated -- irradiation testing 19 activities in a way that reduces schedule and shipping 20 costs but also simplifies data acquisition to ensure 21 that no facility becomes a single point of failure.

22 Sorry, I merged two points there. That simplifies 23 data qualification.

24 And I think, again, this gets to your 25 point, Joy, that when you have a really diverse NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

211 1 network and you have to qualify data at a variety of 2 places, that that sort of can be a downside. So 3 spreading out the testing capabilities, but not so 4 much that you've diffused efforts that are necessary 5 to qualify data as important.

6 DR. CORRADINI: Can I ask a clarification?

7 You said blue is what was expended in those years and 8 yellow is what's planned?

9 MS. BALES: Yes, so on the slide --

10 DR. CORRADINI: So in FY '21, has it come 11 to pass what they need versus what they got?

12 MS. BALES: So I created this slide, and 13 in the weeks that have passed since the slide was 14 created there may be news. But I would have to ask 15 DOE to speak to that specifically.

16 DR. CORRADINI: Okay, all right -- sorry.

17 MR. McCAUGHEY: Would you like this --

18 would you like me to address that? This is Bill 19 McCaughey with the Office of Nuclear Energy.

20 DR. CORRADINI: That's up to the Chairman.

21 I'll let Dr. Rempe decide that.

22 VICE CHAIR REMPE: Sure. And as you're 23 addressing it, I'd like to add on a question: what 24 does that get me if we get out to 2023? Do I have the 25 I-Positions? Do I have something at MIT as well as NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

212 1 the I-Positions? What do I have by 2023 if I get all 2 this funding that's shown here?

3 MR. McCAUGHEY: Okay, so I'm Bill 4 McCaughey, I'm the Director of the Advanced Fuels 5 Technologies Office at the Office of Nuclear Energy.

6 The yellow are, that's estimates still.

7 For 2021, yes, we did get the appropriations. We also 8 got an, also another bill, an authorization bill. We 9 are still working out the funding that flows from 10 that, and that's why it's yellow. So the, starting 11 with 2021, that's still an estimate, and we're working 12 out the details on how we're going to allocate our 13 funds.

14 DR. CORRADINI: Okay, thank you.

15 MR. McCAUGHEY: Dr. Rempe, for your 16 question, what this is going to get is pretty much 17 along the lines of what Michelle just went through in 18 the -- what was needed. So it's going to get -- we're 19 working on a new flowing water loop in the advanced 20 test reactor in one of the I-Positions.

21 We're also working on refabrication 22 capabilities in the hot cells at Idaho National Labs 23 so you can take irradiated fuel from reactors 24 elsewhere, bring them into Idaho Lab and refabricate 25 them to fit the test vehicles that you want and also NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

213 1 to instrument them for insertion into either the 2 transient reactor test facility or the advanced test 3 reactor.

4 And then you're also getting the LOCA test 5 capabilities at the transient reactor test facility.

6 And also advanced instrumentation along the lines of 7 what Halden was capable of.

8 VICE CHAIR REMPE: So by 2023 I'll be able 9 to put a fuel rod in a -- that's been previously 10 irradiated, and put it into the standardized test grid 11 and get thermal conductivity degradation, get crud 12 deposition? What all can I get?

13 MR. McCAUGHEY: Well, now you're getting 14 into some details that I don't want to go out on a 15 limb. But we do have Dan Wachs from Idaho Lab who 16 might be able to answer that, some of those specifics.

17 MS. BALES: Well, can I say just the next 18 couple of slides actually talk about the plans. And 19 so maybe after I go through that, Dan, you can then 20 sort of just say, okay, what Michelle just said will 21 be ready or won't be ready by 2023. Because --

22 VICE CHAIR REMPE: And it's more money 23 that's needed too. Because this, just knowing how 24 much things cost I just am curious. Because again, 25 there's a lot of -- I want to, yeah, understand how NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

214 1 soon.

2 PARTICIPANT: Yeah, that sounds find, 3 Michelle, that will work well.

4 MS. BALES: Okay, because I'll -- I've 5 worked with DOE to extract a lot of the planning here, 6 so I have slides that speak to it a little bit. And 7 then in terms of timing and budget, I think that'll be 8 more meaningful once you see what kind of plans there 9 have been -- what plans have been made.

10 So I mentioned on the previous slide that 11 one of the main recommendations from the report is to 12 expand LWR irradiation capacity. So to fill on the 13 expansion of steady state capabilities, DOE is 14 designing I-Loops, which, you know, have just been 15 mentioned in the ATR and the INL. And the I-Loops are 16 designed to provide additional BWR and PWR water loop 17 testing capability at prototypic linear heat 18 generation rates.

19 The in-pile tubes will allow for highly 20 instrumented testing. And further, the I-Loops will 21 support power ramp testing using helium-3 screens to 22 allow independent control of rod power and full 23 hydraulic control to allow for things like dry-out 24 testing.

25 And power ramp test data is really NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

215 1 critical to address many reactor safety questions.

2 And these testing capabilities have not only been 3 impacted by the Halden reactor closure, but prior to 4 the Halden reactor closure, the closure of the OSIRIS 5 reactor in France and the R2 reactor in Sweden left 6 big holes in the ramp test capabilities. So the 7 establishment of ramp capabilities at INL truly is 8 fulfilling a major capability gap.

9 So on the slide it says that this testing 10 capability is expected to begin operation by 2023.

11 I'll at the end turn it over to Dan to say whether 12 that is funded and -- or not, with the graph that I 13 showed on the previous slide. But you know, in 14 speaking to the plans, I think it's a really exciting 15 capability development.

16 In addition, INL is developing LOCA 17 testing capabilities in treat. So there's been a lot 18 of hot cell LOCA testing capabilities developed around 19 the world in the last decade. But the closure of 20 Halden eliminated a heavily used in-pile LOCA testing 21 capability. The transient water irradiation system 22 for TREAT, or TWIST, provides a truly unique testing 23 capability.

24 Cask fill and rodlet pressure can be 25 measured by the same LVDT Bells (phonetic) approach NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

216 1 that was employed by Halden. Water, fuel, and 2 cladding temperature can be measured with 3 thermocouples and pyrometry. And post-test neutron 4 radiography can reveal details of fuel relocation and 5 conditions after testing, similar to what was done at 6 Halden.

7 But an exciting expansion beyond Halden's 8 capabilities will be available at treat, which is that 9 fuel motion monitoring in real time can be 10 accomplished in the treat facility. And that will 11 allow us a much more sophisticated look at fuel 12 relocation fragmentation relocation and dispersal 13 phenomena.

14 And as this slide indicates, commissioning 15 tests are planned for the TWIST capsule in 2022.

16 Again, I'll leave it to Dan to say at the end whether 17 that is something funded with the allocations 18 requested to date.

19 So two other of the recommendations from 20 DOE's reports were established -- to establish fuel 21 rod refabrication and re-instrumentation capabilities 22 at DOE facilities and to compliment in-pile testing 23 capabilities with in-pile instrumentation, similar to 24 what was available at Halden.

25 So DOE has worked with Halden directly to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

217 1 transfer the world-class technologies that they had 2 developed, transferred those to INL. An 3 instrumentation device for drilling and preparing 4 rodlet ends was procured from Halden, and INL has 5 developed a remote welding system, the so-called in-6 cell weld under pressure systems shown in the upper 7 middle of the page.

8 And these tools will make it possible to 9 allow previously irradiated fuel rods to be repurposed 10 for follow-on R&D with additional irradiation, 11 transient, or other experimental purposes.

12 Replication of Halden's in-pile 13 instrumentation capabilities has also been achieved, 14 thanks to extensive knowledge and technology transfer 15 from Halden, including technologies for measuring 16 plenum pressure, fuel elongation, center line 17 temperatures, and local neutron flux.

18 And there's plans to innovate even beyond 19 what Halden was capable of using optical fibers for 20 measuring in-pile temperatures, pressures, imaging, 21 deformation, etc. So these capabilities are really 22 critical for assessing interval performance data that 23 we have really relied on Halden for in the past so 24 that we can develop models that track these parameters 25 of the function of time and burnup.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

218 1 So I mentioned earlier that there might be 2 in the immediate timeframe some increased reliance on 3 lead test assemblies for fuel performance data. And 4 that's great, that's fully representative, 5 commercially irradiated material is really important.

6 But you only get one data point at the end of life 7 from some of these exams that require destructive 8 examination.

9 And so having in-pile instrumentation 10 capability is really what's needed to follow 11 parameters over the course of irradiation and over the 12 course of life, so that when you go to your modeling 13 and validation exercises that you have more than one 14 data point. You really have a history.

15 So these investments bring world leading 16 capabilities to US facilities, and once complete will 17 represent a huge step towards replacing some of the 18 more unique and relied-upon features of the Halden 19 reactor.

20 So I actually have one more slide before 21 I turn it over to my colleague, but maybe, it's on a 22 separate topic, so maybe this is a good time for Dan 23 to address the question posed by Dr. Rempe earlier 24 about the capabilities that I spoke to on these last 25 two slides and their relationship to the 2023 funding NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

219 1 request so far.

2 VICE CHAIR REMPE: Sounds good to me.

3 MR. WACHS: Thanks, Michelle. Yeah, I 4 think the key point to highlight when you look at the 5 Halden gap report is we reviewed the existing 6 capabilities in the United States already as a part of 7 that -- as a launching point.

8 So you mentioned things like the MIT 9 reactor, you know, work that's going on there. We 10 actually think that's a great place to do some of the 11 corrosion studies and things like that. And we've 12 seen that utilized effectively by a number of the 13 vendors in the United States.

14 We had a pressurized water -- or we have 15 a pressurized water loop in ATR already, a center flux 16 trap that's being used extensively by a number of 17 different sub-parties. And we have separate effect 18 site testing capabilities at both HIFER (phonetic) and 19 ATR is one that we're using.

20 So in that report we're really focused on 21 the places where we were relying on the complementary 22 capabilities available at Halden to fill in our, the 23 spaces that we were missing. So we really -- these 24 are the things that we came up that we needed to have 25 to -- in order to move technology forward in like a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

220 1 general sense.

2 So as Michelle mentioned, so we're adding 3 the pressurized water or the extra LWR loop to the 4 ATR. That could be operated at either a BWR or a PWR 5 mode. We have conceptual design for a second loop so 6 that we could have both operating simultaneously. But 7 the funding to build that infrastructure, 8 infrastructure for both of those is not currently in 9 our near term plan.

10 However, we are running the installation 11 of that new loop in the I-Positions as a capital 12 project. It has a, you know, a project execution plan 13 associated with it, all those kind of thing. It's 14 monitored closely at the DOE level. So assuming the 15 funding allocations, you know, follow what our plan 16 is, we should have that up and running in 2023.

17 Now, that does need to be complemented by 18 an experiments program that's planning to use that, 19 and that plan is incorporated into many of our, the 20 ATF vendors' programs. So we expect that 21 complementary to be there. It is a little bit --

22 VICE CHAIR REMPE: Dan, slow down for just 23 a second. You said that basically the DOE Fiscal Year 24 '21 programmatic or whatever funding doesn't cover the 25 yellow bar that we were shown earlier. It's going to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

221 1 come out of the facilities?

2 MR. WACHS: No, no --

3 VICE CHAIR REMPE: Is that what I heard 4 you say?

5 MR. WACHS: No, the budget line there is 6 the funding allocations from the ASB program direct 7 funding. It's, there's -- and there's a little bit 8 coming in from the advanced sensors and 9 instrumentation program for some of the refab work.

10 So that is the program funding --

11 VICE CHAIR REMPE: So that's still not 12 2021, it's just the programmatic funding. But to get 13 the loops where you could put something in the 14 reactor, unless you fight with existing PWR loops, 15 which is another thing because you've got a lot of 16 people competing for it. To get those two I-Loops, 17 you're going to need to have infrastructure funding.

18 So we don't --

19 MR. WACHS: No, no, I don't think so. The 20 facilities funding is -- so this is the funding to 21 have the up -- the loop up and running and available.

22 The additional funding is simply for a user to provide 23 their sample and do their experiment-specific 24 analysis.

25 VICE CHAIR REMPE: So I'm sorry, I guess NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

222 1 I'm slow here. So the 20 -- the yellow bar for the 2 facility funding that you hope to have --

3 MR. WACHS: It is the advance fuel site --

4 advance fuels campaign allocations to building the 5 experimental infrastructure that would go into the 6 existing infrastructure at the reactors.

7 VICE CHAIR REMPE: So it's the 8 standardized test loop with some instrumentation, 9 what's shown here on the yellow bar. To get the I-10 Loops, you've got to have another pot of money.

11 MR. WACHS: No, no, that is for the I-12 Loops.

13 MR. McCAUGHEY: Can I interject here?

14 This is Bill McCaughey again. So Joy, this is funding 15 the hardware, the design, the installation of the I-16 Loop, as well as the refabrication upgrades and the 17 instrumentation upgrades and the LOCA test 18 capabilities at treat. And it's all getting funded 19 out of the fuels campaign, not the facilities budget 20 but the fuels campaign.

21 VICE CHAIR REMPE: So then I might say, 22 back to the original question, if I got all the yellow 23 bar money, then I would have two I-Loops with 24 instrumented test rigs.

25 (Simultaneous speaking.)

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

223 1 VICE CHAIR REMPE: Is that what I'm 2 hearing?

3 MR. WACHS: No, this is -- we'll have the 4 center flux trap pressurized water loop and one I-Loop 5 on that, this funding.

6 VICE CHAIR REMPE: Okay, that's -- and so, 7 again, you're on track. You got what you wanted, I 8 guess, in 2019 and 2020 to do things? And then you're 9 hoping to have a big plus-up coming here in 2021 and 10 on out. And then you'll be ready to go with one extra 11 I-Loop, the existing PWR loop, and a lot of test rigs 12 and instrumentation.

13 And can I do diameter gauge, can I get 14 crud as a function of time as well as thermal 15 conductivity degradation?

16 MR. WACHS: So those, some of those 17 detailed experiment pieces would go into the 18 experiment-specific design. We'll have a loop with 19 the penetrations that support that branch of 20 experiments, but we're working with the ATF vendors 21 for the design of their specific experiments.

22 VICE CHAIR REMPE: So you'll have the 23 infrastructure but maybe not the instrumentation to 24 get the data.

25 MR. WACHS: Yeah, I think the test train NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

224 1 -- so we broke the test train design off into a 2 separate piece from the experiment platform. And we 3 are designing experiments for these currently, but 4 it's not included in this cost. This is just for all 5 the permit infrastructure that would be in those 6 facilities.

7 VICE CHAIR REMPE: Okay, again, cost isn't 8 my thing, my purview here as an ACRS member. I'm more 9 interested will NRC have what they need to get the 10 data they need for decisionmaking. And this is a 11 partial story is what I'm hearing.

12 MEMBER PETTI: So Dan, is it fair to say 13 that this, the yellow bars in the blue represent what 14 sort of enables an experimenter to come in and execute 15 an experiment but they have to pay for their 16 experiment?

17 MR. WACHS: Yeah, I think that's a good 18 way to describe it, Dave.

19 MEMBER PETTI: Okay, thanks.

20 VICE CHAIR REMPE: True, because having 21 instrumentation takes a little bit more than -- it 22 takes a few years. As you know the GR -- test right.

23 It's, you guys paid for it but it took several years 24 in advance to get it qualified.

25 MR. WACHS: That's correct. And I think NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

225 1 the -- you have active collaborations with the advance 2 centers and the instrumentation program now that are 3 being executed in parallel to the design of the 4 experiment. So we are providing enhanced access to 5 the experiments. We're making some modification to 6 the plan in order to support that.

7 So we should be able to see the timelines 8 that you've historically seen associated with 9 instrumented experiments shortened substantially in 10 these. So we expect these to be fully instrumented.

11 We recognize the primary objective of this was to add 12 access to dynamic testing, and not just the start-and-13 end type testing that we would see with LTAs and 14 things like that. So instrumentation is a core 15 principle for these test rigs.

16 VICE CHAIR REMPE: Okay, the other 17 question I have is we all, many of us live in Idaho.

18 We have the difficulties that are political in nature.

19 Are they being addressed so people can send irradiated 20 fuel to Idaho to have them re-instrumented? How's 21 that going along?

22 MR. WACHS: I know Bill, do you want to 23 answer that, or would you like me to make a stab at 24 it?

25 MR. McCAUGHEY: Well, it's being worked, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

226 1 Joy. It's, you know, it is tied to the operation of 2 the integrated waste treatment unit, with the PM 3 program. We still have a linkage to that with the 4 state of Idaho, and that's being worked. The virus 5 has pushed back some schedules at Idaho. And so it's 6 just something that we're working and it's not totally 7 resolved yet.

8 MR. WACHS: But I would say our goal is to 9 have a functioning LWR materials library to use in 10 support of these capabilities. We've identified the 11 first handful of materials that we would like to 12 acquire for that library, but it's waiting for 13 obviously some of those things to be resolved.

14 VICE CHAIR REMPE: I just am looking for 15 risks that might affect the NRC's ability to get what 16 they need in a timely fashion. Thank you. Go ahead.

17 Or I guess you're done and it's back to Michelle, 18 right. Unless, colleagues, do you have any other 19 questions? Go for it, Michelle.

20 MS. BALES: Okay, thanks. I'm glad that 21 the DOE folks are here to help because I definitely 22 would not have been able to field some of those 23 questions. But I'm glad that we were able to get some 24 more details.

25 CHAIR SUNSERI: And since you've made that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

227 1 comment, let me make an interjection here. And I 2 don't mean to be overly restrictive here, but this 3 meeting was noticed to the public as an interaction 4 between ACRS and NRC staff, so any guest interaction 5 needs to be specifically requested or otherwise 6 acknowledged by the Chairman of the Subcommittee 7 leading the discussion.

8 And I don't mean to be rude on this thing, 9 it's just protocol because we are a FACA group and we 10 have rules and we have to abide by them. So I would 11 appreciate that. Thank you.

12 MR. FURSTENAU: No, that's a fair comment, 13 thanks, thanks for that reminder, Matt. We appreciate 14 it.

15 MS. BALES: Yeah, so I'm going to actually 16 transition to a different topic and then turn it over 17 to my colleagues to talk about other topics.

18 But what I want to say before we go on to 19 some of the other international, new international 20 collaborative research that you're going to hear about 21 is to say that one of the important lessons learned 22 from Halden's closure is that we need to be more 23 deliberative about capturing metadata for nuclear fuel 24 and material research.

25 And by metadata I mean information about NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

228 1 experimental design, the details of the experimental 2 design and experimental protocol and various factors 3 that are crucial for interpreting experimental 4 results. These pieces of information are critical to 5 putting experimental results in the appropriate 6 context for use by regulators and industry.

7 And I think when the closure of the Halden 8 reactor was first announced, part of the panic that 9 reverberated around the nuclear fuel research 10 community came from a realization that the 11 institutional knowledge housed at Halden might be 12 lost.

13 So it wasn't so much that obviously the 14 continued use of the facility was a big deal, but that 15 also we wouldn't be able to call up Halden anymore and 16 say hey, I finally got around to using this data that 17 you sent me five years ago and I don't really 18 understand how it was collected. Can you remind where 19 the neutron detection meter is and how does that 20 affect what I'm seeing in these results.

21 And so DOE's efforts to transfer 22 technology and expertise to the US will go a long way.

23 Later this afternoon, you'll hear from my colleague 24 about how a significant piece of the continuing Halden 25 research project effort since the shutdown is really NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

229 1 dedicated to capturing historic data knowledge, data 2 and knowledge and, quote, metadata.

3 But importantly, this lesson is being 4 applied in current -- in future research programs that 5 are being proposed now. New research proposals are 6 emphasizing the need to capture data and metadata and 7 make them accessible for long-term use. And so as I 8 turn it over to my colleague James, I'll point out 9 that you'll hear that point as a pillar of the FIDES 10 project.

11 So with that, I will turn it over to 12 James.

13 MR. CORSON: As Michelle had said, she 14 gave a nice overview of the US activities. And now 15 FIDES is the international community's response to the 16 closure of Halden. And as she had mentioned, it's not 17 just a way to connect a network of experimental 18 reactors, but to retain the Halden community itself 19 and the vast knowledge that they have on fuel and 20 material research.

21 So FIDES encompasses both the joint 22 experimental programs, so the actual experiments that 23 will be run, as well as cross-cutting activities that 24 make the most of those experimental results. And as 25 Michelle had just mentioned, you know, one of the main NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

230 1 cross-cutting activities is this issue of data 2 preservation and quality assurance. And so FIDES is 3 really trying to focus on that issue.

4 Cross-cutting activities also go into 5 training and education, the next generation of fuels 6 researchers. And also looking to make the most of 7 advances in modeling and simulation to help guide 8 experiments, and as well as have experiments help 9 validate these tools.

10 So the joint experimental program, that's 11 really the crux or the main point of FIDES, is to do 12 these experiments. And the -- which experiments get 13 proposed and funded and so on are guided by the 14 governing board of FIDES. And so each organization 15 that's involved has a say in this in the United States 16 -- sorry, is there a question? Okay.

17 So in the United States, the NRC has been 18 heavily involved, as I believe both Michelle and Ray 19 have said. Ray has been serving or leading the 20 establishment board for FIDES, and has led some of 21 those meetings. We're also working with the 22 Department of Energy on this. A lot of these efforts, 23 we're working in close coordination with them to 24 establish FIDES.

25 So again, yeah, I'll talk about a few NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

231 1 fuels-related JEEPs in just a second. But this is a 2 way to leverage funds from the international community 3 to do this work. So the core group, which are the 4 people who are actually doing the experiments and 5 making the sort of day-to-day decisions, they're the 6 ones who fund half the work roughly. And then the 7 rest comes from the larger pot of FIDES funds.

8 So that's the idea behind this new effort.

9 And now I'd like to talk about three of the upcoming 10 fuels JEEPs that have been proposed. Next slide.

11 MEMBER PETTI: James?

12 MR. CORSON: Yes.

13 MEMBER PETTI: What's the green colored 14 boxes, the light colored boxes?

15 MR. CORSON: So, the green colored boxes, 16 those are the ones that are the most mature proposals 17 at this point. So, they're the ones that would most 18 likely be funded in the first round of FIDES. The 19 white ones represent less fleshed out proposals at 20 this point.

21 And I have to say, for the white ones, 22 they may be slightly out of date, just because I think 23 this is taken from older FIDES materials. But, 24 certainly, the green boxes are the JEEPs that are 25 moving forward at this point.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

232 1 So, the first JEEP that I'd like to 2 highlight is on high burnup experiments in reactivity-3 initiated accidents. And this is primarily led by the 4 United States, at the Department of Energy, EPRI, and 5 the NRC is also a core group member for this. And so 6 I'd like to give some recognition to Bill McCaughey 7 and Dan Wachs for including us in this activity and 8 letting us have our say here. And this also involves 9 our colleagues at IRSN in France and JAEA in Japan.

10 So, this JEEP is going to look at the 11 effect of pulse width deposited energy in PCMI, 12 pellet-cladding mechanical interaction, failure in 13 reactivity-initiated accidents. So, this sort of gets 14 to Dr. Rempe's earlier question about differences 15 between reactors. For RAI testing, it's been done at 16 several different facilities, as highlighted in this 17 slide.

18 So, this JEEP, part of the goal is to look 19 at TREAT versus NSRR in Japan. What are the 20 differences? Does the pulse width make a difference 21 in the behavior that you see? So, it's trying to 22 address some of these issues about having distributed 23 facilities. So, next slide, please.

24 So, the second JEEP I'd like to highlight 25 is power to melt and maneuverability, or P2M. So, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

233 1 this is being conducted at the BR2 reactor in Belgium, 2 and also involves CEA and EDF in France. And so this 3 is looking at the margin to fuel centerline melting.

4 And this is important when you're talking about higher 5 power operations. Specifically, this test is going to 6 look at higher burnup fuel, so it will indirectly 7 address things like thermal conductivity degradation 8 that Dr. Rempe has mentioned several times. So, it 9 will also look at other issues, like fission gas 10 release when you get into these power ramp situations.

11 So, next slide, please.

12 And then the last JEEP I would like to 13 highlight is in-pile creep studies of ATF claddings, 14 or INCA. And this is being conducted primarily in the 15 Czech Republic at the LVR-15 reactor. So, for this 16 JEEP, they'll be looking at the effects of chromium 17 coating on zirconium alloy cladding and how that 18 impacts creep behavior.

19 And so, for the first round of tests, it's 20 primarily going to be capsule tests, where they stick 21 in a bunch of samples, irradiate them, and then take 22 them out and, you know, do their creep measurements.

23 But also as part of this first round they're looking 24 to qualify MELODIE device, which was previously 25 developed in the OSIRIS reactor in France, for the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

234 1 LVR-15 reactor. So, this is an example of using a 2 standardized test rig in multiple reactors, which is 3 something else that Dr. Rempe was interested in.

4 So, those are the three JEEPs, or three 5 fuel JEEPs, that I'd like to highlight. My colleagues 6 will be talking about the materials-related JEEPs 7 later. But, right now, I'd like to see if you have 8 any questions before I turn it over to Lucas 9 Kyriazidis.

10 VICE CHAIR REMPE: So, I'm not hearing any 11 questions from my colleagues, so I'm going to ask you 12 the same question I've been asking others today. Do 13 you see any near-term, or even maybe longer-term, 14 needs that take a while to get addressed because of 15 the need to get facilities and instrumentation and 16 test rigs qualified and you think will have less 17 margin and maybe we ought to be looking at? Or do you 18 have enough tests in the queue that you think things 19 are going okay?

20 MR. CORSON: I mean, I think, for the most 21 part, things are going pretty well. Michelle 22 highlighted a lot of the things that are going on in 23 the United States. Similarly, in the international 24 community, there is this recognition to look at 25 advanced instrumentation, to look at high burnup fuel NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

235 1 behavior, and to look at ATF cladding behavior.

2 There's a lot of talk about, you know, what 3 instrumentation do you use and what's appropriate?

4 So, I think we're on a good path right now. So, to 5 answer your question, I don't see any major issues 6 right now.

7 VICE CHAIR REMPE: If some of these 8 proposals don't go forward, is it going to be a 9 catastrophe? Or are you betting on these things all 10 happening?

11 MR. CORSON: Yeah, I wouldn't say it's 12 going to be a catastrophe; it'll just mean we're going 13 to have to adjust. But I think, you know, Ray brought 14 up this point at the beginning, the fact that we have 15 more of these distributed facilities, that if any one 16 of these proposals doesn't go through, I think there's 17 backup plans. And there might be slight delays, but 18 I wouldn't say it would be a catastrophe.

19 So, that's the benefit of having this 20 distributed network now, as well as the framework, 21 like FIDES, that can help fund these distributed 22 network of reactors.

23 VICE CHAIR REMPE: Okay. Thank you.

24 Colleagues, anything?

25 (No response.)

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

236 1 VICE CHAIR REMPE: Okay.

2 MR. CORSON: Okay. I'll turn it over to 3 Lucas.

4 MR. KYRIAZIDIS: So, good afternoon, 5 everyone. So, today, we talked a little bit about 6 facility upgrades that DOE is doing and some 7 international collaboration work through the various 8 JEEPs that James has highlighted, but another 9 important area that we need to discuss is where this 10 irradiated material is going to come from.

11 So, immediately after the Halden closure, 12 the nuclear community recognized that efforts should 13 be made to save the valuable irradiated samples 14 accumulated throughout the many years of testing at 15 Halden.

16 So, here, Studsvik Nuclear Laboratory had 17 proposed a new international project called SPARE.

18 And here SPARE would fund the transport of the highest 19 priority fuel samples from Norway to Sweden. And, 20 early in SPARE's proposal, NRC staff have been heavily 21 engaged in prioritizing the selection of fuel 22 specimens that would be transported and potentially 23 saved from final disposition and disposal.

24 So, here, NRC staff prioritized fuel 25 segments with enrichments greater than five weight NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

237 1 percent U-235, fuel segments with burnups above the 2 current U.S. limits, and many fuel segments subjected 3 to power ramps and interval testing.

4 But I do want to highlight that SPARE only 5 focuses on the actual transportation, and any future 6 research on transported segments would have to be 7 proposed and funded separately. But SPARE members 8 would control the release and future use of 9 transported fuel for a fixed period of time.

10 So, that's all I have that I wanted to 11 talk on SPARE. Maybe I'll give it a few seconds for 12 questions.

13 MEMBER PETTI: Yes, just a question. So, 14 these will be stored somewhere.

15 MR. KYRIAZIDIS: Yes.

16 MEMBER PETTI: I assume in Sweden. And 17 then, say the U.S. wants to have a specimen and, you 18 know, money is paid to ship it, and it's destructive, 19 so it doesn't go back. Is there some sort of body 20 that will say, yeah, it's okay that you can destroy 21 that one, but if Country X comes in and says, well, I 22 don't think that's the right thing, is there going to 23 be a steering committee, do you know, to kind of 24 adjudicate all this stuff?

25 MR. KYRIAZIDIS: So, there will be a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

238 1 working group with all the signing members of SPARE 2 that will discuss this. But maybe -- I don't know, 3 Michelle, if you want to handle part of this question?

4 MS. BALES: Sure. Yeah, so, the members 5 who pay for the transport have a say in what material 6 can be used, especially when destructive testing is 7 involved. And they have that right for a period of 8 five years from the last transport. After that, they 9 can renew it, but after that, it's up to Studsvik to 10 weigh those proposals.

11 But I think, importantly, for the five 12 years, and then if they choose to renew it, everybody 13 who pays into SPARE to transport the material gets to 14 determine the acceptability of specific proposals and 15 whether the transfer of material to another 16 organization for testing is valuable enough to justify 17 the potential destructive test transfer.

18 MEMBER PETTI: Yeah, okay.

19 MR. KYRIAZIDIS: So, I don't know if 20 there's any more questions. If not, we can go on to 21 the next slide.

22 VICE CHAIR REMPE: Please do.

23 MR. KYRIAZIDIS: Okay. So, here, I want 24 to wrap up some of the fuels work that we talked 25 about. So, with all the activities happening NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

239 1 throughout the U.S. and through international 2 collaboration, the nuclear community as a whole is now 3 a lot less vulnerable to a single future facility 4 closure.

5 Going forward, fuel safety testing will be 6 supported through a well-equipped, diverse, and 7 dispersed network of research facilities. And today, 8 we've heard of just some examples from my colleagues 9 of work being done here within the U.S. at DOE 10 facilities, but also overseas through various 11 international facilities, such as the JEEPs that James 12 had mentioned.

13 So, this work, specifically the 14 investments being made in the U.S., will provide 15 significant autonomy and add significant value for the 16 whole nuclear community, but also especially for U.S.

17 embassies. These investments will allow the NRC a 18 firsthand look and an early opportunity to get 19 involved, provide feedback to test plants, and become 20 intimately familiar with the research being proposed 21 and conducted at these facilities.

22 Even though the future outlook is robust, 23 one feat that Halden did offer the nuclear community 24 that the nuclear community will deeply miss was its 25 economy of scale. And this has been mentioned a few NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

240 1 times now, but Halden was a dedicated research reactor 2 and research center with a relatively large footprint 3 for performing nuclear fuel research for many, many 4 years. Halden provided the required real estate and 5 expertise for performing these long-term steady state 6 irradiation campaigns. And Halden was also equipped 7 with nearby PIE facilities, fuel fabrication 8 capabilities, and offered a robust suite of in-pile 9 instrumentation and measurements.

10 So, many of these replacement capabilities 11 will now be housed at various facilities with various 12 ongoing missions. And these facilities are also 13 typically separated by distance, sometimes 14 organizations, and sometimes staff. This will 15 inevitably introduce competition and inefficiencies 16 for performing safety research.

17 So, it's recognized that the balance 18 between cost and testing capacity will have to be 19 considered when investing in future upgrades and 20 funding new projects. And one example that comes to 21 mind was Halden's ability in performing long-term 22 steady state irradiation, but also its robust suite of 23 in-pile instrumentation.

24 It is known that this is quite difficult 25 and costly to reproduce, whether in the U.S. or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

241 1 overseas. The community as a whole will have to be 2 smarter and innovative in how we go forward, but we do 3 recognize the innovations and the aggressive plans 4 made to date from both DOE and the international 5 community.

6 And, lastly, I just want to state that the 7 community's response to the Halden closure has been 8 robust and swift. That's what I had on this slide.

9 I don't know if there's any questions for either 10 myself --

11 VICE CHAIR REMPE: When I go back and 12 think about the replacement facilities being used 13 right now, the Czech Republic reactor is pretty small, 14 with respect to it's got like one loop location, and 15 the MIT reactor has one loop location, I believe. You 16 have a loop at the ATR, but a lot of people want it.

17 And, yes, there's something in Russia, but it seems to 18 me that might be more difficult for things of interest 19 to the NRC to get. It sure sounds like making sure 20 that you have the ATR loop being established is going 21 to be important.

22 But maybe it isn't, maybe you can rely on 23 sharing the limited number of international 24 capabilities. The Jill Ford (phonetic) I guess I've 25 heard now is not going to go up and run until 2030. So NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

242 1 it seems like locations are going to be very 2 important. But, again, I'm just looking at it as a 3 reviewer. Do you have the same impression?

4 MS. BALES: So, I can respond to this, 5 because we've talked about this a lot with Lucas, and 6 also I think Ray mentioned it at the beginning. I 7 think, with the capacity being substantially less than 8 Halden, we have the requirement to be smarter with 9 what testing we're going to do and be really precise 10 about what data we really need. And so I'm hopeful 11 that the lower capacity will be met with innovation in 12 test planning and really thinking through what we 13 really need.

14 In some ways, the Halden capacity was a 15 luxury that we might have taken for granted. And I 16 think, going forward, we just have to be smart. And 17 I think that that's sort of what Lucas's third point 18 gets at, is, if we wanted to have the capacity that we 19 had at Halden, we could probably do it, if we had 20 infinite resources. But it just means that cost and 21 testing capacity are going to be in competition, and 22 we have to look for the right balance between those to 23 ensure that we're getting the data that we need for a 24 cost that is feasible.

25 VICE CHAIR REMPE: Unless other colleagues NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

243 1 have questions? Please go on, then.

2 MEMBER KIRCHNER: I have an observation, 3 Joy. I was going to save it for the end, but maybe 4 it's appropriate with this viewgraph.

5 I think, positively, the shock of losing 6 Halden -- which we were comfortable with, I think was 7 just described as a luxury -- it's going to force a 8 sharpening of minds, pencils, and budgets. That's not 9 a bad thing. And I see there's opportunity here to 10 reinvigorate both the fuels and materials research 11 communities, as a result, instead of just coasting 12 along as we perhaps were with the luxury, 13 quote/unquote, of Halden. So, I take what I've heard 14 so far very positively.

15 VICE CHAIR REMPE: Thank you. Any other 16 members want to make a comment?

17 Let's go on, then.

18 MR. KYRIAZIDIS: Thank you for those 19 comments. So, I think going on, we're going to 20 transition into the structural materials section. So, 21 with that, I will pass it on.

22 MR. HISER: Thanks, Lucas. My name is 23 Matt Hiser. I'm a materials engineer in the Division 24 of Engineering in the Office of Research. And I've 25 been working on irradiated materials and, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

244 1 specifically, involved with the materials work at 2 Halden for about the last ten years, working alongside 3 Michelle and some of the other folks in the fuels area 4 looking at Halden. So, maybe go ahead to the next 5 slide, Michelle.

6 So, the first slide, I'm sort of taking 7 this in a chronological order and setting the scene 8 for where we were prior to the reactor shutdown. And 9 so Halden was a key aspect of the irradiated materials 10 research that NRC was doing. It wasn't the entirety, 11 or necessarily even a majority, but it was a key 12 piece. And Halden offered some unique capabilities, 13 relative to other facilities, for materials research.

14 It also, as has been mentioned before, offered 15 excellent value and leveraging. And so I just want to 16 note that Halden has contributed in the materials 17 research area quite a lot.

18 So, just touching on the capabilities, and 19 particularly the in-reactor testing capabilities for 20 stress corrosion crack growth rate testing and 21 instrumented creep and stress relaxation testing.

22 Those are, to my knowledge, fairly unique, to have 23 those in an in-pile, and those were some of the 24 capabilities we were using while the Halden reactor 25 operated.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

245 1 On the value and leveraging side, as I 2 note there in the sub-bullet, we split the funding 3 with other countries, other U.S. organizations, and 4 even our other research areas at NRC, to make this 5 program very cost-effective.

6 And then, finally -- and I'll come back to 7 this, and we'll come back to this through the 8 materials slides -- there were three key specific 9 materials research activities that we were pursuing at 10 Halden. And I think we have sort of a strategy to 11 continue those specific activities, so I'll just touch 12 on them here. And then you'll hear more about them in 13 future slides.

14 But they were are further irradiation and 15 testing of ex-plant harvested stainless steel welds, 16 SCC crack growth rate testing of high dose harvested 17 ex-plant stainless steel base materials, and then 18 creep and stress relaxation testing of baffle-former 19 bolt materials. And so I think we'll, as we get into 20 these slides, see how our strategy sort of addresses 21 some of those specific activities, as well as the 22 larger capability development. Next slide.

23 So, as we move into the post-Halden 24 materials research strategy, cooperation is a key 25 aspect of it. And, as with the Halden reactor, we're NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

246 1 looking to focus on the value that we can get for NRC 2 interests, particularly relevant to U.S. nuclear plant 3 operations.

4 So, following the reactor shutdown, NRC 5 staff did look at, you know, what were the values and 6 what were the gaps that were created by losing the 7 Halden reactor, and how can we mitigate those impacts?

8 And so we identified three approaches that would help 9 to address our needs going forward and mitigate the 10 impacts from the reactor shutdown.

11 So, first was preserving knowledge from 12 Halden. And I'll just point out here, you'll notice 13 the color-coded text on the right side of the slide, 14 and you'll notice that carried through in the 15 remaining materials slides. So, you can map back the 16 red referring to Halden activities, blue referring to 17 DOE activities, green for FIDES JEEP activities, and 18 then orange for harvesting activities.

19 So, the first one is preserving knowledge 20 from Halden, which is being addressed by a combination 21 of activities through the Halden project, as well as 22 some DOE initiatives.

23 Performing some of those high priority 24 research items that I identified on the prior slide.

25 That's being addressed both through the Halden NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

247 1 reactor, the finishing up of work at Halden, a couple 2 of proposed JEEPs, materials JEEPs through FIDES, as 3 well as harvesting.

4 And then, finally, developing new 5 capabilities is also a key sort of long-term piece of 6 our strategy. And the FIDES JEEPs, as well as some 7 DOE investments and coordination, is critical there.

8 And so I'd just like to point out one 9 aspect when it comes to the irradiated materials, 10 particularly high dose materials, while we look to 11 cooperate wherever we can, and we have found good 12 international cooperation through Halden and some of 13 the other things that we'll discuss on these slides, 14 I will note that this is becoming increasingly a sort 15 of U.S.-focused concern, as other countries in Europe, 16 in particular, are not necessarily looking at 17 operating plants out to 80, 100 years. Some of these 18 really high dose effects and very long-term aging of 19 reactor internals become a little more of a U.S.

20 focus. And so developing domestic capabilities is 21 probably a prudent long-term strategy for the NRC and 22 for the U.S. as a whole. So that will be part of the 23 theme as well in these slides. Next slide.

24 So, just the first key piece of the 25 strategy. And this slide is just going to sort of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

248 1 touch on the strategy, as well as by really going over 2 what's planned at the Halden reactor post-shutdown.

3 So, there are three key aspects of their post-reactor 4 shutdown activities, and these are true both for the 5 materials and the fuels research.

6 So, first was the orderly completion of 7 ongoing work. This means finishing post-irradiation 8 examination on samples, obviously not getting any more 9 radiation dose once the reactor is shutdown, but 10 finishing up any testing that was planned or 11 characterization after irradiation. Preserving key 12 samples, fuels and materials samples, that are of 13 research value. Lucas touched on the SPARE program, 14 and I'll touch later on this slide on the plans in the 15 materials area. And then, finally, documentation and 16 capture of past Halden data, which also got mentioned 17 during the fuels presentation.

18 So, I'll just note, in the materials area, 19 some of the key ongoing experimental work that was 20 being wrapped up is nearly complete. I don't know 21 that we have final reports, but I think the work has 22 largely been completed. And our main interest was 23 some of the SCC crack growth testing of high dose ex-24 plant materials, high dose stainless steels.

25 Then, the second bullet, we have to credit NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

249 1 our colleagues at the Department of Energy. Through 2 the Nuclear Science User Facilities, they have a fuels 3 and materials library. And they have graciously 4 stepped up to acquire, and are planning to acquire, 5 some of the high-value irradiated materials samples 6 that are at Halden and would otherwise be disposed of 7 in the decommissioning of the Halden facility.

8 So, the intention is that will happen 9 within the next two to three years and be transferred 10 into DOE and then be available for research use. So, 11 NRC has supported that through our role as a Halden 12 member, and we're pleased that that looks like it's on 13 track to take place and not lose valuable research 14 specimens.

15 And then, finally, Halden is, as Michelle 16 indicated in the earlier slides, they are putting 17 quite a bit of effort into capturing prior data and 18 knowledge, and, particularly, developing a legacy 19 database and doing some analyses on some of the key 20 fuels and materials issues, to sort of synthesize the 21 data that has been collected over the past 60 years of 22 Halden reactor operation.

23 So, maybe I'll just pause for a moment and 24 see if there's any questions on these first few slides 25 before I turn it over to my colleague, Eric Focht, to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

250 1 present on the FIDES JEEPs.

2 MEMBER KIRCHNER: Joy, this is Walt again.

3 I'd like to try something on the presenters. I 4 understand the value of having the high dose samples.

5 I'm sure and hopeful that you're going to harvest 6 those. I don't know the details in terms of actual 7 fluences and such. Is there enough information to be 8 gleaned from the Halden operations and samples to kind 9 of address the kind of issues that come up with going 10 plus-20, and then another plus-20, in terms of 11 fluence?

12 So, it's a general question: do we have 13 fluences that capsule samples for stainless steel and 14 other materials used for reactor internals and vessels 15 that encompass, say, an 80-year exposure?

16 MR. HISER: So, the materials from the 17 Halden that we're talking about preserving from 18 Halden, no, those alone do not do it. So we have 19 other initiatives that we're planning, some of which 20 we'll touch on on the later slides, on new harvesting.

21 And just to be clear, in the materials 22 area, Halden would often get specimens from commercial 23 reactors that may have been harvested, in some cases 24 a baffle bolt or other pieces from internals. And so 25 they were not necessarily very super-high dose, but NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

251 1 they had been tested and characterized. And, in some 2 cases, specimens weren't tested and could be available 3 for further irradiation and testing. So that's part 4 of the reason we wanted to preserve them.

5 Maintaining the metadata, if you will, for 6 these specimens is an important piece that DOE is 7 focused on and making sure that they -- you know, they 8 may not be able to get all the materials, but they 9 will focus on the ones that they see as having the 10 most research value in the future.

11 But, no, this is sort of a housekeeping 12 activity we think is important and it's not going to 13 be tremendously costly. But, no, it's definitely --

14 these Halden materials preserved from Halden are not 15 going to cover us to 80 years for reactor internals 16 aging by themselves.

17 MEMBER KIRCHNER: Okay, thank you.

18 MR. HISER: Okay. So, with that, I'll 19 turn it over to Eric to talk about the FIDES JEEPs.

20 VICE CHAIR REMPE: Real quick on the in-21 pile stress corrosion crack growth. Wasn't MIT the 22 one where they were doing some testing for in-pile 23 capabilities in that area? And I, again, have not 24 kept up with some of the things, but how far along are 25 they in getting data?

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

252 1 MR. FOCHT: I'll touch on that. One of 2 our JEEPs is going to benefit from that work.

3 VICE CHAIR REMPE: Okay.

4 MR. FOCHT: I can't speak specifically 5 about it, but I can give you some idea --

6 VICE CHAIR REMPE: That sounds fine.

7 MR. FOCHT: Okay. Well, good afternoon, 8 everyone. My name is Eric Focht from the Division of 9 Engineering in the Office of Research. And, as Matt 10 mentioned, I'm going to be talking about the FIDES 11 JEEPs.

12 And the Halden program definitely offered 13 some unique and valuable testing capabilities. After 14 the program shutdown, the NRC and EPRI sought options 15 for obtaining stress relaxation data on baffle-former 16 bolting materials, which, as Matt mentioned, was 17 testing originally planned to be performed under the 18 Halden program. We also recognized the need for in-19 core mechanical testing capabilities, specifically, as 20 we just said, stress corrosion cracking crack growth 21 rate testing.

22 Thus, there are currently two structural 23 materials JEEPs proposals being developed by INL for 24 consideration by the FIDES program. INL will be the 25 operating agent for both JEEPs, and the NRC will be a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

253 1 part of the core groups, along with EPRI, for both 2 JEEPs. And that will allow us the ability to provide 3 input into the projects.

4 The objective of the stress relaxation and 5 creep research JEEP, or STRETCH JEEP (phonetic), is to 6 obtain irradiation-induced stress relaxation and creep 7 data on baffle-former bolting materials. To get EPRI 8 the data they need as quickly as possible, the testing 9 will be performed in the MIT reactor, which can 10 produce the dose needed in a PWR environment and will 11 utilize stress relaxation testing rigs developed by 12 INL.

13 This JEEP leverages resources from the DOE 14 Advanced Sensors and Instrumentation Program and the 15 Nuclear Materials Discovery and Qualification 16 Initiative, or the NMDQi. And I should mention that 17 the support from these programs will enable this JEEP, 18 both JEEPs, actually, to begin sooner than they 19 otherwise would have without their support.

20 DR. CORRADINI: Just one question. Does 21 the MITR exposure, is it a one-to-one time or is it 22 accelerated?

23 MR. FOCHT: I think it simulates PWR 24 conditions pretty well, so I think it's -- I'm not an 25 expert in that area, but I think it's -- maybe Ron NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

254 1 could chime in, or Matt -- it's on the order of a 2 couple DPA per year. So they're going to be exposed 3 for several years to get the data they need.

4 VICE CHAIR REMPE: I think Ron's been 5 trying --

6 (Simultaneous speaking.)

7 MEMBER BALLINGER: It's one-to-one.

8 MR. FOCHT: Okay. Thank you.

9 MEMBER BALLINGER: There's no 10 acceleration.

11 MEMBER KIRCHNER: I was going to ask Ron 12 or Pete, if they're online, this is Walt, how many 13 plants still use the baffle bolts versus welded?

14 MEMBER RICCARDELLA: Almost all.

15 MEMBER BALLINGER: Yeah. My guess is 16 almost all of them. APR1400 uses welded. AP1000, I 17 think is welded. That's about --

18 MEMBER KIRCHNER: I was just curious 19 whether, in the life extension space, they were 20 replacing the baffle bolts design with the welded 21 baffle-former.

22 MEMBER BALLINGER: Pete probably knows 23 well, but I don't think a U.S. plant would replace the 24 shrouds, or whatever they call it, the barrel.

25 MEMBER RICCARDELLA: I think that would be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

255 1 prohibitive, Walt. But they are replacing baffle 2 bolts.

3 MEMBER BALLINGER: Yeah.

4 MEMBER RICCARDELLA: They can do that when 5 they find them, but --

6 (Simultaneous speaking.)

7 MEMBER RICCARDELLA: And I would say, 8 probably 90 percent of PWRs right now have baffle 9 bolts.

10 MEMBER KIRCHNER: Still have them? Okay, 11 thank you.

12 MEMBER BALLINGER: And I think one of the 13 Studsvik reactors, and maybe one of the Japanese 14 reactors, have replaced shrouds or core barrels, but 15 not in the U.S.

16 VICE CHAIR REMPE: So, let's go ahead and 17 keep going, because I'm a little worried about time, 18 because --

19 MEMBER BALLINGER: Okay.

20 MR. FOCHT: Sure, thanks. The development 21 of in-core mechanical testing capabilities, or the 22 ENCORE JEEP, is focused on developing in-core testing 23 capabilities to benefit both light water reactors and 24 advanced non-light water reactors. The goal is to 25 develop testing capabilities at the ATR that not only NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

256 1 retain those lost due to the Halden program shutdown, 2 but to go beyond that, with capabilities for testing 3 in non-light water environments, such as at elevated 4 temperatures and in coolants such as molten salts.

5 An important aspect of this project is 6 that developing these capabilities at a DOE lab will 7 provide access and expertise over the long-term to 8 meet the data needs of both the U.S. nuclear community 9 and the NRC, particularly the need for data on 10 structural materials performance in advanced non-light 11 water reactor environments.

12 The ENCORE JEEP will focus initially on 13 stress corrosion cracking crack growth rate testing, 14 and takes advantages of previous work at INL and MIT 15 performed to develop an SCC crack growth rate testing 16 rig based on the Halden design. And, as I mentioned, 17 just like the STRETCH JEEP, this JEEP also leverages 18 resources from the DOE ASI Program and the NMDQi.

19 So, yes, Dr. Rempe, we are using the test 20 rig very similar to the Halden design that was being 21 developed at MIT.

22 VICE CHAIR REMPE: Yeah, I think --

23 MR. FOCHT: They did a performance test on 24 that, I think, and in their -- it may have been an 25 autoclave -- no, actually, it was in-core, they have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

257 1 done in-core testing --

2 VICE CHAIR REMPE: I'm pretty sure Joe 3 Palmer got this with the MIT reactor. I don't know 4 how it turned out, but I know it went in.

5 On the prior one, with the LVDTs, has that 6 ever been irradiated or is this the first time? I'm 7 sure they've done autoclave testing, but I don't know 8 if they've ever irradiated it. So, it's kind of 9 interesting that this is a JEEP where, unless it --

10 did it go in a reactor yet, or do you know, Eric? On 11 Slide 25.

12 MR. FOCHT: I don't think so. I think 13 it's been autoclave tested. I don't know if Colin 14 Judge is on the line, he could confirm that.

15 MR. JUDGE: I know the rig's been put into 16 MIT and they've done testing. I'm not sure if they've 17 done them with the LVDTs in place.

18 VICE CHAIR REMPE: Okay. Because I just 19 was curious. So, the JEEP might be helping to fund 20 some of this instrumentation development in a way, 21 because it's putting it in a reactor the first time, 22 which I hadn't caught on from what I've read. But, 23 anyway, go ahead, because, again, I'm delaying you.

24 MR. FOCHT: No, that's okay. I appreciate 25 the questions. So, I guess, one thing I'd like to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

258 1 point out -- I don't know if Matt will follow up with 2 this -- but, traditionally, it's been my understanding 3 that the Halden program was a balance of materials, as 4 well as fuels testing. So we feel these JEEPs are 5 important for the FIDES program to help kind of 6 maintain that balance of fuels and materials testing.

7 So, hopefully they'll agree and we'll get these JEEPs 8 funded.

9 So, with that, if there's no other 10 questions, I can pass it back over to Matt.

11 MR. HISER: Thanks, Eric. So, I just have 12 a couple slides on harvesting, and then we'll wrap 13 things up on the materials side.

14 So, I just want to introduce the sort of 15 final pillar, if you will, of our materials post-16 Halden research strategy, is looking at ex-plant 17 materials harvesting. And just to clarify 18 terminology, this means taking previously or service-19 irradiated pieces of material and reactor internal 20 components out, and then doing testing and 21 characterization on them to confirm our understanding 22 of their performance.

23 And so, you know, harvesting can be done 24 both on irradiated and unirradiated materials. We've 25 actually put quite a bit of time and energy into NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

259 1 harvesting over the past few years and trying to do it 2 strategically in a way that will give the most value.

3 So, again, it can provide highly 4 representative aged materials for research. And this 5 is particularly true for irradiation effects on 6 materials. It can be challenging to replicate 30, 40, 7 well, 50, 60, 70, 80 years of operation in 8 irradiation, and, particularly, flux effects and 9 neutron energy spectrum through test reactors. So, if 10 nothing else, harvesting provides a good validation 11 and confirmation of what we learn through other types 12 of irradiation studies.

13 And so I just want to note, on the broader 14 topic of harvesting, NRC staff has spent some energy 15 developing harvesting priorities. And that's for not 16 even just metallic components or irradiated 17 components, but also for concrete and electrical 18 components, which are also in the scope of license 19 renewal and subject to long-term aging effects.

20 We've also been working to identify 21 harvesting opportunities and been engaged with 22 partners, both DOE and EPRI, as well as the 23 international community. We've hosted a couple of 24 international workshops focused on harvesting, one at 25 NRC headquarters about four, almost four years ago, in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

260 1 March 2017, and then another one at NEA headquarters 2 just about a year ago, just before we stopped being 3 able to meet in-person.

4 And so, just to shift this discussion back 5 to the focus of this meeting, in the irradiated 6 materials arena we've identified a few key harvesting 7 priorities from an NRC perspective. So, the first one 8 is high fluence stainless steel welds. The second one 9 is high fluence and high temperature stainless steel 10 materials. And you'll note the high fluence is 11 defined a little differently, and this has to do with 12 where welds -- the doses that welds see in most plants 13 versus just base materials. And then, finally, 14 irradiated cast austenitic stainless steel materials.

15 So, those are some of the key irradiated 16 materials harvesting priorities that NRC has. So, 17 maybe, Michelle, you can go to the next slide?

18 So, just shifting gears from the broader 19 materials harvesting discussion, there is a key 20 international cooperation that is getting ready to 21 kick off here known as the Studsvik Materials 22 Integrity for Life Extension Project. And this is a 23 planned cooperation through the NEA that's being led 24 by Studsvik, which is also involved quite a bit in the 25 fuels research worlds, as you heard earlier.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

261 1 The program is structured similarly to the 2 Studsvik Cladding Integrity Project. It's a five-year 3 scope of work, anticipated to begin this year. And we 4 think SMILE will provide a lot of value to some of our 5 key priorities for metallic light water reactor 6 components, both irradiated and unirradiated 7 materials.

8 So, it includes the harvesting, 9 characterization, and testing of metallic components 10 from both a BWR and a PWR in Sweden. And from a 11 materials standpoint, it covers an array from the 12 reactor pressure vessel low alloy steel to stainless 13 steel internals, including the core shroud and barrel, 14 respectively, in a BWR and a PWR; the baffle plate and 15 baffle-former bolts from the PWR; and then, finally, 16 nickel alloy penetrations, piping, and alloy-690 steam 17 generator tubes are also a part of this planned 18 program.

19 And just to give a little information on 20 the reactors that the harvesting will be done from.

21 The first, the PWR is Ringhals 2, which is a 22 Westinghouse design three-loop PWR with about 30 EFPY.

23 And then the second is Oskarshamn 2, which is a ABB 24 Atom design BWR with approximately 30 EFPY.

25 MEMBER BALLINGER: This is Ron. I'm glad NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

262 1 to see you're working with that Studsvik Materials 2 Laboratory. It is outstanding. It's really 3 outstanding.

4 MR. HISER: Yes, we've been involved with 5 some previous projects that just wrapped up on some of 6 those irradiated materials with Studsvik. Yes. We've 7 had good experience with Studsvik.

8 MEMBER BALLINGER: Yeah, you can stick a 9 block of irradiated material in one end and out comes 10 specimens and TEM and everything at the other end.

11 It's really very highly integrated.

12 MR. HISER: Any other questions on this 13 slide? Or we can move on to the next one, which sort 14 of burrows down a little further into SMILE.

15 So, just, again, to the scope of this 16 meeting, focused on irradiated materials, this slide 17 just gives a little more information on the irradiated 18 materials that we see of most interest. So, we're 19 primarily interested in some of the PWR materials that 20 are higher dose.

21 And so this table gives a little flavor of 22 the grade of stainless steel, as well as the maximum 23 dose that's estimated from some of the different 24 components, the baffle plate, baffle bolts, core 25 barrel, which is where you're going to see your NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

263 1 highest dose stainless steel welds, and then some of 2 the other internal components.

3 So, we're particularly interested in the 4 very high doses on stainless steel plate and baffle 5 bolt materials, which would be about the highest we're 6 aware of being harvested from an LWR where we can have 7 enough material to do crack growth rate testing, as 8 well as the fairly healthy dose on the stainless steel 9 welds with 30 years of thermal exposure as well.

10 And I'll just note, lower in the slide, we 11 have both characterization, microscopy, retrospective 12 dosimetry, and hydrogen helium content, as well as 13 mechanical testing, tensile, crack growth rate, and 14 fracture toughness is all planned as part of the scope 15 of this program.

16 So, if no other questions on SMILE, we can 17 move, and I just have one last slide to wrap up the 18 materials part of the presentation.

19 So, again, just to echo back to our 20 strategy on Slide 23, sort of some key aspects are 21 preserving knowledge from Halden, and that's being 22 done both through the Halden activities, as well as 23 the DOE initiative to preserve samples and move them 24 into the DOE library. And then for performing high 25 priority research, we have contributions from SMILE, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

264 1 from the FIDES JEEPs led by DOE, as well as some of 2 the Halden activities that are wrapping up.

3 And then developing new capabilities is 4 really focused on the DOE-led FIDES JEEPs, and we see 5 those as hopefully providing valuable research in 6 their programs, but then also the capability to 7 provide benefits down the road.

8 And I don't know if it got emphasized 9 earlier, but we also see these capabilities as being 10 important for non-light water reactors, too. That's 11 part of the plan that DOE has in place, is to first 12 demonstrate some of these capabilities with light 13 water reactor conditions, but with the vision to 14 expand. And that can truly be a value to the NRC and 15 to the U.S. nuclear research community down the road.

16 And then, finally, just long-term testing 17 of highly representative materials is how NRC is 18 focused, particularly for irradiation effects and 19 long-term aging of reactor internals. And so 20 harvesting is going to be a key aspect of that. We 21 try to be selective and make sure we're -- harvesting 22 is very challenging and expensive, and so we try to 23 pick our spots and identify the highest value items.

24 But we also see the value of pursuing that where it 25 makes sense, and SMILE appears to be a good example of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

265 1 that.

2 And so, in summary, just to wrap up on the 3 materials side, we think the closure of the Halden 4 reactor has caused some delays and will probably 5 increase costs modestly to fill some of our short-term 6 needs, but it's really just accelerated probably some 7 transitions and adjustments that we were going to need 8 to make anyway, because Halden was not going to 9 operate forever and we need to have domestic 10 capabilities on some of these things.

11 And so this is really giving us a push 12 into the direction that we need to go to be prepared 13 for non-light water reactors, and the long-term aging 14 of light water reactors as well.

15 So, any questions?

16 MEMBER KIRCHNER: This is Walt Kirchner.

17 What are the major codes that Research holds or 18 maintains and develops to actually take this 19 information and use it to benchmark and verify and 20 validate the codes so that if, indeed, we cannot get 21 80 years of fluence, we have a reasonable feeling 22 about the codes' abilities to project performance 23 under -- I'm thinking first of the LWR fleet, and then 24 I'm thinking of the higher irradiations that you'll 25 see in some of these fast reactor designs and such.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

266 1 Is there a close coupling between your different 2 branches on this aspect?

3 MR. HISER: So, I would say that -- you're 4 talking about, like, modeling and simulation codes, I 5 presume?

6 MEMBER KIRCHNER: Yes, exactly.

7 MR. HISER: Yeah, I was going to say, we 8 don't -- particularly for reactor internals, there 9 have not -- I think DOE has developed some codes, you 10 know, with challenges in validation, for irradiation 11 effects. It's really a challenging phenomenon to 12 model. And the safety significance is -- you know, 13 it's not insignificant, but these are not pressure 14 boundary components. So, there have been failures of 15 baffle bolts, for instance, and these things have 16 tended to be managed through experimental research, 17 engineering judgment, and then inspections, has been 18 generally the strategy that the industry has followed 19 and that we have found acceptable.

20 Now, we certainly are open to looking at 21 validating and benchmarking, and developing codes that 22 could be validated and benchmarked, but that hasn't 23 been really the industry or NRC approach in this area.

24 MEMBER KIRCHNER: No, I appreciate that, 25 and I appreciate the value of having the empirical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

267 1 evidence. I'm just thinking that, as we go forward, 2 or as research goes forward -- and I hope the industry 3 is thinking of this, because if they're not prepared 4 to do the integral testing at representative fluences, 5 et cetera, and prototypical operating conditions, then 6 they need some kind of convincing and validated tools 7 to make the projection that the particular design 8 aspect is going to meet the requirements and satisfy 9 you, the staff, in a review, in a licensing review, 10 that, in the absence of an integral test with many 11 years of irradiation, this is an acceptable design.

12 Otherwise, it throws you back on saying, 13 well -- and this is making it too simplistic, but you 14 need to have a prototype, which was the old way of 15 going forward on advanced reactors.

16 MR. HISER: Right. Well, and what I will 17 say is, you know, when you look at the spectrum of 18 dose on reactor internals, you have the re-entrant 19 corners on baffle plates, and those see a significant 20 acceleration in dose relative to even just the 21 exterior-facing corners of the baffle plate. And then 22 there's a significant reduction in dose out from 23 there. So, through harvesting, you can harvest sort 24 of the highest dose corners and really be able to 25 bound a large fraction of the internals out to a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

268 1 significant lead factor, if you will. So, there are 2 approaches. We're not --

3 MEMBER KIRCHNER: Well, the whole basis 4 for PWR -- and it's true, I think, also for the BWRs 5 -- to give confidence and conservativism in terms like 6 the boiling pressure vessel code, and so on and so 7 forth, I mean, you design the reactor with the baffle 8 for a number of reasons, flow as well as having a 9 space there that's filled with water that reduces the 10 fluence on the vessel.

11 But as we go forward to some of these 12 advanced concepts, they're not going to have that, how 13 should I say, buffer baffle kind of effect, and we're 14 going to see fluences at high temperatures, much 15 higher than you would see in a LWR fleet.

16 And that's where I was going with this.

17 If we're not going to see with these advanced reactors 18 the kind of integral testing out through the exposure, 19 lifetime exposure, then it suggests that they're going 20 to have to rely on some kind of means to extrapolate 21 performance, and that sooner or later gets you either 22 to a kind of simplistic estimates of exposure and all 23 the other factors, or you have some kind of physical-24 based code to project performance further out.

25 So, that's what I was thinking about when NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

269 1 you were presenting this material section of the 2 presentation.

3 MR. HISER: No, and that makes sense. And 4 I have to say, I personally have lived more in the LWR 5 world and been focused more on, yeah, the current 6 fleet. I know that there are other staff in my branch 7 that are more focused on the advanced reactor piece.

8 And I'm sure that, yeah, probably modeling and codes 9 will need to be a bigger piece of the puzzle on that 10 side.

11 But I will say, I think the ENCORE JEEP is 12 designed to develop some of the experimental testing 13 capabilities, which there will at least need to be 14 some experimental testing, I'm sure, but maybe not as 15 much, as was stated earlier.

16 VICE CHAIR REMPE: So, we are kind of 17 short on time. And so, if my colleagues will let me, 18 I'd like to go ahead and go to the next presenter 19 directly. I believe this is your last one, right?

20 Your last slide?

21 MR. HISER: Yeah, that was my last slide.

22 Yeah.

23 MS. LUND: Yeah, good afternoon. And we 24 very much appreciate the opportunity to discuss this 25 important topic with you this afternoon. And, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

270 1 summarizing what you've heard today, while we 2 recognize the tremendous research benefits we've 3 gained by our participation in the Halden Reactor 4 Project, and are grateful for its uniqueness and 5 longevity, we're committed to being ready as an agency 6 to meet our future regulatory challenges.

7 The closure of the reactor has certainly 8 caused us some short-term pain and has led us to 9 refocus and broaden our efforts to fill those 10 immediate research gaps that were planned to be 11 addressed by Halden.

12 One immediate focus is, as you've heard 13 today, the orderly termination of the Halden Reactor 14 Project to best preserve the knowledge gained over the 15 course of the project. And also consistent with what 16 you've heard today, the reactor's closure has spurred 17 us to rigorously reevaluate our research and the 18 associated regulatory needs, and has accelerated our 19 strategic planning to meet those needs, for both the 20 existing fleet of light water reactors and future 21 reactors that will have a broader array of 22 characteristics and capabilities. Future fuels and 23 materials challenges will require experimental 24 facilities beyond those possible with Halden.

25 So, these expanded research needs, coupled NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

271 1 with economic realities that make it infeasible for 2 the NRC to independently support extensive and 3 expensive fuels and materials irradiation and testing, 4 are the principal factors driving our current pursuits 5 of focused, innovative, aggressive, and diversified 6 activities.

7 To focus even sharper on identifying and 8 addressing the most safety-significant concerns 9 through our research portfolio, technical innovation 10 is imperative to better couple advanced experimental 11 techniques with state-of-the-art modeling and 12 simulation tools to optimize experimental programs and 13 more intelligently mine the results. This innovation 14 is necessary in our pursuit of aggressive goals to 15 both accelerate fuels and materials qualification and 16 certification and implement experimental findings more 17 efficiently in technical practice and associated 18 regulatory requirements.

19 Diversification is imperative in realizing 20 these goals. We will need a variety of domestic and 21 international partnerships, which you've heard about 22 today, some of which are well-established, but others 23 that are still conceptual, to best leverage funding, 24 capabilities, and expertise going into the future.

25 An array of programs are being planned to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

272 1 address our more immediate research needs. Some of 2 these are leveraging existing programs at DOE, and 3 some activities have been initiated to fill the void 4 left by Halden. An array of experimental capabilities 5 will be needed to address future needs, and those 6 needs are being individually matched to the optimal 7 facility to create a dispersed research network that 8 will allow us to nimbly adapt to the expected 9 accelerated future pace of change.

10 DOE's investment in U.S. infrastructure is 11 central to our entire post-Halden strategy. These 12 past investments have allowed us to mitigate the near-13 term impact of the Halden closure. However, continual 14 investment is critical for our long-term success. NRC 15 hopes to work closely with DOE to align on an optimal 16 investment strategy.

17 In closing, as we look to the future, we 18 realize we can't do things exactly the way that 19 they've been done in the past, and I think that's 20 already been mentioned many times in this 21 presentation. However, we believe that the activities 22 we've presented today illustrate how we will continue 23 to meet the agency's fuel and materials research 24 mission going forward, despite the loss of the Halden 25 reactor.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

273 1 And that closes my remarks.

2 VICE CHAIR REMPE: Thank you, Louise.

3 Colleagues, does anyone have a question 4 for Louise?

5 So, not hearing anyone, I'll ask a 6 question. My earlier comment about the need for some 7 domestic locations are tied to this bullet that you 8 have on the far right, the lower one. And I often 9 mention the fact that when they reorganized the Atomic 10 Energy Commission, that there's wording in that act 11 that discusses the fact that the head of what is now 12 the Department of Energy, as well as other federal 13 agencies, are expected to provide the Commission the 14 facilities that it needs to accomplish their mission.

15 And so I assume that this point is being 16 conveyed to the Commissioners and to the DOE and 17 decision-makers, because I'm not even sure if DOE 18 (audio interference) affect that, right?

19 MS. LUND: There was an interruption in 20 your last sentence.

21 VICE CHAIR REMPE: Well, I assume that 22 this bullet is being conveyed to the Commissioners, 23 because -- I think you got the point about what I'm 24 saying is --

25 MS. LUND: Yes. Yes.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

274 1 VICE CHAIR REMPE: It's important that 2 decision-makers that affect the DOE investments are 3 aware and communicate that to people who affect what 4 the DOE does, right?

5 MS. LUND: Yes. I think that, to the 6 extent -- especially as the planning for this evolves 7 and there's communications -- Ray, with his periodics, 8 with the Commissioners -- and we have, Ray and 9 Stephanie, as do others, we have a lot of coordination 10 discussions with DOE to try to make sure that the 11 decision-makers on both sides are aware of the plans, 12 the needs, what it takes going forward.

13 So I think, to your point, we are trying 14 to make the needs and what it's going to take to get 15 us there clear to everybody involved.

16 VICE CHAIR REMPE: Thank you. And I think 17 now it's time for Ray to do his last --

18 (Simultaneous speaking.)

19 MR. FURSTENAU: I'll just wrap it up 20 really quickly here. Chairman and Vice Chair, I 21 really appreciate you taking time out of the busy ACRS 22 schedule to allow us to come in and make these 23 informational briefings. I hope we plan to continue 24 that in the future on different topics. These are 25 some potential ones, but we can add further NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

275 1 discussions on what others might be appropriate.

2 Just the conversation and the discussion 3 and the questions we get are just valuable to us to 4 kind of think about -- causes us to think more where 5 we may need to do better or do more or do less. So, 6 thanks for your support of us on the briefing.

7 VICE CHAIR REMPE: And I need to thank you 8 and your staff for taking time and to prepare the 9 presentation and give it to us, because it helps us to 10 accomplish what we're supposed to be doing as an 11 Advisory Committee.

12 Colleagues, do you have any last minute 13 comments before I open the public line?

14 MEMBER BALLINGER: Yeah, this is Ron.

15 Ray, I'll put you on the spot. Is the NRC a paid, 16 active member of the ICG-EAC group?

17 MR. FURSTENAU: You did put me on the 18 spot, Ron.

19 MEMBER BALLINGER: It was intentional.

20 MR. FURSTENAU: Okay. Is there any of my 21 staff on the line that can help me on that?

22 Otherwise, we'll take it as a look-up, Ron.

23 MR. JUDGE: I can tell you that, yes, they 24 are, speaking as an ICG member.

25 MR. FURSTENAU: Okay, good. Sorry, Ron, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

276 1 you got me. I'll be better prepared next time.

2 MEMBER BALLINGER: No, you're fine.

3 MR. FURSTENAU: Okay.

4 VICE CHAIR REMPE: Tom, can you verify 5 that -- oh, Dave, are you trying to say something?

6 You're flashing.

7 MEMBER PETTI: I just wanted to thank the 8 staff for the slides. Very informative, nice view of 9 the lay of the land and the challenges and the 10 opportunities going forward.

11 VICE CHAIR REMPE: Okay. Let's open up 12 the public line.

13 MR. DASHIELL: The public line is open for 14 comment.

15 VICE CHAIR REMPE: So, are there any 16 comments from members of the public? Please state 17 your name and provide your comment.

18 (Pause.)

19 VICE CHAIR REMPE: I think we've given 20 then the ten-second rule, right? So, let's close the 21 public line.

22 And, Ray, with respect to your potential 23 topics, we'll talk and plan and do this. I also 24 wanted to remind you that we need to get going on our 25 biannual review. And so we'll be talking to you and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

277 1 your staff about that, too, okay?

2 MR. FURSTENAU: Okay. All right, sounds 3 good.

4 VICE CHAIR REMPE: And thank you, again, 5 very much. It was very informative and it was exactly 6 what I wanted to hear.

7 MR. FURSTENAU: Yeah, my staff did all the 8 work, Joy. It was a pleasure, so, thanks.

9 VICE CHAIR REMPE: Okay. And then I'll 10 turn it back to you, Mr. Chairman. We're only a 11 couple of minutes late.

12 CHAIR SUNSERI: Thank you, Joy. And let 13 me extend my appreciation to Ray and your team for the 14 thorough and comprehensive and informative 15 presentation. It's clear that you all are staying on 16 top of this matter and applying a lot of details. So, 17 appreciate that.

18 Okay. Members, you will thank me tomorrow 19 for what I'm going to do right now, but we're going to 20 roll right into the GEH containment letter. I'll give 21 five minutes or so, time for you to take a break, as 22 we transition to get the letter report up.

23 Jose March-Leuba, the lead member for 24 this, is prepared to go, we've got other people 25 standing by, so we need to get this done. The goal NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

278 1 will be to read in the letter, have high level 2 comments, and we should be able to end by around 6:00 3 per our schedule.

4 So, let's start that transition right now.

5 (Whereupon, the above-entitled matter went 6 off the record at 5:33 p.m.)

7 8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.

(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433

Post-Halden Research Strategies &

Activities in Fuels and Structural Materials A briefing from the Office of Nuclear Regulatory Research to the Advisory Committee for Reactor Safeguards February 4, 2021

Agenda

• Opening Remarks - R. Furstenau

• Post-Halden Research Strategies & Activities in Fuels - M. Bales, L. Kyriazidis, J. Corson

• Post-Halden Research Strategies & Activities in Structural Materials - E. Focht, M. Hiser

• Vision for the Future - L. Lund

• Closing Remarks - R. Furstenau 2

Agenda

• Opening Remarks - R. Furstenau

• Post-Halden Research Strategies & Activities in Fuels - M. Bales, L. Kyriazidis, J. Corson

• Post-Halden Research Strategies & Activities in Structural Materials - E. Focht, M. Hiser

• Vision for the Future - L. Lund

• Closing Remarks - R. Furstenau 3

Motivation & Scope

• Motivation: Replace the general RES quality reviews with briefings on targeted technical subjects

• Scope: Post-Halden plans for the Fuel and Materials research areas. The Man-Technology-Organization area, including human factors and digital I&C, was not impacted by the reactor closure and will continue as a stand-alone OECD joint project.

4

Closure of Halden Reactor caused swift, significant and broad response around the world 2011-2018 Funding challenges, significant reduction in revenue from bilateral projects March 2018 Costly maintenance work identified, requiring prolonged shutdown Spring 2018 NRC, DOE, EPRI, and fuel vendors, expressed support for continued operation of the Halden reactor into the mid-2020s to support development of ATF May 2018

• US-Norway Government Mtg

• NEA-Norway Government Mtg

• NRC, DOE, EPRI, NEI Mtg

• DOE seeks impact assessment from ATF fuel vendors June 2018 Halden Reactor announces permanent closure July 2018 INL hosts Halden Capability Gap Assessment Workshop with participation from U.S. national laboratories, Halden, DOE, NRC, NEI, EPRI, NEA, SCK-CEN, NRG, MIT, and industry teams from Westinghouse, GA, GE, Framatome, and Lightbridge October 2018 NEA hosts Workshop on "Building Multinational Material and Testing Capacities for Science, Safety and Industry December 2018 INL issues Post-Halden Reactor Irradiation Testing for ATF: Final Recommendations March 2019 5 NEA NEA hosts Introduction to the Multinational Framework for In-Pile Fuel and Materials Testing

A lot has happened in the two years since the Halden Reactor shutdown

• Innovative solutions, new ideas and collaboration have been abundant in the two years since the Halden Reactor shutdown

• NRC staff have been in a leadership role in many of the initiatives

• Presentations from the staff will provide exciting details of whats been accomplished to date including:

- Archiving of Halden data

- New collaborative research projects

- Retrieval of valuable Halden test materials

- New domestic experimental capabilities 6

The new fuel and materials landscape offers many benefits, but some challenges remain

• Benefit: The dispersed network of research facilities is overall positive, leaving the nuclear fuel and materials community less vulnerable to a single future facility closure

• Benefit: The investment in US facilities will provide significant autonomy and value for the US nuclear community

• Challenge: Cost of multiple research initiatives will likely be higher than Halden Reactor Project for the near future

• Presentations from the staff will elaborate on these benefits and challenges 7

The closure of the Halden Reactor isnt the only factor driving change in fuel and materials research

• Fuels and materials testing is complex and expensive - its becoming increasingly difficult for NRC to sponsor irradiated research projects alone, we need partners and collaborations

• DOEs NEUP and NSUF programs are offering opportunities for NRC to easily partner with DOE Labs and Universities for meaningful, timely research

• The Integrated University Program (IUP), offers another mechanism to align university research with NRC emerging research needs

• As the industry pursues innovative fuel and materials, to improve economics in the LWR fleet and design advanced reactors, there is a growing demand to optimize the number and types of experiments needed to qualify fuel and material

• Presentations from the staff will elaborate on our vision for the future and explain why the future calls for more creativity and more partnerships 8

Agenda

• Opening Remarks - R. Furstenau

• Post-Halden Research Strategies & Activities in Fuels - M. Bales, L. Kyriazidis, J. Corson

• Post-Halden Research Strategies & Activities in Structural Materials - E. Focht, M. Hiser

• Vision for the Future - L. Lund

• Closing Remarks - R. Furstenau 9

A lot has happened in the two years since the Halden Reactor shutdown

• DOE has made significant investments to address Halden capability gap

• NEA has launched the FIDES Framework, which is designed to connect a network of experimental reactors and re-establish the community formed via Halden research

• Studsvik Nuclear Laboratory has proposed a small program to recover valuable fuel specimens from the Halden Reactor 10

DOE has made significant investments to address Halden Capability gap Estimated DOE Investment to Address Halden Closure Blue is spent in previous FYs Yellow are estimates, subject to Congressional appropriations 11

DOE has made significant investments to address Halden Capability gap (continued)

Establish additional LWR test loops in ATR

- Design of I-Loops underway, loop operating is expected to begin 2023 Establish LOCA testing capability at TREAT

- Commissioning tests for a Transient Water Irradiation System for TREAT (TWIST) planned for 2022 12

DOE has made significant investments to address Halden Capability gap (continued)

Establish refabrication capabilities

- Remote welding and instrumentation installation Develop advanced in-pile instrumentation INL Developed Remote End Welding System for capabilities fuel rod refabrication, currently undergoing out In-pile temperatures data from temperature of cell testing. measurement during RIA Transient Test at TREAT.

- Replicating Haldens capabilities and pushing further Above: Fuel pin from recently completed RIA transient test at INL which included advanced instrumentation (LVDTs, boiling detectors, TCs, Optical Pyrometry, Fuel Motion Monitoring System) 13

Lessons Learned on capturing experimental metadata

• The closure of the Halden Reactor clarified: we need to be more deliberate about capturing the metadata of nuclear fuel and materials research

- Metadata includes experimental design details and key parameters

- These are critical to putting the results in appropriate context for use by regulators and industry

• This lesson is being applied in future research programs

- New research proposals are emphasizing the need to capture data (and metadata) and make them accessible for long-term use 14

FIDES is designed to connect a network of experimental reactors and retain the Halden research community 15

FIDES JEEP High-burnup Experiments in Reactivity Initiated Accidents (HERA)

NRC Core Group Member In-Pile RIA Testing Database as a function of Pulse Width.

TREAT Reactor United States 16

FIDES JEEP Power to Melt and Maneuverability (P2M)

Schematic of the envelope BR2 Reactor configuration of the slow transient Belgium that the Project proposes to investigate 17

FIDES JEEP In-pile Creep Studies of ATF Claddings (INCA)

Melodie sample holder LVR-15 Reactor Czech Republic 18

Transferring the valuable fuel and material library at Halden to other research facilities has been a priority for the entire nuclear community Studsvik Nuclear Laboratory has proposed project SPARE to fund the transport of the most valuable fuel specimens from Norway to Sweden 19

The dispersed network of research facilities is an overall positive, but also presents some challenges

• Less vulnerable to facility closure

• The investment in US facilities will provide significant autonomy and value for the US nuclear community

• Halden offered significant economy of scale that will be hard to replicate - a balance between cost &

testing capacity will have to be considered in the dispersed network

• Long-term, steady-state irradiation of instrumented fuel and materials is difficult to replace 20

Agenda

• Opening Remarks - R. Furstenau

• Post-Halden Research Strategies & Activities in Fuels - M. Bales, L. Kyriazidis, J. Corson

• Post-Halden Research Strategies & Activities in Structural Materials - E. Focht, M. Hiser

• Vision for the Future - L. Lund

• Closing Remarks - R. Furstenau 21

Halden was a key aspect of NRC s irradiated materials research prior to reactor shutdown

• Unique Capabilities

- Multiple in-reactor testing loops dedicated to materials research

• In-pile stress corrosion crack growth rate (SCC CGR) testing

• In-pile instrumented creep / stress relaxation testing

• Excellent Value / Leveraging

- Halden funding split with other countries, US organizations, and other NRC research areas made research very cost-effective

• Specific Planned Materials Research Activities

- Further irradiation and testing of ex-plant stainless steel (SS) welds

- SCC CGR testing of high dose ex-plant SS materials

- Creep and stress relaxation testing of baffle-former bolt (BFB) materials 22

NRCs post-Halden materials research strategy emphasizes cooperation

• Following the Halden reactor shutdown, NRC staff assessed the impacts to NRC interests

• Focused on addressing needs and mitigating impacts by:

- Preserving knowledge from Halden Halden 2018 - 2023 / DOE

- Performing high-priority research Halden 2018 - 2023, FIDES JEEPs, harvesting

- Developing new capabilities FIDES JEEPs / DOE 23

Halden Materials Research 2018-2023

• Halden activities post-reactor shutdown focused on:

- Orderly completion of ongoing work

- Preserving fuels and materials samples of research value

- Documentation and capture of past Halden data

• Ongoing experimental work is nearly complete

- Included high-priority NRC interest of SCC CGR testing of high dose ex-plant SS materials

• DOE is expected to acquire high-value irradiated materials samples from Halden

- Will be available to the research community through the existing Nuclear Science User Facilities (NSUF) process

• Halden focused on capturing historic data and knowledge by developing a legacy database 24

Proposed FIDES Structural Materials JEEP INL Proposal (under development)

Stress-Relaxation and Creep Research (STRETCH) JEEP Obtain irradiation-induced stress relaxation and creep data for baffle former bolting materials Testing previously planned under Halden program Leverages DOE projects at INL and INL-MITR working relationship MIT Reactor United States 25

Proposed FIDES Structural Materials JEEP INL Proposal (under development)

Development of In-Core Mechanical Testing Capabilities (ENCORE) JEEP Develop in-core mechanical testing capabilities at INL Advanced Test Reactor (ATR)

Long-term solution to retain and expand lost Halden capabilities SCC crack growth rate testing in LWR and ANLWR environments ATR United States 26

Ex-plant materials harvesting is important for understanding long-term aging

• Harvesting provides highly representative aged materials for research

- Particularly valuable for irradiated material research

• NRC staff has developed harvesting priorities, identified opportunities, and engaged partners

- International harvesting workshops held at NRC HQ in March 2017 and OECD/NEA HQ in January 2020

• Key irradiated materials harvesting priorities for NRC:

- High fluence (>2 dpa) SS welds

- High fluence (>30 dpa) / high temperature SS materials

- Irradiated cast austenitic SS materials 27

SMILE will address key long-term aging issues for metallic LWR components

• Studsvik Materials Integrity for Life Extension (SMILE) project is a planned international cooperation through OECD/NEA

- Structured similarly to Studsvik Cladding Integrity project (SCIP) year planned scope of work beginning in 2021

• Harvesting, characterization and testing of metallic components from decommissioning Swedish BWR and PWR

- Reactor pressure vessel low alloy steel

- SS internals: core shroud / barrel welds, baffle plate, BFBs

- Nickel alloy penetrations, piping, and steam generator tubes

• Reactors:

- Ringhals 2: Westinghouse 3-loop PWR with ~30 EFPY

- Oskarshamn 2: ABB-Atom BWR with ~30 EFPY 28

SMILE Irradiated Materials Research

• SMILE addresses multiple high-priority NRC interests:

- Very high dose SS plate and BFB materials

- High dose SS welds

• Scope includes:

- Microscopy, retrospective dosimetry, H/He content

- Tensile, IASCC CGR, and fracture toughness (FT) testing 29

Materials research strategy focuses on supporting regulatory decision-making

• Preserving knowledge from Halden

• Performing high-priority research

• Developing new capabilities

• Long-term testing of highly representative materials 30

Agenda

• Opening Remarks - R. Furstenau

• Post-Halden Research Strategies & Activities in Fuels - M. Bales, L. Kyriazidis, J. Corson

• Post-Halden Research Strategies & Activities in Structural Materials - E. Focht, M. Hiser

• Vision for the Future - L. Lund

• Closing Remarks - R. Furstenau 31

Vision for the Future Closure of the Halden Reactor The closure of the Halden Diversifying, partnering, and has accelerated strategic Reactor isnt the only factor enhancing the U.S. infrastructure planning to address our future driving change in fuel and are the tenets of NRCs post-fuel and materials research materials research Halden strategy needs Economic trends are driving change

• Preserving the knowledge gained during

• Growing need to pursue more focused, the Halden program innovative, and aggressive campaigns,

• Utilizing array of ventures to address i.e., Accelerated Fuel Qualification our short-term research needs

• Developing new experimental The future will require more partnerships capabilities to address our long-term needs

• Domestically: NEUP, NSUF and IUP Grant Programs

• Significant investments in U.S.

• Internationally: FIDES, QUENCH-ATF, infrastructure made by DOE are critical SMILE and SPARE to meeting future needs

Agenda

• Opening Remarks - R. Furstenau

• Post-Halden Research Strategies & Activities in Fuels - M. Bales, L. Kyriazidis, J. Corson

• Post-Halden Research Strategies & Activities in Structural Materials - E. Focht, M. Hiser

• Vision for the Future - L. Lund

• Closing Remarks - R. Furstenau 33

Potential Topics for Future Meetings

• Updates on the Future-Focused Research Initiative

• Updates on Integrated University Program grants for mission-related R&D

• Non-LWR research activities 34

Closing Remarks 35

Oral Comments by Mr. John Stetkar on IDHEAS Agenda Item 6 at the 682nd ACRS Meeting February 4, 2021

1. My name is John Stetkar. I am a former member of the ACRS, speaking today as a member of the public.

2. I would first like to comment briefly on the IDHEAS general methodology in NUREG-2198. I understand that the Committee has received a copy of my September 24, 2020 detailed written comments on the previous draft version of the report. The current version of the report has addressed the majority of those comments. I very much appreciate the staff's stamina and their extensive efforts to consider my comments and make those changes.

3. I have a few remaining high-level concerns about NUREG-2198. Considering the brief time available for these oral comments, I will highlight only a couple of my most important issues.

• First, I remain concerned about the lack of technical justification for the quantification model that is represented by Equation 4.6 and Equation 4.7. In particular, I do not know why the primary influence on human performance is determined by three specific "base" performance-influencing factors, while the other 17 factors are cumulative modifiers. I also do not understand why the assumed linear summation of those performance-influencing factor weights is justified. I understand that the Committee has also received a copy of my November 2, 2020 detailed written comments on the IDHEAS-DATA report. I could not find compelling justification in that report for the format of the quantification model. Furthermore, Section 6.3 and Appendix D in the current version of NUREG-2198 have removed examples which were originally intended to support the conclusion that linear addition of the performance-influencing factor weights provides the best method to account for their composite effects. The current version of NUREG-2198 relies primarily on qualitative assertions that the model is justified from reviews of other studies, without quantitative examples that clearly support that justification.

• Second, the examples in Appendix M of NUREG-2198 are very important for prospective analysts to understand how the methodology is applied in practice. I have two comments on those examples.

• My first comment is related to documentation of the analysts' decisions that are made in each example. One of the most important objectives of the IDHEAS methodology is to reduce sources of analyst-to-analyst variability in the human reliability analyses. As noted on Slide 6 of Dr. Xing's presentation, a critical task to achieve that objective is clear documentation of the rationale and justification for numerous analyst decisions that are made throughout the analyses. The examples in Appendix M only partially document those decisions. For example, they simply list the specific cognitive failure modes (CFMs) and performance-influencing factors (PIFs) that the analysts selected for each evaluation. They do not document why other CFMs and PIFs were excluded as not relevant. It is very important to document the rationale for those exclusions, because it can often reveal key sources of disagreement between different analysts and the reasons for that disagreement. In practice, the documentation process by itself also often prompts each analyst to more carefully consider the basis for their own judgment and selections. The same comment applies to lack of a documented rationale for selection of a particular form of the uncertainty distributions for the time estimates and

the assignments of specific parameters in those distributions (for example, the 5th and 95th percentiles). So, in summary, to provide instructive examples that demonstrate the expectations of how the IDHEAS methodology should be implemented, I think that the examples in Appendix M should better document the analysts' rationale for their decisions.

• My second comment on Appendix M is that the quantification example in Section M.2.6 was revised, but it is not yet correct. The combined uncertainty distribution in that example evaluates the time at which power is restored from the FLEX generator. That distribution shows that there is a small, but non-zero, probability that power is restored before the crew begins to use the extended loss of AC power (ELAP) guidance. In other words, there is some probability that power is restored before 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> in that scenario.

That is not realistic, and the example should be corrected. Perhaps the staff will need to use other forms of the uncertainty distributions to provide realistic estimates that are consistent with the physical constraints of the scenario. This comment is not intended to be an obsessively detailed critique of probabilistic arithmetic. I think that it is conceptually important for the methodology, because the examples should clearly demonstrate to prospective analysts that the uncertainty quantification is not an abstract mathematical after-thought. The uncertainties should be an integral part of the analysis.

They should account for a realistic engineering and operational assessment of personnel performance within the physical and functional constraints of the event scenario. If the uncertainty analysis results provide non-physical conclusions, then something is drastically wrong with those analyses.

4. Finally, if the Committee will indulge me, I would simply like to emphasize the fact that RIL-2020-13 is the product of several key methods and reports that deserve careful attention before conclusions are made about the example FLEX analyses. The relationships among the IDHEAS general methodology (NUREG-2198), the IDHEAS-DATA report, the IDHEAS-ECA application (RIL-2020-02), and finally the use of IDHEAS-ECA to evaluate FLEX actions are shown on Slides 7 and 8 of Dr. Xing's presentation. Those relationships are complex and somewhat convoluted. Thus, until the Committee has an opportunity to carefully examine the IDHEAS-DATA and IDHEAS-ECA reports, specific conclusions or recommendations about RIL-2020-13 may be premature.

The Integrated Human Event Analysis System (IDHEAS) Program Introduction Sean E. Peters Advisory Committee on Reactor Safeguards February 4, 2021

Why are we here?

SRM-M061020 The Committee should work with the staff and external stakeholders to evaluate the different Human Reliability models in an effort to propose either a single model for the agency to use or guidance on which model(s) should to be used in specific circumstances.

2

Timeline of HRA Development PRA Policy IDHEAS At-Power Evaluation of Statement HRA Methods HRA Cognitive Basis HRA Good vs Good International HRA U.S. Empirical Study Practices Practices Empirical Study 2011 2020 1995 2005 2006 2009-2014 2016 2017 SRM M061020 Fukushima IDHEAS-G Daiichi IDHEAS-ECA FLEX Expert Elicitation FLEX HRA 3

Timeline References

• PRA Policy Statement (60 FR 42622)

• NUREG-1792 - Good Practices for Implementing [HRA]

(ML051160213)

• NUREG-1842 - Evaluation of [HRA] Methods Against Good Practices (ML063200058)

• NUREG/IA-0216 - International HRA Empirical Study (ML093380283, ML11250A010, ML14358A254)

• NUREG-2127 - The International HRA Empirical Study: Lessons Learned from Comparing HRA Methods Predictions to HAMMLAB Simulator Data (ML14227A197)

• NUREG-2156 - The U.S. HRA Empirical Study (ML16179A124) 4

Timeline References (Cont.)

• NUREG-2114 - Cognitive Basis for [HRA] (ML16014A045)

• NUREG-2199, Vol. 1 - [IDHEAS] for [NPP] Internal Events At-Power Application (ML17073A041)

• NUREG-2198 - IDHEAS - General Methodology (ML20329A428)

• RIL 2020-02, Integrated Human Event Analysis System for Event and Condition Assessment (IDHEAS-ECA) (ML20016A481)

• RIL 2020 Vols. 1 and 2 - Applying HRA to FLEX Operations -

Expert elicitation and Using IDHEAS-ECA - (ML21033A529, ML20345A318, ML21032A119) 5

IDHEAS Development Process

• US and International Benchmarking Projects -

determined existing methods strengths and weaknesses

• Cognitive Basis Report

- Extensive Literature Review, Scientific Basis for Structure

• IDHEAS at-Power

- Industry/NRC Collaboration - goal of reducing variability

• Fukushima Event - March 2011 6

Development Process (cont.)

• IDHEAS-G

- Guidance for developing application-specific HRA methods or tools

- Framework to generalize and integrate human error data

- Structure to analyze human events and identify human failures and root causes

• IDHEAS-ECA

- Built from IDHEAS-G to handle all NRC applications

- Can be used for in/ex control room activities and other nuclear/non-nuclear domains (human centered method)

- Quantification model and software tool included 7

Development Process (cont.)

• IDHEAS-Data

- Data basis for IDHEAS quantification

- Constantly evolving and tied to NRC data collection activities

• Scenario Authoring, Characterization, and Debriefing Application - SACADA

• NRCs Human Performance Test Facility

• Halden Reactor Project 8

IDHEAS Reviews

• IDHEAS-G

- Multiple ACRS Subcommittee reviews

- 3 external peer reviews, 2 internal peer reviews

- Used on: Fukushima, US Benchmarking Events, Fuel Cycle Facility Events

• IDHEAS-ECA

- Used on FLEX Scenarios (NRC and industry studies), ASP and SDP Events

- Currently taking user comments to incorporate into revised report/tool

• IDHEAS-DATA

- Data review (underway)

- Plans for regular updates 9

IDHEAS - An Integrated Human Event Analysis System Jing Xing, Y. James Chang, Jonathan DeJesus Segarra, U.S. Nuclear Regulatory Commission Presented by Jing Xing to ACRS Full Committee Feb-04-2021

Outline I. Overview of IDHEAS II. Introduction to IDHEAS-G, IDHEAS-ECA, and IDHEAS-DATA III. Examples of IDHEAS applications IV. Revision to IDHEAS reports after 9-23-2020 ACRS Subcommittee meeting 2

Where we were Scope

- Ex-CR actions HRA Use of Human

- Shutdown Variability Performance

- Severe accidents Data Uncertainties Analysts HRA in the scenario practices method Qualitative Qualitative to PIFs - explicit Cognitive and analysis guidance quantification description data basis 3

Where we are now Scope

- Ex-CR actions HRA Use of Human

- Shutdown Variability Performance

- Severe accidents Data Uncertainties Analysts HRA in the scenario practices methods Qualitative Qualitative to PIFs - explicit Cognitive and analysis guidance quantification description data basis 4

What we have achieved

• Expanded scope - IDHEAS is an HRA method suite for all nuclear HRA applications

• Use of human performance data - Human error data were explicitly used in IDHEAS

- The method and data structure are based on the same cognitive basis model such that data can be generalized and used by the method.

• HRA variability - IDHEAS improves HRA method variability by enhancing the four areas (identified in HRA benchmarking studies)

- Systematic qualitative analysis guidance

- Links between qualitative analysis outcomes and quantification of human error probabilities (HEPs)

- Explicit attributes for every performance influencing factor (PIF)

- Cognitive and data basis that links PIF attributes to cognitive failure modes (CFMs) 5

Other sources of HRA variability HRA Variability Uncertainties Analysts HRA in the scenario practices methods

• Uncertainties in the scenario resulting in different analysis assumptions

- IDHEAS provides guidance on identifying uncertainties in the scenario and tracing the assumptions in the HRA.

• Analysts practices resulting in different interpretations of the scenario

- The documentation of IDHEAS structured process provides transparency of analysts interpretations.

6

Development of IDHEAS

- An Integrated Human Event Analysis System Scientific Cognitive Basis for HRA (NUREG-2114)

SACADA and other Literature data sources

Research, IDHEAS General Methodology IDHEAS-operation experience (IDHEAS-G) (NUREG-2198) DATA IDHEAS Internal At- IDHEAS-ECA (RIL-2020-02) power Application (NUREG-2199)

Testing the Event FLEX HRA Evaluation SDP/ASP method analysis Expert of FLEX analysis Elicitation actions (RIL-2020-13) 7

Development of IDHEAS

- An Integrated Human Event Analysis System Scientific Cognitive Basis for HRA (NUREG-2114)

SACADA and all Literature data sources

Research, IDHEAS General Methodology IDHEAS-operation experience (IDHEAS-G) (NUREG-2198) DATA IDHEAS Internal At- IDHEAS-ECA (RIL-2020-02) power Application (NUREG-2199)

Testing the Event FLEX HRA Evaluation SDP/ASP method analysis Expert of FLEX analysis Elicitation actions (RIL-2020-13) 8

Outline I. Overview of IDHEAS II. Introduction to IDHEAS-G, IDHEAS-ECA, and IDHEAS-DATA III. Examples of IDHEAS applications IV. Revision to IDHEAS reports after 9-23-2020 ACRS Subcommittee meeting 9

What is IDHEAS-G

• A methodology for developing application-specific HRA methods

• A platform for generalizing and integrating human error data to support HEP estimation

• A general HRA method for human event analysis and human error root causal analysis 10

Overview of IDHEAS-G IDHEAS-G consists of a cognition model as the framework for HRA, its implementation in an HRA process, and detailed guidance for HRA applications.

Cognitive Basis Structure Cognition Model PIF Structure Stage 1 Stage 2 Stage 3 Stage 4 Scenario Modeling of HEP Integrative analysis important quantification analysis human actions 11

Cognitive Basis Structure Macrocognitive Cognitive PIFs Processors functions mechanisms Processor - D1 Cognitive Detection mechanism Processor - D5 PIF 1 Cognitive Processor - U1 mechanism PIF 2 Understanding Processor - U5 Cognitive PIF 3 mechanism Human Decision- Processor - DM1 task Cognitive making Processor - DM6 PIF 17 mechanism Action Processor - E1 PIF 18 Cognitive execution mechanism Processor - E5 PIF 19 Interteam Processor - T1 Cognitive coordination mechanism Processor - T7 12

PIF Structure Environment Personnel and System Task Context and Situation organization

• Information

• Staffing availability and

• Accessibility/habitabi *Procedures, reliability

• System and I&C lity of workplace and guidance, and *Scenario familiarity transparency to PIF travel paths personnel instructions *Multitasking,

• Workplace visibility *Training interruptions, and

• Human system

• Workplace Noise *Team and distractions interface Cold/heat/humidity organizational *Task complexity

• Equipment and

• Resistance to physical factors *Mental fatigue tools movement *Work processes *Time pressure and stress

• Physical demands PIF *Poor lighting in

• Tools are difficult to

• Procedure is inadequate use *Sustained high-attributes workplace *Procedure is difficult to demand cognitive

• Tools are unfamiliar to Note: The PIF *Glare or reflection use activities personnel attributes shown are on physical structure *Procedure is available, *Long working hours

• Tools do not work examples and *Smoke or fog- but does not fit the *Sleep deprivation

• Tools or parts are correspond to the PIFs induced low visibility situation highlighted in red. unavailable 13

Example PIF - Human-System Interface

• Definition: HSI refers to indications (e.g., displays, indicators, alarms) and controls for detecting information and executing actions on systems.

• Attributes:.

- The source of indication (e.g., indicators, labels) is similar to other sources nearby.

- The indications have low salience.

- Indications are confusing or nonintuitive.

- Controls are difficult to maneuver.

- Labels on the controls do not agree with document nomenclature.

- Controls are not reliable, and personnel are unaware of the problem.

14

How IDHEAS-G models human failure events (Stage-1 and Stage-2)

• 5 Macrocognitive functions, or Human actions /

• 5-7 processor for each function, or human failure tasks

• Application-specific failure modes event Event context

• 20 PIFs & Time adequacy

• PIF attributes

• Five macrocognitive functions model failure of human actions

• 20 PIFs model the context that affects human performance of an action 15

IDHEAS-G Stage 3 HEP QuantificationOverview Time available

=

Time required CFMs PIF attributes HEP = = (, , )

CFM 1 Critical

= (, , )

task 1 CFM 2 Critical = (, , )

task 2 CFM 3 Critical = (, , )

CFM 4 task 3 16

HEP QuantificationPc

• Probability of CFM, , can be estimated in one or a combination of the following three ways:

- Calculation from the number of errors divided by number of occurrences

- Expert judgment

- HEP quantification model

• IDHEAS-G provides a data structure of generalizing human error data to support the three ways.

17

IDHEAS-G Stage 4 - Integrative analysis IDHEAS Dependency Model

1. Identify the dependency context Cut set with

• Consequential dependency multiple HFEs (HFE1, HFE2)

• Resource-sharing dependency

• Cognitive dependency HFE1 and HFE2 are 2. Model the dependency context HFE2lHFE1 independent All no Any Are there changes to HFE2s:

Definition? Time required and time available?

P(HFE1,HFE2)= yes?

Feasibility? CFMs?

P(HFE1)*P(HFE2) Critical tasks? PIF attributes?

Yes

3. Calculate P(HFE2lHFE1) based on P(HFE1,HFE2) =

context changes to HFE2 and using P(HFE1)*P(HFE2lHFE1) same method as individual HFEs HFE2lHFE1 means the occurrence of event HFE2 given the occurrence of event HFE1, where HFE1 is the first event and HFE2 is the second event. 18

Summary of IDHEAS-G

• A methodology for developing application-specific HRA methods

• A platform to generalize and integrate human error data from various sources for HEP estimation

• A method to systematically analyze human events, including identification of human failures and root causes

• Applicable to all nuclear applications 19

Outline I. Overview of IDHEAS suite II. Introduction to IDHEAS-G, IDHEAS-ECA, and IDHEAS-DATA III. Examples of IDHEAS applications IV. Revision to IDHEAS reports after 9-23-2020 ACRS Subcommittee meeting 20

IDHEAS-G as a methodology for developing application-specific HRA methods

• IDHEAS-G has comprehensive sets of CFMs and PIFs, using all of those for HRA is time-consuming

• HEP quantification may be adapted for specific HRA applications IDHEAS-G vs. application-specific method IDHEAS-G Application-specific method Applicable to all nuclear Specific for the application applications Comprehensive but low usability Concise and easy to use Referencing the Human Error Data Calculating HEPs of human actions Tables 21

From IDHEAS-G to IDHEAS-ECA Developing application-specific IDHEAS method:

• Define the scope of application, requirements, and available sources for the intended use

• Keep the qualitative analysis the same as that in IDHEAS-G

• Develop application-specific sets of CFMs and PIFs

• Provide HEP estimate using one or the combination of the HEP quantification approaches and generalized human error data Define IDHEAS-ECA by NRR users:

• Scope: Perform Event and Conditions Assessment (ECA) for all NRCs risk-informed applications; specifically, be applicable for FLEX HRA

• Requirements: Easy to use, not over-burden HRA analysts

• Resources: Human error data, NRC 2018 FLEX-HRA Expert Elicitation 22

IDHEAS-ECA DELTA between IDHEAS-G and IDHEAS-ECA - modeling failures and calculating HEPs IDHEAS-G IDHEAS-ECA Qualitative analysis guidance Same as that of IDHEAS-G A basic set of CFMs in three Five high-level CFMs levels of details 20 PIFs and their attributes

• All 20 PIFs preserved

• A compressed set of PIF (combining attributes)

Three approaches to HEP

• HEP quantification model; estimate

• The base HEPs and PIF weights integrated from IDHEAS-DATA;

• Allowing HEP calculation for given failure modes and PIF attributes 23

HEP Quantification in IDHEAS-ECAPc

• HEP quantification model 1

= 1 + 1

=1 Base HEPs Recovery factor; set PIF weight factors from to 1 unless data Modification PIFs suggest otherwise

=

PIF interaction error rate at a given PIF attribute factor; set to 1 with error rate when the PIF linear combination attribute has no or low impact IDHEAS-ECA needs:

• Lowest HEPs for the 5 CFMs

• Base HEPs of every CFM for every associated attribute of the 3 base PIFs

• PIF weights of every CFM for every associated attribute of the 17 modification 24 PIFs

IDHEAS-ECA Processsame as IDHEAS-G PIF attributes of every CFM Scenario for every CT PRA context and Calculate model list of (3.3.2) applicable Assess attributes of Analyze scenario context PIFs every applicable PIF (3.1.2)

(3.2.3)

List of Develop scenario narrative HFE and its applicable Calculate Develop scenario timeline definition CFM(s) for overall HEP (3.1.1) the CT(s) (3.3)

HFE and its List of definition Break down CT(s) Characterize the CT(s) and Identify and define HFE (3.1.3) HFE into CT(s) select applicable CFMs (3.2.1) (3.2.1 and 3.2.2)

HFE and its definition Estimate parameters and of distribution Analyze HFE timeline (subset of scenario timeline, if there (3.3.1)

Calculate are multiple HFEs in the scenario) Estimate parameters (3.3.1) of distribution (3.3.1) and Uncertainty and dependency analysis

= time required and documentation

= time available CFM = cognitive failure mode PIF = performance-influencing factor and = mean and standard deviation of CT = critical task PRA = probabilistic risk assessment

= error probability due to CFMs and = mean and standard deviation of HEP = human error probability 25 HFE = human failure event = error probability due variability in and

IDHEAS-ECA Products

• IDHEAS-ECA report - including guidance, worksheets, base HEPs and PIF weights, and three HRA examples

• IDHEAS-ECA training materials

• IDHEAS-ECA Software - A computer interface implementing IDHEAS-ECA for HEP calculation

- Recommended use:

• Analyze the event and document the results in IDHEAS-ECA worksheets

• Enter the information from the Worksheets to calculate the HEP 26

IDHEAS-ECA Software 27

Outline I. Overview of IDHEAS suite II. Introduction to IDHEAS-G, IDHEAS-ECA, and IDHEAS-DATA III. Examples of IDHEAS applications IV. Revision to IDHEAS reports after 9-23-2020 ACRS Subcommittee meeting 28

Generalizing human error data to inform HEPs HEP = f(states of performance influencing factors)

• Human error data exist from various domains, in different formats, varying context and levels of detail.

Data source 1 Data source 2 Tasks Context Tasks Context Failure PIFs Failure PIFs modes modes A generic, adaptable set of failure modes and PIFs 29

Use human error data to inform HEPs

1. Evaluation - 2. Generalization - 3. Integration -

Assess data Represent source data Integrate the source with the CFMs and PIFs generalized data for HEP calculation

• Context and Human tasks -> Error rates - Base HEPs task Cognitive failure modes (CFMs) Change of error rates -

• Variables and PIF weights ( )

Measurements Context ->

Performance Others (e.g., PIF

• Uncertainties influencing interaction, time factors (PIFs) distribution, dependency) 30

II. Data sources A. Nuclear simulator data and operational data (e.g., SACADA, HuREX, German NPP maintenance database analysis)

B. Operation performance data from other domains (e.g., transportation, off-shore oil, military operations, manufacturing)

C. Experimental studies in the literature (e.g., cognitive and behavior science, human factors, neuroscience)

D. Expert judgment of human reliability in the nuclear domain E. Unspecified context (e.g., statistical data, ranking, frequencies of errors or causal analysis) 31

Data source evaluation

• Participants - Normal adults, trained for the tasks, good sample size

• Measurements - Human error rates preferred, task performance measures related to human error rates

• Specificity - CFMs and PIFs identifiable

• Uncertainties - Controlled, known, or traceable

• Breath of representation - Repetitive and representative 32

IDHEAS-DATA Structure

• IDHEAS-DATA has 27 tables (IDTABLEs) documenting generalized human error data and empirical evidence

• Human error data are generalized to IDHEAS-G CFMs and PIF attributes IDHEAS-DATA IDTABLE IDTABLE 1-3 Base HEPs IDTABLE-21 Lowest HEPs of CFMs IDTABLE-1 Scenario Familiarity IDTABLE-22 PIF Interaction IDTABLE-2 Information IDTABLE-23 Distribution of Task Needed IDTABLE-3 Task Complexity IDTABLE-24 Modification to Time Needed IDTABLE 4--20 PIF Weights IDTABLE-25 Dependency of Human IDTABLE 4-8 Environment PIFs Actions IDTABLE 9-11 System PIFs IDTABLE-26 Recovery of Human Actions IDTABLE 11-16 Personnel PIFs IDTABLE-27 Main drivers to human events IDTABLE 17-20 Task PIFs 33 33

Summary of IDHEAS-DATA By 2020:

• Use of nuclear operation/simulation data (SACADA, HuREX, Halden studies)

• ~300+ literature generalized; another 200+ evaluated and selected for generalization

• The generalized data were independently verified and reviewed.

In the future:

• Human error data needed in teamwork and organizational factors

• Data generalization is an on-going, continuous effort; Data integration should be periodically updated.

34

Outline I. Overview of IDHEAS suite II. Introduction to IDHEAS-G, IDHEAS-ECA, and IDHEAS-DATA III. IDHEAS applications

- 2018 FLEX HRA Expert Elicitation

- 2019 FLEX HRA Evaluation using IDHEAS-ECA

- Several SDP analysis

- Industrys comparison

- IDHEAS Dependency Group applying IDHEAS-ECA to SDP/ASP events and PRA models IV. Revision to IDHEAS reports after 9-23-2020 ACRS Subcommittee meeting 35

Objectives of 2018 FLEX-HRA Expert Elicitation

1. Quantify the HEPs of representative FLEX action

- for using FLEX equipment during FLEX-designed scenarios and for added defense-in-depth during non-FLEX-designed applications

2. Evaluate the unique performance influencing factors (PIFs) associated with the use of FLEX equipment

3. Quantify the contribution of these PIFs on the HEPs 36

FLEX-HRA Expert Elicitation Process

• Sponsor: NRC

• Process: Implement the principles and 10-step process in the NRCs White Paper Expert Elicitation Guidance o Extensive datasets disseminated on HEPs of surrogate human actions and effects of performance influencing factors o Five tele-meetings, one face-to-face workshop

• Expert panel: Three NRC staff and three industry experts who are knowledgeable in PRA/HRA, implementation / audits of FLEX strategies, and maintenance practices at nuclear power plants.

37

Estimate the HEPs of representative FLEX actions Estimate HEPs of representative FLEX actions in two scenarios:

i) a non-FLEX-designed scenario (one EDG is down followed by SBO in a design-basis accident), and ii) a FLEX-designed scenario (SBO caused by a severe external event -

strong wind and flooding)

Action 1: Use of portable generators Action 2: Use of portable pumps Action 3: Refilling water storage tanks using alternate water sources Action 4: ELAP declaration Action 5: Deep DC load shed 38

Scenario definition and context Scenario context is characterized with IDHEAS-G performance influencing factors.

Environment and System Personnel and Tasks situation organization

- Accessibility - Information - Training - Scenario

- Visibility - Tools and parts - Procedure familiarity

- Cold, heat, and - Human- - Teamwork - Multitasking humidity system- factors - Task complexity interfaces - Mental fatigue (indications & and stress controls) - Physical demands 39

Characterization of scenario context Example: environment context Non-FLEX-designed scenario FLEX-designed scenario No impact - no weather Moderate impact

• Normal day

• Visibility - Poor lighting (e.g., darkness,

• Water - May be loss of fog, smoke, dust) upstream dam bringing debris

• Water level - water in some work places into contact with plant or travel paths in water (1-3 feet)

• Cold

• Wind - Strong winds that would focus debris to the intake structure.

• Difficult to access some sites or travel paths

• Very cold 40 16

Result - HEPs for Declaration of ELAP

• Non-FLEX-designed scenario - Non-hazard SBO leads to ELAP declaration

• FLEX-designed scenario - Strong wind and flooding result in SBO

• The action is declaring ELAP by 60mins if power is not back within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.

Non-FLEX-designed scenario FLEX-designed scenario 1th 50th 99th 1th 50th 99th 0.03 0.31 0.57 0.02 0.19 0.48 Justifications:

• Information incomplete and uncertain - I dont have info yet, I need more info to make decision.

• More preferred alternative exists - Restoring the power instead of going to ELAP

• FLEX-designed scenario has fewer uncertainties, thus easier for the decision 41

HEP variability due to uncertainties in the scenarios Example: The action Load Shed (open 18 breakers in two locations) in the non-FLEX-designed scenario Uncertainties in the scenarios:

• Layout and labels of the breakers

• Who does the work

• Travel path to the breakers

• Effect of stress Experts judgments of Load Shed:

Expert 1th 50th 99th Justifications A 0.06 0.2 0.4 High stress, variation in ergonomics, unfamiliar B 0.01 0.1 0.3 C 0 .01 0.05 0.1 Similar to the actions operators perform routinely, stress should have no impact D 0.015 0.04 0.1 E 0.001 0.01 0.1 A simple action modeled in SPAR-H, poor lighting, some stress impact 42

Insights from 2018 FLEX HRA expert elicitation

• The expert judgments captured the technical communitys state-of-knowledge about uncertainties, challenges, and opportunities in FLEX human actions.

• The estimated HEPs are valid only for the assumptions and specifications made for the scenarios and actions in this study.

• IDHEAS-G PIF structure was capable of modeling the context of using FLEX equipment in FLEX-designed and non-FLEX-designed scenarios.

• The human error data in the information package were helpful for HEP estimation and should be used in the IDHEAS method.

43

2019 FLEX HRA Evaluation using IDHEAS-ECA

- Evaluate several representative FLEX actions using IDHEAS-ECA

- Provide feedback for improving IDHEAS-ECA

- Led by NRC (RES and NRR) and the industry (EPRI). EPRIs involvement facilitated participation of industry as FLEX and operations experts, HRA experts, and hosts for two plant site visits.

- The plant visits were the predominant sources of detailed HRA-relevant FLEX information for the HRA analysts to reference.

- Information from a small group of PWR Owners Group and BWR Owners Group representatives, and FLEX experts (both NRC and industry) supplemented the plant-specific information to provide a more generic operational understanding of FLEX strategies and equipment analyzed in the scenarios.

44

Overview

• Teams of FLEX and human reliability analysis (HRA) experts modeled a sample of FLEX human actions.

• Both teams had plant site visits to better understand FLEX strategies, associated equipment and operator actions;

• The FLEX experts created a set of realistic scenarios and human failure event (HFE) descriptions using FLEX equipment.

• The HRA experts further modified the scenarios then quantified the HFEs using the IDHEAS-ECA quantification tool.

• The HRA experts participated in a 3-day workshop to perform and/or finalize their HRA quantification using IDHEAS-ECA.

45

Scenarios and HFEs Evaluated Three scenarios, four FLEX human actions

1. Beyond design basis (BDB) seismic event at a BWR that results in a Station Blackout (SBO)

- Operators fail to declare extended loss of alternating current (AC) power (ELAP) with variations

- Operators fail to perform FLEX direct current (DC) load shed

- Operators fail to deploy FLEX diesel generator

- Operators fail to perform containment venting

2. Loss of all feedwater with deployment of FLEX pump

3. SBO with pre-staged FLEX plus diesel generator 46

IDHEAS-ECA HEP Results Scenario

Description:

A BDB seismic event occurs that causes an SBO. It is obvious that power cannot be restored quickly.

HFE Critical Task HEP Estimate Case 1: Definitive Wording 1.1E-3 to 2.7E-3 Case 2: Wording Requires Judgment 1.1E-3 to 3E-2 Fail to Declare ELAP Case 3: Wording Requires Judgement and Diagnosis 1.6E-2 to 1E-1 is not Obvious Fail to Perform FLEX DC Load Shed 2E-3 to 6E-3 Fail to Deploy FLEX Transport Diesel Generator 1E-3 to 3E-3 Diesel Generator Connect and Start Diesel Generator 1E-3 to 1.2E-2 Case 1: IF AC power cannot be restored within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, declare ELAP within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of losing all AC power.

Case 2: IF AC power cannot be restored within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, declare ELAP within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of losing all AC power.

Case 3: Same as Case 2 except that it is less obvious that power cannot be restored.

47

HRA variability - Declare ELAP Case 2 Ana- CFM PIF attributes Justification HEP lyst A DM Information 3E-2 Completeness and Reliability - Information is unreliable or uncertain

(**INF2; Level 2)

B DM Info Completeness and INF2 level would vary depending on details 3E-2 Reliability - Information of what the procedure guidance would say.

is unreliable or uncertain Level would range from 3 to 5 given the

(**INF2; Level 2) example presented to the team.

C DM No Impact; No PIF (Operators understand that during 1E-3 selection evaluation of loss of all AC power, the 1-hour time frame is set in stone and cannot be deviated from due to the importance of getting a FLEX generator deployed and started.)

D D No impact: (A decision must be made. The cue is 1.1E-3 DM No impact: based on the expectation that AC power to any 4.16 kV bus cannot be restored)

E48 U No impact: 2E-3 DM No impact:

Insights from 2019 FLEX HRA evaluation using IDHEAS-ECA

• HRA variability -

o Analyst-to-analyst variability is within an order of magnitude for most human actions o Analyst-to-analyst variability remains a concern when using IDHEAS-ECA

- due to uncertainties in the scenarios and variation in HRA practices o Method traceability supports understanding and reconciling variability

• HRA informs plant risk understanding and mitigation strategies -

o Procedural cues for using FLEX equipment in non-FLEX scenarios are important for crediting FLEX in PRAs.

o The integration of FLEX into plant accident response has improved substantially since FLEX was initially implemented.

49

Preliminary user feedback on IDHEAS-ECA

- from applying IDHEAS-ECA to SDP, ASP, and basic PRA models.

• In general, results from IDHEAS-ECA are aligned with HRA analysts perceptions

• IDHEAS-ECA explicitly addresses the effects of subtle factors in SDP analyses, such as group-thinking and crediting the supplemental cues that occurred later in scenarios.

• It can be difficult assessing applicable PIF attributes with limited information in basic PRA models.

50

Final summary

- What IDHEAS has achieved

• Built on cognitive science.

• Generalized and integrated human error data for HEP estimation.

• Reduced HRA variability.

• Improved HRA trackability tracing uncertainties in the scenario; improved HRA transparency discovering analysts interpretations.

• Produced systematic understanding about human failure events:

what can go wrong, what are the causes, and how to militate the causes.

• Increased the applicability of HRA to all nuclear risk-informed applications.

51

Outline I. Overview of IDHEAS suite II. Introduction to IDHEAS-G, IDHEAS-ECA, and IDHEAS-DATA III. Examples of IDHEAS applications IV. Revision to IDHEAS reports after 9-23-2020 ACRS Subcommittee meeting 52

IDHEAS-G review and development IDHEAS-G was developed with inputs from reviewers:

• 10+ ACRS reviews, 3 external reviews, 2 internal reviews Development

• 7 tear-downs/rewrites of the IDHEAS-G report

• Each rewrite included new developments initiated from Review review inputs Examples of review-inspired IDHEAS-G development:

• ACRS recommendation on modeling timing effect Time uncertainty model as a part of HEP quantification

• Dr. E. Roth comments on teamwork the fifth macrocognitve function Interteam Coordination

• Drs. N Siu and K. Coynes comments on having a cohesive methodology the 8-step IDHEAS-G process as a stand-alone method for human event analysis 53

2020 IDHEAS-G report revision Comments from

- 9/18/2019 ACRS Subcommittee meeting

- External peer review comments

- NRC management review and project team review Summary of the revision:

1) All the comments were addressed except for ones that were outside the scope of IDHEAS-G report

2) Additional revisions were made to address some comments on draft reports of IDHEAS-ECA, IDHEAS-DATA, and FLEX HRA evaluation.

3) Major updates to Chapter 6 on data generalization were made to be consistent with IDHEAS-DATA report.

4) The PIF Interaction section in Ch.6 and Appendix D were rewritten using the new materials in the IDHEAS-DATA report.

54

2020 FLEX HRA report revision Revision of Volume 1: FLEX HRA Expert Elicitation

1) Most comments were addressed except the ones that were outside the project scope (e.g., assumptions made about the scenarios and context)

2) Several comments were about a major caveat in the expert elicitation process: The experts were uncomfortable estimating the HEP distributions, thus they only estimated the most likely, lower bound, and upper bound of the HEPs. The revision discussed this caveat.

Revision of Volume 2: FLEX HRA evaluation using IDHEAS-ECA

1) All comments provided to the project team that were within project scope and related to the results and conclusions were addressed.

2) A table was used internally to track comments and their resolution. Some comments overlapped or conflicted which is documented in the table.

3) The NRCs resolution of the comments were documented 55

Revision of IDHEAS-ECA report and IDHEAS-DATA report in 2021 Revision plan to IDHEAS-ECA report

1) Address comments from 2019, 2020, 2021 ACRS meetings as well as internal and external reviewers;

2) Clarify some PIF attribute definitions based on lessons learned from using IDHEAS-ECA in FLEX evaluation, SDP/ASP practices, and Dependency Workgroup;

3) Add a new chapter on guidance of using IDHEAS dependency model.

Revision plan to IDHEAS-DATA report:

1) Address comments from 2019, 2020, 2021 ACRS meetings as well as internal and external reviewers;

2) Incorporate corrections and recommendations from PNNLs independent verification and review of 2020 draft IDHEAS-DATA report.

56

The Integrated Human Event Analysis System (IDHEAS) Program Path Forward Sean E. Peters Advisory Committee on Reactor Safeguards February 4, 2021

IDHEAS Future Work

• IDHEAS-G - Publication

• IDHEAS-ECA - Refinement/Rollout

- Dependency

- Recovery

- Integrate with SAPHIRE/SPAR Models

- Publication of revision

• IDHEAS-DATA

- Completion/Publication

- Revision 11

Other HRA Work

• Minimum joint human error probabilities

• Data!

• Wish List

- Errors of commission

- Data for Org Factors

- Security (Physical and Cyber) 12

Path Forward

• Complete/Practical HRA Method

• Improvement to the current state of practice at the NRC

• Human-centered, scientific and data-based

• Program for periodic updates based on user feedback and data

• Can be applied to all NRC applications

• Closure of SRM-M061020?

13

QUESTIONS/DISCUSSION 14

Implementation Action Plan (IAP)

Strategy 2 Volume 4 - Licensing and Siting Dose Assessment Codes Volume 5 - Plans for Radionuclide Characterization, Criticality, Shielding, and Transport in the Nuclear Fuel Cycle February 4, 2021 Kimberly A. Webber, Ph.D.

Division of Systems Analysis Office of Nuclear Regulatory Research

Agenda

• Staff Introduction

• Overview

• Advanced Reactor Code Development Plans

- Volume 4 - Licensing and Siting Dose Assessment Codes

- Volume 5 - Plans for Radionuclide Characterization, Criticality, Shielding, and Transport in the Nuclear Fuel Cycle 2

NRCs Be Ready Attitude

• Improve mission value while enabling safe operations BlueCRAB

- Deliver cost savings

- Develop regulatory tools

- Build staff expertise

- Leverage collaborations 3

NRCs Integrated Action Plan (IAP) for Advanced Reactors Near-Term Implementation Action Plan Strategy 1 Strategy 4 Knowledge, Skills, Industry Codes and Capacity and Standards Strategy 5 Strategy 2 Technology Analytical Tools Inclusive Issues ML17165A069 Strategy 3 Strategy 6 Flexible Review Communication Process 4

IAP Strategy 2 Volumes to Date These Volumes outline the specific analytical tools to enable independent analysis of non-LWRs, gaps in code capabilities and data, V&V needs and code development tasks.

Introduction Volume 1 Volume 2 Volume 3 ML20030A174 ML20030A176 ML20030A177 ML20030A178 Volume 4 Volume 5 5 ML20028F255 ML20308A744

NRCs Integrated Action Plan (IAP) Status 6

Volume 4 - Licensing and Siting Dose Assessment Codes John Tomon, CHP Chief, Radiation Protection Branch Office of Nuclear Regulatory Research 2/04/2021

Volume 4: Licensing and Siting Dose Assessment Codes

• Technical Issues

- Potential for a spectrum of Non-LWR and fuel designs

- Over 10 licensing and siting dose assessment codes

- Inconsistent code development practices, by various contractors, over decades

- Overlap in code capabilities and need to use resources pragmatically

• Approach (Tasks)

- Consolidate/Modernize Dose Assessment Codes

- Improve characterization of Source Terms

- Improve Atmospheric Transport & Dispersion Models

- Update Dose Coefficient values

- Update Environmental Pathway Model 02/04/2021 2

Licensing and Siting Dose Assessment Codes Over 10+ codes used for NPP licensing and siting based on various regulations. Image adapted from BNWL-1754, Models and Computer Codes for Evaluating Environmental Radiation Doses.

02/04/2021 3

Code Consolidation and Modernization (Task 1)

• Given the large number of Non-LWR technologies being conceived and developed, it will be resource intensive to modify each of the siting, licensing, and emergency response codes for each design type.

• Therefore, the staff is proposing to consolidate and integrate them into several codes (i.e., two or three) that are modular, flexible, efficient, and user-friendly.

02/04/2021 4

Code Consolidation Approach The three pillars to the dose assessment code consolidation process:

- Create consolidated engines

- Develop a standardized data transfer schema

- Build a single user interface PNNL-29717, Health Physics Codes Consolidation and Modernization 02/04/2021 5

Conceptual Model for the Consolidated Code PNNL-29717, Health Physics Codes Consolidation and Modernization 02/04/2021 6

Source Term (Task 2)

• Identify source terms inputs (i.e., radionuclide fuel inventories, reactor coolant inventories, plant design and operational data) for each of the Non-LWR designs.

- Normal (Routine) source terms

- Severe Accident source terms

- Design-Basis Accident source terms

- Transportation source terms 02/04/2021 7

Source Term (Task 2)

• Source Term Considerations:

- Source term/release rate framework database will:

• Leverage activities from Volumes 3 and 5

• Estimate inventory in core/release from core

• Identify dominate release pathways

• Characterize mechanism to reduce release (e.g. filters)

• Estimate release rates,

• Use operational data where applicable 02/04/2021 8

Normal Operation Source Term 02/04/2021 9

ATD Module (Task 3)

• ATD consolidation in Phases:

- Phase 1: Consolidate ARCON, PAVAN & XOQDOQ.

- Phase 2: Evaluate the applicability of the near-field and ATD models for Non-LWR technologies.

• These phases would leverage the experience of the NRC-meteorology staff and any near-field modelling efforts from Volume 3.

02/04/2021 10

Atmospheric Engine Prototype 02/04/2021 11

Dose Coefficient Module (Task 4)

• This task involves:

- Developing dosimetry modules/engines that have the flexibility to use different dose models and dose coefficient values

- Examining dose coefficient models with respect to aerosol particle size in addition to exploring the impact of tritium and carbon-14 biokinetics since these radionuclides may be in higher quantities in non-LWRs.

02/04/2021 12

Dose Coefficient Considerations

• Vision for module:

- Flexible Engines for different dose coefficient values

- Dose Coefficient Package Code (DCFPAK)

- Aerosol particle size relative to dose coefficients

- H-3 and C-14 relationship to dose coefficients

• Current State

- Some codes can choose different data sets.

- Leveraging DCFPAK datasets with US EPA.

- Possibly acquiring international dosimetry codes.

- Training RPB staff on specific designs where internal dosimetry could be significant such as MSRs.

02/04/2021 13

Environmental Pathways (Task 5)

• Further developing the aquatic pathways (river/lake/ocean dispersion), environmental accumulation, and human/non-human biota consequence modules for codes.

• Lower priority because they are less dependent on Non-LWR designs and fuel types.

• Explore the feasibility of radionuclide particle size behavior in the environment for some non-LWR designs.

02/04/2021 14

Volume 4 Implementation Plan 02/04/2021 15

Thank You 02/04/2021 16

Back Up Slides 02/04/2021 17

Non-LWR Technologies Non-LWR Plant Description Examples Fuel Types 1 HTGR; prismatic core, thermal spectrum Framatome TRISO (rods or plates)

X-energy, 2 PBMR; pebble bed core, thermal spectrum TRISO (pebbles)

Starcore 3 GCFR; prismatic core, fast spectrum GA SIC clad UC (plates)

PRISM, ARC, 4 SFR; sodium cooled, fast spectrum Metallic (U-10Zr)

TerraPower Westinghouse, (Possibly nitride 5 LMR; lead cooled, fast spectrum Columbia Basin, fuel.)

Hydromine 6 HPR; heat pipe cooled, fast spectrum Oklo, Westinghouse Metallic (U-10Zr) 7 MSR; prismatic core, thermal spectrum AHTR TRISO (plates) 8 MSPR; pebble bed, thermal spectrum Kairos TRISO (pebbles)

MFSR; fluoride fuel salt, Terrestrial Thorcon, 9 Fuel salt thermal/epithermal spectrum FliBe 10 MCSR; chloride fuel salt, fast spectrum TerraPower, Elysium Fuel salt 02/04/2021 18 Back

Regulatory Needs for Dose Assessment Codes 02/04/2021 19 Back

Safety & Environmental Review Codes ARCON SNAP/RADTRAD (DBA)

(/Q)

LPZ Control EAB Room PAVAN

(/Q)

Image adapted from BNWL-1754, Models and Computer Codes for Evaluating Environmental Radiation Doses.

02/04/2021 20

Safety & Environmental Review Codes Control Room EXTRAN CHEM DEGADIS SLAB HABIT (Chemical)

Image adapted from BNWL-1754, Models and Computer Codes for Evaluating Environmental Radiation Doses.

02/04/2021 21

Safety & Environmental Review Codes NRCDose GASPAR GALE 80 km XOQDOQ

(/Q)

RADTRAN GALE LADTAP Image adapted from BNWL-1754, Models and Computer Codes for Evaluating Environmental Radiation Doses.

02/04/2021 22

Dose Consequences Code RASCAL Accident 129 km 129 km Image adapted from BNWL-1754, Models and Computer Codes for Evaluating Environmental Radiation Doses.

02/04/2021 23

Decommissioning Codes

• DandD (Decontamination and Decommissioning): compliance with the dose criteria of 10 CFR Part 20, Subpart E. Perform simple estimates of the annual dose from residual radioactivity in soils and on building surfaces.

• RESRAD (Residual Radioactivity): Family of codes used to analyze human and biota radiation exposures from environmental contamination of residual radioactive materials.

02/04/2021 24

Research and Other Purposes

• GENII (Generation II computer code): a set of programs for estimating radionuclide concentrations in the environment and dose to humans from acute or chronic exposures from radiological releases to the environment or initial contamination conditions.

• Dose Coefficient File Package (DCFPAK):

that includes nuclear decay data and dose and risk coefficients for exposure to radionuclides.

• SCALE and MELCOR are used in development of core radionuclide inventory and severe reactor accident source terms as described in Volume 3. Plan to leverage work done for Volume 3 in the licensing and siting dose assessment codes.

02/04/2021 25

Other Considerations/Challenges

• Timing of Non-LWR submittals vs code readiness

• Consolidation vs no consolidation

• Wide range of program office participation and input

• Managing expectations 02/04/2021 26

Code Readiness

• Next Steps for Volume 4: (Near- & Mid-Term)

Activity Date Brief SC and Full ACRS Sept 2020/Feb 2021 Build Consolidate Code Framework FY 2021 Obtain Source Terms from Most Probably Designs Ongoing Pilot of Atmospheric Models FY 2021 Include Non-LWR HP Operational Experience FY 21 and beyond (Domestic and International)

Dose and Environmental Engines FY 23 and beyond 02/04/2021 27

NRC nonLight Water Reactor Vision and Strategy, Volume 5:

Radionuclide Characterization, Criticality, Shielding, and Transport in the Nuclear Fuel Cycle Presented by Don Algama (RES) and Drew Barto (NMSS)

United States Nuclear Regulatory Commission Office of Nuclear Regulatory Research (RES)

Nuclear Materials Safety and Safeguards (NMSS) 02/04/2021 1

Acknowledgements

• This work was completed thanks to many contributors from NMSS, NRR and RES.

• Dr. David Luxat (Sandia) and Dr. William Wieselquist (ORNL) were instrumental in the plan development.

2

IAP Strategy 2 Volumes to Date Introduction Volume 1 Volume 2 ML20030A174 ML20030A176 ML20030A177 Volume 5 ML20308A744 Volume 3 Volume 4 ML20030A178 ML20028F255 3

Objectives

• Elements of the fuel cycle plan

- Evaluate and demonstrate computer code readiness

- Evaluation and use of existing NRC computational tools for accident analysis (Volume 3) and consequence assessment (Volumes 3/4)

- Incremental development approach based on existing LWR fuel cycle as reference

- Staff experience with anticipated nonLWR fuel cycle and use of computer codes

- Development of nonLWR fuel cycle reports and publicly available input decks 4

Analysis Approach Develop accident scenarios by reviewing available information including documents such as:

• NUREG/CR6410 Nuclear Fuel Cycle Facility Accident Accident Analysis Handbook

• NUREG1520 Standard Review Plan for Fuel Cycle Facilities License Applications

• NUREG2215 Standard Review Plan for Spent Fuel Dry Storage Systems and Facilities - Final Report

• NUREG2216, Standard Review Plan for Spent Fuel Transportation

• DOEHDBK12242018: Hazard and Accident Analysis Handbook 5

Severe Accident & Consequence Analysis (LWR/nonLWR example) analysis endpoints ENDF/B Physics data Thermal scattering law, resonance data, energy distributions, fission yields, decay constants, etc. AMPX Validated cross section libraries in multigroup (O(100g)) or continuousenergy JEFF Activation (O(100,000g); depletion and Isomeric cross sections, decay data activation reactions ORIGEN General depletion, Sources4C decay, source term neutron emission data (alpha,n)

TRITON/SHIFT Kinetics Data General reactor fuel neutron MELCOR nuclidespecific beta transport + depletion effective, precursor data Severe accident consequence progression and analysis NIST ORIGAMI mechanistic Reactorspecific radioactive source terms natural abundance, atomic isotopics/source term characterization mass NRC NonLight Water Reactor Vision and Strategy, Volume 3 - Computer Code Development Plans for Severe Accident Progression , Source Term, and Consequence Analysis, Revision 1, January 2020, ML20030A178 6

Scope of Analysis Follow these analysis steps used in Volume 3

• Assess existing codes to cover neutronics and previous fuel cycle work for LWRs and radionuclide and nonradionuclide

1. Define scenario hazards throughout nonLWR fuel cycles 2. Identify safety related item(s) of

• Consequence and radiation protection interest

3. Ask the right safety questions /

methods are covered under Volumes 3/4 Phenomena of interest / Understand

• Mining, milling, long term storage and the dominant features

4. Survey experiments available that disposal are not considered in this activity provide fundamental information

• Leverage Volume 3 nonLWR designs 5. Develop physics models to capture dominant feature and allow prediction

- FluorideSaltCooled (SolidFuel) High 6. Translate physics models into computer Temperature Reactor (FHR) code

7. Perform verification testing (unit

- Heat Pipe Reactor (HPR) testing; and integrated testing as code complexity increases)

- Sodium Fast Reactor (SFR) 8. Perform validation with experiments.

- High Temperature Gas Reactor (HTGR) Capture the integrated codes performance (with uncertainty analysis)

- Molten Salt Reactor (MSR) 9. Document findings 7

Deliverables

• 10 reports are defined as a result of this plan

- Each report defines a set of accident scenarios during a portion of the fuel cycle

- Perform assessment, analysis, and generate demonstration input files

• 5 nonLWRs currently considered, and openly available reference designs are defined in Volume 3:

1. FHR Fuel Cycle Analysis (Berkeley Mk. 1)

2. HPR Fuel Cycle Analysis (INL Design AMET) This organization of deliverables allows

3. SFR Fuel Cycle Analysis (MET1000/VTR) prioritizing specific designs and

4. HTGR Fuel Cycle Analysis (PBMR400)

5. MSR Fuel Cycle Analysis (MSRE) reducing overlap. For example:

• 5 front end (FE) reports centralize FE

• HTGR analysis requires the analysis among these nonLWRs following reports

6. Enrichment and UF6 Handling up to 20 wt.% 67104.

7. TRISO Fuel Kernel Fabrication

• For FHR, it would require

8. Uranium Metallic Fuel Fabrication 67101. 6,7, and 10 are

9. Fast Reactor Fuel Assembly Fabrication

10. Pebble TRISO Fuel Fabrication already available!

8

Reference LWR Cycle Each analysis report tackles one or more of the equivalent fuel cycle stages for each nonLWR.

NOTE: Transportation offsite and off site storage (T3 and S1) are currently not considered in this fuel cycle assessment plan due to uncertainty with this part of the back end.

9

HTGR Fuel Cycle Report The HTGR fuel cycle report develops and analyzes new accident scenarios related to stages U1 and U4 and links them to frontend stages (E1, T1, F1, F2, T2) analyzed in this project and inreactor accident scenarios U2 from volume 3.

Front end analysis is basically the same as for FHR. 10

Concluding Remarks

• Relying on a reasonable and flexible approach

• Sufficient capabilities to support nonLWR fuel cycle analyses

• Decades of model development and validation can be applied to nonLWR analyses as in Volume 3 and other programs

• Plan will be updated as more experience is gained and as new information becomes available 11

Back Up 12

Regulatory Application of Codes Reactor Aux. Systems; Transport; Storage Fuel Mechanical Thermal-Fluid Materials Nuclear Design Criticality/Shielding Design Design Accounting 10 CFR 50.68 (GCD 62); 10 CFR 70, 71 GDC 2 GDC 10 GDC 10, 11, 12, 26, 27, 28 GDC 10 GDC 12 10 CFR 74 and 72 Dynamic Thermo- Reactor Systems Criticality and Shielding Analyses Inventory Stability Stress Mechanical Physics Analysis/Thermal Analysis Analysis Performance Analysis Margin SCALE Engineered Safety FAST SCALE/PARCS/TRACE Features Containment Core Cooling Protection Against Radiological Release 10 CFR 50.44 10 CFR 50.46 Safety Review GDC 16, 38, 50 GDC 10, 34, 35 10 CFR 50.34; 10 CFR 50.67 Environmental Review 10 CFR 52.47; 10 CFR 100 10 CFR 51.30 GDC 19 10 CFR 51.50 Systems Analysis 10 CFR 50.160 (EP) 10 CFR 51.70 (inputs from fuel thermo- 10 CFR 52.17 (ESP) 10 CFR 51.75 Containment Analysis mechanical and reactor 10 CFR 52.79 (COL) physics analyses)

MELCOR FAST, Source Term/Dose Consequence Analysis SCALE/TRACE/PARCS MELCOR/ RADTRAD/

MACCS SCALE RASCAL 13

Transportation and Storage Licensing (LWR) analysis endpoints ENDF/B Physics data Thermal scattering law, resonance data, energy distributions, fission yields, decay Thermal Analysis constants, etc.

AMPX Structural and Validated cross section JEFF Activation libraries in multigroup containment Isomeric cross sections, (O(100g)) or continuous SHIFT/MAVRIC activation reactions energy (O(100,000g);

depletion and decay data 3D shielding and dose rate (cask) analyses analysis Sources4C neutron emission data ORIGEN CSAS (alpha,n) General depletion, 3D criticality safety analysis decay, source term ICRP dose conversion factors, radiotoxicity TRITON/SHIFT ORIGAMI General reactor fuel Reactorspecific radioactive NIST neutron transport + isotopics/source term natural abundance, atomic depletion characterization mass 14

Examples of Existing Fuel Cycle Analysis

• Level 3 PRA Project

- SCALE/MELCOR are used to support PRA development of accident sequences and source terms including nonreactor scenarios for the spent fuel pool

• NUREG2161

- SCALE/MELCOR was used to study the performance of a SFP under severe accident conditions

• NUREG/CR7108/7109

- Here SCALE was used to estimate isotopic depletion and criticality code, and cross section data bias related to burnup credit in spent fuel storage and transportation systems 15

Examples of Existing Fuel Cycle Analysis

• Barnwell - NonReactor Safety Assessment

• SCALE/MELCOR utilized as part of bestestimate analysis methodology in NUREG/CR7266

• Spent fuel inventories developed in SCALE package

• Aerosol transport modeling

- Integral analyses estimate radiological transport and release

- Aerosol modeling enables estimation of transport of hazardous material within facility and to environment

• Accident scenarios considered relevant to broad range of facility accidents

- Explosion scenario

- Fire scenario

- Combined explosion and fire scenario 16

nonLWR Characteristics Table 11. Comparison Between LWR and NonLWR Enrichment Typical Discharge OnSite Fuel Fuel Storage /

Reactor Type Fuel Form Fuel Residence Time (wt.%) Burnup Processing Transport Peak Rod Average: Storage:

<62 GWd/MTU Fresh and spent fuel LWR Assemblies burned for

<5 U Oxide No storage onsite or (Ref.) approximately 3 to 4 cycles Max Assembly Average: offsite

<55 GWd/MTU Transport:

FE: UF6 solid transport in 30B cylinders, fresh fuel Peak Rod Average:

assembly and fuel LWR: HALEU ~75 Wd/MTU Assemblies burned for component (UO2

/HBU 5 - 10 U Oxide No approximately 3 to 4 cycles powder/pellet)

(Ref.) Max Assembly Average:

transportation

~6070 GWd/MTU packages BE: Used fuel transport and dry storage containers U Oxide HPR 5 - 20 210 GWd/MTU Up to 7yrs No To be evaluated*

U Metal SFR 5 - 20 U Metal Up to 300 GWd/MTU To be evaluated* No To be evaluated*

TRISO (UCO or UO2)

HTGR 5 - 20 in pebble bed or 100200 GWd/MTU To be evaluated* No To be evaluated*

prismatic array TRISO (UCO or UO2)

FHR 5 - 20 100200 GWd/MTU To be evaluated* No To be evaluated*

in pebble bed 235U dissolved in MSR 5 - 20 To be evaluated 23yrs Yes To be evaluated*

molten salt 17

• • Will be evaluated 1 atom% burnup isbased on information approximately available at the time work is undertaken, e.g. based on current DOE and industry input.

9.4 GWd/MTU.

FHR Fuel Cycle Report The FHR fuel cycle report develops and analyzes new accident scenarios related to stages U1 and U4 and links them to earlier frontend stages (E1, T1, F1, F2, T2) analyzed in this project and inreactor scenarios U2 from volume 3.

18

HPR Fuel Cycle Report The HPR fuel cycle report develops and analyzes new accident scenarios related to stages F2, T2, U1 and U4 but also requires re analysis of U2 for a metallic fuel system (current source term demo calcs using oxidic fuel). NOTE: The F2 and T2 front end stages are included in this report because fabrication and transportation of an HPR core will be specific to that design and thus nothing is gained from putting those stages in their own analysis reports. 19

SFR Fuel Cycle Report The SFR fuel cycle report develops and analyzes new accident scenarios related to stages U1, U3, and U4 and links them to previously studied E1, T1, F1, F2, and T2. NOTE: The F2 and T2 front end stages are not developed as separate reports, since fabrication and transportation of an HPR core will be specific to the reactor design. Nothing is gained from putting those stages in their own analysis reports. 20

MSR Fuel Cycle Report The MSR fuel cycle report has the least overlap with any other design and develops and analyzes new accident scenarios for F1, T2, U1, and U4 in the main MSR analysis and links them only to front end E1 and T1 for UF6 enrichment and transportation. 21

Leveraged Programs

• HALEU

- UF6 transport packages

- Fresh fuel transport packages

• Volume 3 (codes and plant models)

- Capabilities to characterize utilization stage

- Hazardous material transport for nonwater systems

• DOE Programs

- DOENE spent fuel and waste science and technology program

- Support hazard identification and characterization 22