ML22105A418
| ML22105A418 | |
| Person / Time | |
|---|---|
| Issue date: | 03/08/2022 |
| From: | NRC/NRR/DRA |
| To: | |
| Pressley L | |
| References | |
| NRC-1861 | |
| Download: ML22105A418 (72) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
34th Regulatory Information Conference (RIC)
Technical Session T8 Docket Number:
(n/a)
Location:
teleconference Date:
Tuesday, March 8, 2022 Work Order No.:
NRC-1861 Pages 1-71 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1716 14th Street, N.W.
Washington, D.C. 20009 (202) 234-4433
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 34TH REGULATORY INFORMATION CONFERENCE (RIC) 4
+ + + + +
5 TECHNICAL SESSION - T8 6
MAKING EFFECTIVE DECISIONS IN USING DEFENSE IN 7
DEPTH, SAFETY MARGINS, AND RISK!
8
+ + + + +
9
- TUESDAY, 10 MARCH 8, 2022 11
+ + + + +
12 The Technical Session met via Video-13 Teleconference, at 3:00 p.m. EST, Andrea Kock, Deputy 14 Office Director for Engineering, Office of Nuclear 15 Reactor Regulation, presiding.
16 17 PRESENT:
18 ANDREA KOCK, Deputy Office Director for Engineering, 19 NRR/NRC 20 DOUG TRUE, Chief Nuclear Officer and Senior Vice 21 President, Nuclear Energy Institute 22 SMAIN YALAOUI, Senior PSA Technical Specialist, 23 Directorate of Assessment and
- Analysis, 24 Canadian Nuclear Safety Commission 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
2 MARK THAGGARD, Director, Division of Risk Analysis, 1
RES/NRC 2
MIKE FRANOVICH, Division of Risk Assessment, NRR/NRC 3
LUNDY PRESSLEY, Reliability and Risk Analyst, PRA 4
Oversight Branch, Division of Risk Assessment, 5
NRR/NRC 6
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 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
3 TABLE OF CONTENTS 1
PAGE 2
Introductory Remarks 3
4 Risk-informed Decision-making: Greater 5
Than the Sum of its Parts 6
Doug True 9
7 CNSC Risk Informed Decision Making 8
Smain Yalaoui
................. 20 9
Polling Questions
............... 31 10 Safety Marker Study 11 Mark Thaggard
................. 34 12 Polling Question................ 42 13 Safety Improvements Using Risk Insights 14 Mike Franovich
................. 43 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
4 P-R-O-C-E-E-D-I-N-G-S 1
MS. KOCK: Good afternoon, everybody.
2 Welcome to the NRC's RIC technical session on making 3
effective decisions in using defense-in-depth safety 4
margins in risk. This session is a joint effort by 5
the NRC's Office of Nuclear Regulatory Research and 6
the Office of Nuclear Reactor Regulation.
7 Before I start, I just want to acknowledge 8
and support our friends in the Ukraine and their 9
valiant efforts to defend their country. We want you 10 to know that we stand with you in this time of 11 challenge. Next slide, please.
12 I just wanted to go over the agenda 13 quickly. I'm going to start with some quick 14 introductions and then our panel members will present 15 and discuss the effective application of risk-informed 16 decision making.
17 They're going to talk about some practical 18 examples, best practices, and lessons learned, 19 successes, challenges, and other considerations in how 20 risk has helped us keep our focus on safety.
21 This morning, Chairman Hanson touched on 22 the importance of risk in decisions in a wide range of 23 views. This session will facilitate understanding of 24 what drives risk and the differences of perspectives 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
5 on risk.
1 For example, we'll cover misconceptions 2
about the role of risk considerations in defense-in-3 depth and how we integrate the use of risk-informed 4
decision making with deterministic reviews for 5
effective decision making.
6 These are historically some of the most 7
challenging issues, and today you're lucky because you 8
have the experts here to assist with untangling the 9
ins and outs of risk-informed decision making.
10 Additionally, the NRC and our external 11 stakeholders will highlight both the positive benefits 12 and potential pitfalls of using risk-informed decision 13 making and it will provide you a great opportunity to 14 engage with the panelists who are experts in this very 15 important area on the issues during the panel 16 discussion.
17 We have about an hour and a half for our 18 discussion and we plan to spend about half of that 19 time in discussion, so we ask that you hold your 20 questions until all of the speakers have finished with 21 their presentations.
22 Just a few acknowledgments before we 23 start. I want to thank everyone for their preparation 24 and participation in this session. In particular, I 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
6 want to thank the session speakers for agreeing to 1
participate in this session and for their effort taken 2
to prepare for the session.
3 And a special thanks to Lundy Pressley and 4
Matt Humberstone of the NRC for their work in 5
coordinating this session, and, of course, our AV 6
staff that's making sure that everything goes very 7
smoothly. Thank you.
8 A quick overview, and I think this has 9
already been said before, but the journey to become a 10 more modern risk-informed regulator is a really 11 important one and it will help us to more effectively 12 accomplish our mission to protect people and the 13 environment by helping us focus on safety-significant 14 aspects of our work.
15 This is an important topic as risk-16 informed decision making is at the heart of sound 17 regulatory practices, and for the NRC, it's a critical 18 part of our regulatory transformation.
19 The importance of this topic, I think, is 20 reflected in the number of people who registered for 21 this session. At last count, I think it was over 800.
22 That's quite impressive.
23 I just want to start by introducing our 24 panelists. First, we're going to hear from Doug True.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
7 He's the Nuclear Energy Institute's chief nuclear 1
officer and senior vice president of generation and 2
suppliers at the NEI.
3 He has more than 40 years of nuclear 4
industry experience in nuclear safety, and prior to 5
joining NEI, he contributed to many of the major 6
milestones in risk-informed regulation and he was 7
responsible for one of the largest specialty 8
engineering organizations in the nuclear industry, 9
including being president of the largest nuclear PRA 10 firm in the world.
11 Secondly, we're going to hear from Smain 12 Yalaoui.
He's a
senior probabilistic safety 13 assessment technical specialist with the Canadian 14 Nuclear Safety Commission.
15 Mr. Yalaoui has a Master's in Nuclear 16 Engineering. He specialized in probabilistic risk 17 assessment, and he joined the Canadian Nuclear Safety 18 Commission in 2008.
19 Internationally, Mr. Yalaoui contributed 20 to the development of IAEA safety report series, and 21 he is a member of the NEA working group on risk. He 22 specifically participated in the IAEA and WGRISK 23 projects on multi-unit and site-level PSA. Mr.
24 Yalaoui took part in the International Seismic 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
8 Probabilistic Seismic Assessment peer reviews.
1 Our third presenter will be Mark Thaggard.
2 He serves at the deputy director of the Division of 3
Risk Analysis here at the NRC in the Office of Nuclear 4
Reactor Research.
5 Mr. Thaggard joined the NRC in 1989, and 6
in 2009, he was selected for the Senior Executive 7
Service, serving in the Office of Federal and State 8
Materials and Environmental Management Office and also 9
in the Office of Nuclear Security and Incident 10 Response, and most recently in the Office of Research 11 as a director of the Division of Risk Analysis.
12 Our last presenter will be Mike Franovich 13 and he is the director of the Division of Risk 14 Assessment in the Office of Nuclear Reactor 15 Regulation.
16 He has over 30 years of nuclear 17 experience. Mr. Franovich is a member of the Senior 18 Executive Service and he previously served as deputy 19 and acting director of the Fukushima Lessons-Learned 20 Division, enhancing defenses against extreme natural 21 events.
22 He currently leads the Division of Risk 23 Assessment, which conducts probabilistic risk analysis 24 and establishes regulatory standards for risk-informed 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
9 nuclear reactor licensing oversight, accident 1
consequence analysis, health physics, and fire 2
protection engineering.
3 And without further ado, I'm going to turn 4
the panel session over to our first presenter, Mr.
5 Doug True.
6 MR. TRUE: Thanks, Andrea. I'm glad to be 7
here today and thanks for inviting me. I look forward 8
to the discussion today. I titled my presentation 9
today Risk-Informed Decision-Making: Greater Than the 10 Sum of its Parts, because I really believe that that's 11 true.
12 In fact, I think it was over 25 years ago 13 I wrote a paper that talked about how, when you use 14 PRA information, it's important to consider defense-15 in-depth and safety margins and the performance of 16 equipment in those considerations, and that was two 17 years even before Reg. Guide 1174 was invented and the 18 term risk-informed was coined.
19 So, this is a subject that's near and dear 20 to my heart and I really believe that there are some 21 misconceptions and some misunderstandings about this 22 side, but I'm going to try to talk our way through.
23 And I want to start with a discussion on 24 safety versus risk. A lot of times, we think of those 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
10 as being different things. I actually believe they 1
are intimately related, so let's go to the next slide.
2 So, as we all know, our regulations for 3
the current plants are based on deterministic 4
requirements to provide that foundation for ensuring 5
the safety of our nuclear power plants.
6 Risk analysis gives us a tool that allows 7
us to assess the risks that remain when the 8
regulations are met, and that risk is never zero.
9 The risk analysis not only gives us a 10 chance to estimate those residual risks or a level of 11 safety, quantitative level of safety, but it also lets 12 us understand what contributes to them, and through 13 that, we can understand also chances in requirements, 14 either additional requirements or relaxations in 15 requirements, and how they impact that risk, that 16 residual risk, and in the case of the STP process 17 under the ROP, what non-compliances look like and what 18 the significance of those are.
19 This means that really what we do in the 20 safety side is tied directly to what we measure on the 21 risk, and I want to turn in the next slide to an 22 example of that.
23 So, as we think about the relationship 24 between risk and safety, you can have different levels 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
11 of safety requirements across the bottom of this graph 1
and different levels of risk on the left axis.
2 Here, we're looking at redundancy. You 3
can have no safety systems and very high risk, one, or 4
two, or three, or four safety systems and have lower 5
and lower risk.
6 You can add a whole bunch of safety 7
systems, but there will always be a residual risk, and 8
what we're trying to do in understanding what that 9
residual risk looks like is to understanding what 10 contributes to it, how in this case redundancy 11 applies, but it works in all kinds of different 12 directions, not just in the sense of redundancy.
13 You could talk about how much safety 14 margin we need in our containment, how much shielding 15 you need for radiation safety. It's understanding 16 that residual that's left after we've decided what 17 level of deterministic requirement we're going to 18 have.
19 The benefit of that understanding of what 20 contributes allows us to make risk-informed decisions.
21 Let's go to the next slide.
22 And that understanding allows us to 23 actually improve safety. It gives us the ability to 24 focus on what's truly safety significant. It allows 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
12 us to allocate resources in the manner that most 1
effectively improves safety so we can focus on the 2
things that are the biggest contributors to risk and 3
maybe minimize the amount of effort we put on things 4
that are less important to make sure we're staying 5
focused on the things that are most important.
6 Risk informing also incentivizes licensees 7
to focus on what's important to safety. If they 8
understand what's important to safety, they know where 9
they run the risk with the regulator of having a 10 significant safety issue. As I said, it also allows 11 us to know where we can spend less time on things of 12 less importance.
13 Overall, we've seen across the last 25 or 14 30 years that this focus on safety significance that 15 risk allows us to do actually stimulates a net 16 improvement in safety. What you focus on actually 17 improves, and therefore risk goes down.
18 You've seen me and others at NEI present 19 that curve of PF versus time that shows how we've 20 driven risk down. That curve happens to apply just to 21 internal events, but the same thing is true of 22 external events. All of the work we did in NFPA 805 23 greatly reduced fire risk at those plants. Other 24 enhancements we've made in plants have reduced other 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
13 contributors to risk.
1 But what's important when we do this is 2
that when we're using PRA as a tool is that we 3
understand its limitations. As much as I'm a PRA 4
practitioner and have been one for now 40-odd years, 5
it's a tool that can only be used within its 6
limitations and it has to be used appropriately.
7 It's neither omnipotent, nor omniscient.
8 It doesn't always come out with an answer that can be 9
trusted as just a flat number. You've got to 10 understand what contributes to it.
11 But when I say that about PRA, the same 12 thing applies to the deterministic approaches.
13 There's nothing perfect about a deterministic approach 14 either. It has its own limitations that we've learned 15 over the years, which is one of the reasons why we 16 brought risk into our decision-making process. So, 17 let's move to the next slide and talk about that.
18 Reg. Guide 1174 outlines the risk-informed 19 decision-making process that's shown here on the 20 right-hand side of this graphic. The PRA results are 21 one input into that process and they're a product of 22 a model that certainly contains uncertainties.
23 Uncertainties exist whether you're using 24 a
deterministic approach or you're using a
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
14 probabilistic approach. It's just the PRA allows us 1
to quantify those and illuminate them in a way that is 2
very difficult to do if you're dealing with strictly 3
a deterministic basis.
4 The brilliance of Reg. Guide 1174 was that 5
it outlined an integrated decision-making process, a 6
process that used PRA as one piece, but also asked us 7
to consider defense-in-depth and safety margins in 8
that decision-making process.
9 And it's important that they use the term 10 integrated. It's not that PRA is a gate, that 11 defense-in-depth is a gate, and safety margins is a 12 gate, that you have to pass through each of those.
13 They need to be considered in an integrated manner 14 where you understand what the risk analysis is telling 15 you, what the defense-in-depth considerations are 16 telling you, and understand how those fit together.
17 Sometimes PRA isn't very good at modeling 18 things. Sometimes it's quite good at it and that 19 should be taken into account, and that goes to the 20 point of uncertainties, and what's important about 21 using PRA in a
decision-making process is 22 understanding the uncertainties that are important to 23 that decision.
24 Typically, when we're looking at a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
15 particular use of PRA, we're looking at one particular 1
slide, whether it's a piece of equipment that had 2
failed and had a performance deficiency associated 3
with it, or it's a
change in a
technical 4
specification, or some other plant configuration 5
change.
6 We need to understand what the role that 7
change makes in the overall risk profile and which 8
uncertainties contribute to that. If we do that, then 9
we can understand how that plays into the defense-in-10 depth and safety margin considerations.
11 And a good example of this is when we were 12 looking at what to do after Fukushima. We could have 13 gone after that with a strictly numerical approach and 14 tried to devise some method to say once we get below 15 ten to the minus X, we're good, but instead, we said 16 there's a lot of uncertainty here.
17 These box one events like happened at 18 Fukushima are rare. It's better for us to think about 19 this in a defense-in-depth posture and that's how we 20 ended up with FLEX.
21 So, PRA has its role. In certain things, 22 it's very good. In other cases, we need to think 23 about what the uncertainties are and put more emphasis 24 on other pieces of this decision-making process.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
16 And I just at the bottom there took a 1
quote right directly out of Reg. Guide 1174 that talks 2
about how important it is to take all of these pieces 3
together in an integrated manner so that we understand 4
the relationships between the risk, the safety 5
margins, defense-in-depth, how we can monitor that 6
performance, and what we're talking about in terms of 7
the regulation.
8 And I'll finish up with one last slide.
9 The next one talks about this and some examples. So, 10 from a regulatory perspective, we've had some fairly 11 significant decisions that have used the risk insights 12 to drive that.
13 The ATWS rule for sure was driven by our 14 understanding of ATWS events back at the time it was 15 promulgated, the Station Blackout rule where we 16 actually identified sort of a hole in the whole 17 defense-in-depth process where we realized that loss 18 of onsite power events were more likely and we needed 19 more than just a limited amount of redundancy.
20 We actually needed an ability to cope, and 21 in some cases, we needed an alternative AC power 22 supply in order to be able to mitigate those risks.
23 That insight wouldn't have come about from 24 a strictly deterministic understanding. It came about 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
17 because we understood the likelihood and the 1
consequences, and we understood the defense-in-depth 2
we had.
3 Another good example is in the Fukushima 4
period where we installed severe accident vents and 5
the water addition to BWR Mark 1 and Mark II.
6 In that process, we looked at a lot of 7
different options, but we learned that without the 8
water addition, having a severe accident vent wasn't 9
going to help us much because the containment was 10 ultimately going to fail, and that vent pathway that 11 we thought was going to give us benefit would actually 12 be compromised, and that understanding came about by 13 understanding the nature of the events that occur in 14 the severe accident regime and understanding how to 15 mitigate those and the uncertainties associated with 16 them.
17 And lastly, I won't go through all of 18 these in detail, but utilities and licensees have made 19 a lot of voluntary changes. I can't count the number 20 of plants who have found things that were compliant 21 with the regulations, but from a risk perspective, 22 were driving risk results, and they made enhancements 23 to their plants to address those.
24 A good example is fire protection piping, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
18 and a number of plants found a flooding risk from that 1
piping could delay the impacts on AC and DC power.
2 Plants weren't comfortable with that and they made 3
changes to mitigate that flooding risk and reduce 4
their overall risk profile because of that.
5 Plants have installed non-safety equipment 6
to perform functions that are important, made lots of 7
procedural changes to the user systems in different 8
ways, and fed back the insights from their PRAs into 9
training so that the human actions are in a context 10 that's without residual risk, so to make sure we're 11 training our operators on the actions that are really 12 going to be necessary in order to address some of the 13 most important things.
14 So, as you can tell, I'm very bullish on 15 the risk-informed concept. I think if we do it well, 16 we can really enhance and get a better value out of 17 our regulations by getting the best out of the risk 18 insights, the defense-in-depth, and safety margin 19 understandings that we already have.
20 With that, Andrea, I'll turn it back to 21 you.
22 MS. KOCK: Thanks, Doug. You gave us some 23 really good insights based on your experience, so I 24 know I appreciated that and I'm sure the audience did 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
19 too.
1 I think it's time for our first polling 2
question, so that's kind of exciting. So, these are 3
going to be a couple of questions about your, the 4
audience's experience in using risk-informed decision 5
making and how extensively you've used it and has it 6
benefitted your work?
7 So, we're going to ask these questions and 8
we're going to save the answers to the polling 9
questions to the end of the panel session, so you can 10 wait in anticipation for those.
11 And then just a note, in order to access 12 the polling questions, you can go to the right of your 13 screen and toggle from the Q&A space into the polling 14 questions so that you can answer those.
15 Okay, I can't see the polling questions.
16 Can anybody else on the panel see them? No? Okay, so 17 I think what we'll do is maybe move onto our next 18 speaker and maybe we'll catch up with the polling 19 questions after the second speaker.
20 So, our second speaker will be Smain 21 Yalaoui and he's going to be talking to us about risk-22 informed decision making in Canada, so I'm going to 23 turn it over to Smain.
24 MR. YALAOUI: Thank you very much, Ms.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
20 Chair. Thank you for having me at this technical 1
session. So, my presentation is about the risk-2 informed decision making in Canada. Next slide, 3
please.
4 So, this is the outline of my 5
presentation. First, I will discuss the RIDM within 6
the CNSC regulatory framework. I will discuss the key 7
principles of the RIDM. I'll provide a brief history 8
of RIDM in Canada.
9 I will discuss shortly the staff procedure 10 for risk evaluation, estimation and evaluation. I 11 will talk also about the PSA use to support the RIDM.
12 I will then elaborate on the CNSC risk handbook tool 13 that we have developed at the CNSC.
14 I'll talk about the emergency mitigating 15 equipment consideration known as FLEX in the U.S.
16 Next, I will discuss the benefits and pitfalls of PSA 17 use in RIDM, and then I'll finish with a short summary 18 of my presentation. Next slide, please.
19 So, for the RIDM regulatory framework, 20 here are some highlights. We have a regulatory 21
- document, REGDOC-3.5.3, which is regulatory 22 fundamentals. This regulatory document describes the 23 risk-informed approach to licensing and compliance 24 activities. It emphasizes that the focus is on issues 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
21 of higher risk.
1 There is also an internal RIDM procedure 2
at the CNSC that further elaborates on situations 3
where staff can apply a risk-informed approach for 4
regulatory requirements and
- guidance, and for 5
regulatory decisions as well.
6 This CNSC RIDM approach emphasizes that 7
PSA can be used to complement the DSA and other RIDM 8
key principles with due consideration of 9
uncertainties. We can go to the next slide, please.
10 This slide shows the overall approach of 11 risk-informed decision making which integrates 12 insights from the deterministic safety analysis, the 13 PSA, operating experience, and mandatory requirements.
14 The chart, if you just go to the next, it 15 will show the chart that describes the different 16 elements of the RIDM. The type of decisions and 17 candidates for RIDM in Canada is the same as in other 18 countries and regulatory bodies.
19 Here, I just named a few of them, and 20 these include, for example, the design, siting, 21 licensing, radiologic safety reviews and life 22 extension projects, and decommissioning. Next?
23 Yeah, this is the chart I was talking 24 about, I was referring to showing the different 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
22 elements that goes within the RIDM, and the same was 1
presented by our first presenter, Dr. Doug True. Next 2
slide, please.
3 So, here I'm showing the RIDM key 4
principles as introduced in the chart from the 5
previous slide. The key principles include first the 6
demonstration that the relevant legislation and 7
requirements are met.
8 That defense-in-depth is maintained is the 9
second key principle, and generally this principle is 10 assessed without invoking the PSA, but we know that 11 the PSA can offer some insights like the cutsets can 12 inform by revealing how many failures may occur before 13 we can get to core damage frequency or large range 14 frequency.
15 The third key element or the key principle 16 is about the safety margins should be maintained, and 17 there are also instances where the PSA can be used to 18 show that the safety margins are maintained.
19 The fourth principle is acceptable risk 20 impact, and here PSA can provide the calculation of 21 incremental risk such as the delta CDF and delta LRF, 22 and the last principle is to monitor the performance.
23 However, the challenge that we face with 24 these key principles, I think, is just to find is 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
23 there any balanced way for the consideration of these 1
key principles or is there any weighting factors that 2
we can use to consider the five elements? Next slide, 3
please.
4 In this slide, I will provide a brief 5
history of RIDM in Canada. So, traditionally decision 6
making has heavily relied on defense-in-depth and 7
expert judgment in the past, but in the last decades, 8
we show increasing use of PSA in Canada, use of PSA in 9
RIDM.
10 And this is mainly because back in 2005, 11 we issued at the CNSC two regulatory documents, one on 12 PSA and the other one is on the reliability program 13 for nuclear power plants, and in parallel, the staff 14 were developing a procedure, RIDM procedure based on 15 the CSA standard which is called risk management.
16 In 2018, the CNSC issued this regulatory 17 document, REGDOC-3.5.3, and in 2019, about three years 18 ago, the CSA standards issued CSA N290.19 which is 19 called RIDM for NPPs, building on staff procedures.
20 And right now currently, CSA is conducting 21 a survey on the use of this standard and to which 22 extent it helps the industry and the regulators in 23 applying this standard. Next?
24 Oh, I think I covered all of this bullet, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
24 yeah. I think I covered all of these bullets. Next?
1 Next? Next? Next? Next? Yeah, so now on this 2
slide, I'm showing the -- can you just press the next, 3
please? Next? So, yeah, and next, another one, 4
another time? Okay, perfect.
5 So, in this slide, I'm showing the staff 6
procedure for risk evaluation. This is based on the 7
risk tolerability scale for determining the risk 8
significance levels.
9 It's almost, I should say, deterministic, 10 and the risk evaluation is based on using matrices, as 11 you can see in the second chart. We chose the 12 consequence and likelihoods, and we defined the risk 13 significance levels dependent on the likelihood and 14 the consequences of each.
15 So, this procedure, which was based on 16 risk tolerability, was successfully used in the past 17 for the reclassification of what we call CANDU generic 18 safety issues.
19 And I'm providing just in the blue box 20 some examples, like for the reclassification of the 21 generic issue of pressure tube failure coincident with 22 moderator heat sink failure, which we call LOCA/LOMA, 23 loss of coolant accident and loss of moderator, safety 24 improvements for steam line breaks in multi-unit 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
25 nuclear power
- plants, and the large LOCA 1
reclassification for certain break sizes to beyond 2
design basis accidents. Next, please.
3 So, in this slide, the PSA use to support 4
RIDM, as a direct use of the PSA and the PSA result 5
that we get from the licensees, CNSC staff have 6
developed what we call the risk handbook tool which is 7
a web-based application.
8 In this tool, the PSA and reliability 9
program results and insights are used to risk inform 10 the licensing and compliance verification activities.
11 Other PSA uses to support RIDM include 12 risk management for outage planning for online 13 maintenance, for example, what we call risk monitor or 14 risk watch.
15 We use it also for life extension 16 projects. PSA can help identify safety improvement 17 opportunities if a plant needs to go through the life 18 extension.
19 We also use it for SAMG, severe accident 20 management guidelines development, emergency 21 preparedness drills and exercises because all of the 22 programs and diagnostic analysis are done using the 23 PSA and analysis derived from the level two PSA. Next 24 slide, please.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
26 So, in this slide, I'm going to develop a 1
little bit further on the CNSC risk handbook. First, 2
the purpose of this handbook is to support the 3
regulatory compliance program, focusing primarily on 4
applications for inspections, which means that this is 5
mainly developed to support site inspectors.
6 This tool is used to optimize the 7
inspection planning and improve efficiency. For 8
example, we get requests for inspections of, let's say 9
inspections for electrical, mechanical, or whatever 10 system qualification inspections, then we use the PSA 11 insights and results to provide a focused set of 12 equipment, or human actions, or specific hazard 13 information for the inspection purposes.
14 We do also use the PSA to evaluate 15 inspection results. For example, the site inspector 16 may have a sense of the risk, incremental risk if a 17 piece of equipment is taken out of service.
18 It also helps in the determination of 19 safety significance of operational events. If some 20 event happened, to have some sort of idea what is the 21 incremental core damage frequency or large range 22 frequency. Next slide, please.
23 This slide is about the emergency 24 mitigating equipment credits in the PSA. Emergency 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
27 mitigating equipment once again is known as FLEX in 1
the U.S.
2 Emergency mitigating equipment functions 3
that we have is first to prevent a severe accident.
4 The second objective is to repower instrumentation and 5
monitoring of critical safety parameters.
6 The third objective is to prevent severe 7
core damage. The fourth one is the in-vessel 8
retention of collapsed core, and the fifth objective 9
is to repower containment supporting functions. Next?
10 Next?
11 So, if we can see in the graph that I show 12 here, the three first objectives, they are applicable 13 to level one PSA, and the two last objectives, they 14 are applicable to level two PSA. Next slide, please.
15 So, the prerequisites for EME credits in 16 PSA, of course, as everything that appears in the PSA, 17 there should be a clear guidance to deploy emergency 18 mitigating equipment and decisions are also made 19 within the main control room or secondary control room 20 by authorized staff, because when the operation shifts 21 to the emergency operating center, probably to deal 22 with the human actions is more cumbersome in this 23 situation.
24 The challenges that we faced with 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
28 crediting the EMEs in the PSA is the first one is the 1
use of PSA models, with or without EME, for different 2
applications. Let's say if we want to compare against 3
safety goals, do we consider the improvements from 4
EMEs, yes or no?
5 The second question which is also 6
discussed now at the CNSC and with the industry is the 7
identification and classification of systems important 8
to safety. So, if we consider emergency mitigating 9
improvement as systems important to safety with all 10 reliability program that should be applied to it, yes 11 or no.
12 We have also the issue of EME credits in 13 multi-unit PSA, sequential EMEs or the FLEX equipment 14 for different units going into accidents, and there is 15 also the challenge of surveillance requirements for 16 EME because most of the time they are just industry-17 grade equipment. Next slide, please.
18 In this slide, I will discuss the benefits 19 and pitfalls of PSA. I think Dr. Doug True has 20 provided very insightful about the benefits of PSA.
21 The benefits, as we know, that PSA can 22 provide a rigorous and reproducible assessment of 23 incremental risk, delta CDF and delta LRF, compared to 24 risk significance levels which are based on subjective 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
29 judgment. Let's say we think that consequences are 1
high and the likelihood is low. PSA will provide a 2
very good quantitative assessment of the incremental 3
risk.
4 I'm just giving an example here. For 5
example, in our operating policies and procedures, 6
which is the same as tech specs in the U.S., we have, 7
in the traditional deterministic approach, if the 8
system redundancy is reduced, the repairs shall be 9
made promptly or other actions taken to ensure 10 adequate system reliability and capability.
11 So, if we want to interpret this 12 requirement, what do we mean by promptly? Does it 13 mean minutes, hours, days? So, whereas the PSA, we 14 can calculate, if we have the time at risk or the 15 completion time, we can calculate exactly what's the 16 incremental risk and compare it to the guidelines.
17 The pitfalls of the PSA is most of the 18 PSA, sometimes there is a lack of cause and effect 19 relationship in some cases. For example, if an 20 emergency core cooling valve is passing, this is not 21 modeled in the PSA.
22 If we want to change the methodology for 23 trip set points determination, the PSA may not be a 24 useful tool. For piping inspections, you need to do 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
30 some work on PSA to help you use PSA.
1 There is also the issue of uncertainties 2
in the PSA, and we always caution about the over 3
reliance on PSA to address all safety concerns. PSA 4
is not the solution for every issue that we may have.
5 Next slide, please?
6 This is the summary of my presentation.
7 I think in summary I would say that guidance is needed 8
on how to assess the impacts on defense-in-depth and 9
safety margins, as well as on benefit cost analysis.
10 If we don't have clear guidance on how to 11 assess or how PSA can help assess the impacts on 12 defense-in-depth and safety margins, we may not have 13 a clear and reproducible scheme for RIDM.
14 I think we all understand that PSA is a 15 valuable tool to complement deterministic and expert 16 judgment in the RIDM process.
17 Uncertainties, of course, should be 18 accounted for, especially when the PSA is also close 19 to the acceptance guidelines.
20 And I believe that development of RIDM 21 guidance will allow a transparent and reproducible 22 process for regulatory decisions. Next?
23 So, this is the end of my presentation.
24 Thank you for your attention.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
31 MS. KOCK: Thank you, Smain. Okay, I 1
think now we are going to go to polling questions, so 2
we're going to do -- we have a total of four questions 3
and we're going to do the first two now to kind of 4
catch up.
5 So, the first question is about your 6
experiences in using risk-informed decision making.
7 How much have you used it and how has it benefitted 8
your work? So, I'd really like to hear from you and 9
maybe join the discussion at the end. We can get some 10 feedback on this.
11 So, oh, there's the results already. How 12 do you use risk-informed decision making as part of an 13 official technical process? The vast majority of you 14 have and just a few, 13 percent, have not.
15 So, it would be interesting to hear 16 experiences for those of you who have, how you got to 17 the point of using those risk-informed decision-making 18 tools that you've talked about, and if you haven't, 19 maybe we'll hear a little bit in the discussion 20 session about why some folks have not used it, so 21 that's kind of an interesting result there and I see 22 it's fluctuating a little bit.
23 We can go onto polling question two. So, 24 polling question two is going to have to do with how 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
32 risk-informed decision making has benefitted your 1
work. And I think we've already heard that there are 2
differences of opinions on the benefit of risk-3 informed decision making, so it will be interesting to 4
see the results here.
5 Okay, so we're still fluctuating a little 6
bit, but it looks like most of the folks in the 7
audience have the opinion that risk-informed decision 8
making has benefitted their work, and again, a smaller 9
percentage says it hasn't.
10 So, it will be interesting to hear as part 11 of the discussion at the end where perhaps it has 12 significantly benefitted our work, and if it hasn't, 13 why not, or maybe there are certain areas where it's 14 not as beneficial, so that could be an interesting 15 result.
16 We're going to move on now to the second 17 polling questions, so there's an A and B aspect to 18 those as well. The next two questions have to do with 19 your opinion about how both the industry and the NRC 20 or other regulatory agencies are using risk-informed 21 decision making. 2A here is a question about whether 22 industry is doing enough with regard to risk-informed 23 decision making.
24 Okay, and the results are fluctuating just 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
33 a little bit at this point. So, it looks like the 1
majority opinion here is that there's more to do on 2
risk-informed decision making, which really isn't 3
surprising because I think we've already heard it's a 4
complex topic, and there's about 20 percent of you who 5
think that industry has done just the right amount 6
there.
7 Okay, and then the last polling question 8
for now is going to have to do with how regulators 9
like the NRC use risk-informed decision making and 10 whether there's been enough there.
11 So, kind of interesting, the same result 12 here for regulatory agencies as we saw for the 13 industry. The vast consensus seems to be that there's 14 more to do, and just to me personally, that's not 15 surprising.
16 I think we've made great progress, but, 17 you know, given some of the uncertainties and 18 challenges that both Doug and Smain just talked about, 19 it's not really surprising to hear there's more to do.
20 Okay, so that's interesting and maybe we 21 can build off the answers to some of those questions 22 as we move along in the presentations, but at this 23 point, we're going to go to our next speaker who is 24 Mark Thaggard, who is going to be talking about the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
34 safety marker study. Mark?
1 MR. THAGGARD: Good afternoon. The NRC 2
has a long history of incorporating risk insights into 3
its decision making. Some of that you're going to 4
hear about from Mike in the following presentation.
5 However, the focus of my presentation is looking at 6
the potential impacts of safety from the use of risk-7 informed decision making.
8 I plan to go over some insights from a 9
fairly comprehensive assessment conducted by the NRC 10 staff recently to look at safety trends within the 11 U.S. nuclear industry.
12 The staff looked at a broad range of 13 measures and markers to see if safety was trending in 14 a positive direction, negative direction, remaining 15 steady, or were indeterminate.
16 Keep in mind that this was a trending 17 assessment and that the focus was in looking for 18 trends. There was no attempt to make a comparison 19 against a particular performance metric.
20 The staff also made no attempt to 21 ascertain or infer compliance or noncompliance with 22 our regulations. Next slide.
23 NRC maintains numerous safety activities 24 and programs to monitor, assess, and reinforce safety.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
35 For the staff assessment, as noted, the staff looked 1
at a broad range of these activities to get a 2
comprehensive safety picture of U.S. commercial 3
nuclear power plants over the last 20 to 30 years, a 4
time period that included the issuance of the 5
Commission's probabilistic assessment policy statement 6
issued in the 1990s.
7 Some of the measures and activities 8
reviewed included looking at the number of scrams, the 9
number of license event reports, inspection findings, 10 performance indicators, risk measures, and insights 11 from studies such as the State-of-the-Art Consequence 12 Analysis or SOARCA, just to name a few.
13 The staff assessment focused only on 14 safety and did not look at other possible trends such 15 as economics or electric output. Next slide.
16 A key consideration in the staff's 17 assessment is that there are several aspects of trends 18 that can affect the overall conclusion. One clear 19 example is the time frame you consider. Some 20 performance measures might show improvements over the 21 last 30 years, but may not show an improvement if you 22 only looked at a 20-year period.
23 Another consideration is the advancements 24 in our technical and modeling capabilities. Some 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
36 performance measure improvements might be a result of 1
our advancement in these areas and not an actual 2
safety improvement.
3 Considerations such as these required care 4
in making a definitive conclusion on the trending 5
direction of some activities. Accordingly, for a 6
number of activities, the conclusion was indeterminate 7
even when the staff could see a trend. For a number 8
of the safety measures, measure and performance came 9
down to a collective engineering judgment. Next 10 slide.
11 One of the considerations discussed in the 12 last slide is the time period chosen impacts the 13 overall conclusions. Looking at the period between 14 1988 to 2000, the staff saw several performance 15 measures that clearly showed safety improvements.
16 This period saw many changes in the U.S.
17 nuclear industry, including implementing the Station 18 Blackout
- rule, the maintenance
- rule, and the 19 individual plant examination program. Next slide.
20 This slide highlights an example of a 21 performance measure that showed clear safety 22 improvements during the 1988 to 2000 time period. If 23 you look at the chart on the left, it shows scrams 24 while critical per unit over time, which shows a clear 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
37 positive trend over this time period. Please note 1
that the reduction in the number of scrams over time 2
is viewed as positive in this context.
3 The chart on the right reflects the 4
difficulty in ascertaining safety improvements even 5
when the data clearly indicate that things are moving 6
in the right direction. It shows significant events 7
per unit over time, which also shows a clear positive 8
trend over this period. Again, a reduction in the 9
number of events over time is viewed as positive.
10 Even with the apparent positive trend, the 11 staff couldn't definitively determine that this was 12 reflective of safety improvements because the Agency 13 changed its reportability guidance during the period 14 which could have affected the trend.
15 However, taken together, the two graphs 16 help support an overall conclusion that nuclear plant 17 safety improved during this period. For both of these 18 performance measures, the positive trend after 2000 is 19 not as clear. Next slide.
20 As stated, when looking at the last 20-21 year time period, the performance measures do not show 22 the same level of safety improvements. There is still 23 a trend in the right direction, but we do not see the 24 same level as seen during the '90s.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
38 There are several possible reasons for 1
this. One is the safety improvements from the '90s 2
may have made further safety impacts less obvious, 3
that is the law of diminishing returns.
4 This last 20-year period does include 5
several changes that have a positive safety impact on 6
the industry, including the reactor oversight program, 7
the B.5.b security compensatory measure requirements, 8
use of FLEX, and design enhancements such as the 9
reactor cooling pump shutdown seals.
10 Therefore, another possible reason that we 11 see less improvements in the last 20-year time period 12 is that safety significant changes may not be fully 13 realized. Next slide.
14 There are several different ways to look 15 at all of the performance measures. These are six 16 categories of performance measures used by staff. As 17 you can see from the different category activities, 18 some were more easy to ascertain a quantitative trend.
19 Next slide.
20 Overall, the staff looked at roughly 50 21 performance measures. These roughly 50 performance 22 measures, they clearly show a positive trend. Let's 23 look at each of these.
24 The number of scrams has dropped to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
39 historically low levels. The total number of scrams 1
and scrams per unit showed a 20-year decrease in 2
trends, while the number of plants with zero scrams 3
showed a positive 20-year trend.
4 The number of reactor inspections has been 5
trending steadily downward when looking at two 6
different time periods. The current rate of all 7
precursors exhibits a statistically significant 8
decrease in trend for the 2000 to 2020 time period.
9 The collection radiation dose per unit 10 shows a 20-year decrease in trend. The number of 11 greater than green findings show a decrease in trend 12 since 2014.
13 A micro level fleet average internal event 14 core damage frequency or CDF based upon the Agency's 15 risk models is trending downward.
16 Installation of the reactor coolant pump 17 seals in Westinghouse PWR nuclear power plants has 18 resulted in a reduction in both CDF and large early 19 release alerts in our risk models.
20 Lastly, there is a highly statistically 21 significant decrease in trend in the frequency of 22 overall loss of offsite power.
23 Only one performance measure showed a 24 clearly negative trend, which is loss of offsite power 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
40 recovery time. This marker showed a statistically 1
significant increase in trend. Next slide.
2 Of the roughly 40-some remaining measures 3
looked at by staff, nine were viewed as positive, but 4
the assessment was more qualitative. In the interests 5
of time, I won't go over the nine which are listed on 6
the slide.
7 The remaining 30-plus measures were either 8
steady or staff could not say definitively one way or 9
the other the trending direction. An example where 10 staff could not make a determination was the number of 11 license event reports even though it showed a positive 12 trend as I previously noted.
13 There were a number of the 30-plus 14 performance measures like this that the staff 15 concluded as indeterminate. Next slide.
16 Several points noted by the staff in their 17 assessment were that there is a reduction in risk over 18 the 30-year time period as measured by calculated 19 average core damage frequency.
20 However, they noted this only considers 21 the contribution of risks from internal events.
22 External event hazards could in some cases 23 significantly add to plant risk. The staff also noted 24 a reduction in performance issues.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
41 Based upon the staff's assessment, there 1
is some margin between calculated risk and the NRC's 2
safety goals. However, it is important to note that 3
both uncertainty and external hazards need to be 4
considered in determining the amount of the margin.
5 Next slide.
6 In conclusion, based on the performance 7
measures looked at by the staff, the use of risk 8
insights in decision making at the Agency is having an 9
overall positive impact on the safety of the industry.
10 Notwithstanding the need for consideration 11 of the influence of external hazards in our assessment 12 and accounting for uncertainties, there have been 13 clear safety improvements.
14 The performance measures show a more 15 prominent trend in the 1990s, but a more gradual trend 16 after 2000. Next slide.
17 This concludes my presentation. We'll 18 save questions for the end, all right. I'll turn it 19 back to you, Andrea.
20 MS. KOCK: Thank you, Mark. I just want 21 to put a plug in for everybody to get their questions 22 in. We have one more speaker to go, but obviously you 23 have a little bit of time, but if you have questions, 24 go ahead and submit those and we'll cover those during 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
42 the discussion period.
1 Now we're going to go to our third polling 2
question which also has two subparts, and these 3
questions have to do with your opinion on the state of 4
industry safety, nuclear safety today versus ten years 5
ago and the role of PRA.
6 So, the first question, in your opinion, 7
is the nuclear industry safer than it was ten years 8
ago?
9
- Okay, it looks like there's less 10 fluctuation now in the numbers, but the vast majority 11 of you feel like the nuclear industry is safer than it 12 was ten years ago, and then there's a small percentage 13 of folks who are either in the no or depends category.
14 Okay, and then 3B, polling question 3B is 15 about the role of PRA with regard to industry safety 16 and whether PRA has played a role. So, in your 17 opinion, has the use of PRA and risk-informed decision 18 making made the nuclear industry safer today than it 19 was ten years ago?
20 Okay, those results look like they've kind 21 of stopped fluctuating. So, it looks like these 22 results follow closely the question right before this 23 on the state of nuclear safety in that the vast 24 majority of you feel that PRA and risk-informed 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
43 decision making has contributed to safety, and then 1
there's a smaller percentage who either answered no or 2
maybe. Okay, so that was interesting.
3 We're going to move onto our last speaker, 4
Mike Franovich, and he's going to talk about safety 5
improvements using risk insights, so onto you, Mike.
6 MR. FRANOVICH: Thank you, Andrea. If we 7
can have my first slide, please? Next slide? All 8
right, I'll try to get us back on track a little bit.
9 We're running a little bit behind schedule here.
10 So, what I'm going to cover in my 11 discussion is to give some context and a little bit of 12 historical background of issues or policies that are 13 active today and still affecting and shaping the way 14 we do our business.
15 And I also want to give you a few 16
- tangible, more contemporary examples of where 17 advancements in risk assessment and other engineering 18 analysis has actually helped enable us to make better 19 decisions in some additional work going on, in 20 particular with weather events, HEAF events, and also 21 in new reactors, but more so what I'm going to talk 22 about is how it's shaping our work in the operating 23 reactor side, in particular, licensing.
24 And then a shameless plug here for an 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
44 Agency initiative that I am a very strong proponent 1
of, and that is the Be Risk Smart framework. It's an 2
initiative that is looking to try to unify and provide 3
more uniform application of risk concepts in all of 4
our work, not just the technical work, but some of the 5
more corporate support work, and using the risk 6
triplet, and looking at the rewards that may come out 7
of doing various projects, just not the downsides of 8
them. So, if we could go to the next slide?
9 Okay, meeting the challenge of becoming a 10 more modern risk-informed regulator, we could spend a 11 lot of time on that topic alone. The journey is 12 continuing.
13 A couple of insights I want to share with 14 you is that the technology in terms of risk assessment 15 does continue to mature. Risks are dynamic. We do 16 have to have effective risk management by maintaining 17 our vigilance and assessment of operating experience.
18 The technology and insights do complement 19 the traditional defense-in-depth framework that we 20 have used for many, many decades to achieve an 21 acceptable level of residual risk, not zero risk, but 22 residual risk, and this journey does continue. So, if 23 we can go to the next slide?
24 I mentioned the Be Risk Smart framework 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
45 and a couple of other things that are going on in the 1
Agency. Really what we've seen is a great push here 2
in the last few years to try to come up with a more 3
uniform application of guidance that crosscuts 4
different offices and business lines.
5 Really, looking at the risk proposition, 6
not just the negative side of what are the risks of 7
certain decisions and not just all of the plants 8
themselves, but also different projects, but what are 9
the rewards?
10 What are the gains that we may gain by 11 pursuing certain activities in the Agency? And try 12 doing them in a way that shows that we are getting a 13 good return on our investment.
14 And if we are applying these approaches, 15 in particular, the Be Risk Smart framework, it should 16 enable us to become more agile and adaptive when we 17 look at different projects within an office or across 18 offices in the NRC.
19 So, I'm going to give you a flavor of a 20 few of those things that are going on that might 21 provide more insight into what's going on inside the 22 NRC. You can go to the next slide.
23 Now, Mark mentioned that I would talk 24 about some of the major policies that are affecting 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
46 our thinking in the Agency. My point of this slide is 1
not to go into a deep dive of these particular 2
policies, but to share with you that they are active.
3 They are live. They do shape the way our work is 4
going on today.
5 And periodically we need to go back and 6
refresh our memories as to why we have these policies, 7
in particular, starting in the '80s with the severe 8
accident policy statement on how we treat severe 9
accidents, also some anchoring guidance and 10 expectations out of the safety goal policy statement 11 in terms of the qualitative safety goals, as well as 12 the QHOs.
13 And then moving forward in time to the PRA 14 policy statement where there's one particular aspect 15 I want to highlight out of that policy statement 16 that's still alive and well, and that is we should be 17 using PRA technology to the extent it increased in all 18 regulatory matters as supports by the state-of-the-19 art, and it goes onto say to complement that of the 20 defense-in-depth philosophy that we use in the Agency.
21 And my next few examples are going to 22 illustrate a little bit more how the state-of-the-art 23 or what we would say today, state-of-the-practice, has 24 actually advanced quite a bit.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
47 And then lastly, I want to note a very 1
important Commission decision that was made in 1999 2
based on a Commission paper, a White Paper sent up in 3
1998, and that is an overarching framework for how we 4
are to use risk-informed, performance-based 5
regulation, not just in operating reactors.
6 If you take a close look at the paper and 7
the SRM, it actually speaks to our work in the 8
materials area as well as the fuel facilities area.
9 That paper is alive and well and is germane to a topic 10 I'll talk here about when I get toward the end of my 11 discussion. Next slide. If you can advance to the 12 next slide, please?
13 Okay, my screen is frozen, so I don't know 14 if you all are seeing the next slide, but if not, I 15 will try to use my local copy. Okay, it looks like we 16 have a little bit of a technical problem going on at 17 the moment, that all the slides are frozen.
18 So, I'm just going to speak to what would 19 be on the next slide that you would be seeing and that 20 is the integrative decision making principles that 21 Doug spoke of, and that there are five key principles.
22 I'm not really intending to go into depth 23 of each of those principles of defense-in-depth, and 24 safety margins, and performance monitoring. I will 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
48 highlight that the objective also is that the risk 1
changes for facilities should be small and within the 2
safety goals.
3 But really, to me, it's something more 4
important about how we go about making these 5
integrated decisions. You can take each one of those 6
decision-making principles and do them in isolation.
7 And that's been a real challenge in the 8
Agency to try to break down some of the silos and do 9
more integrating through what we call integrated 10 review teams, where you start a project with the 11 principles in mind, not trying to bring them together 12 through a long review process, and seeing how they all 13 fit, and that they really are not mutually inclusive 14 of each other.
15 To me, the integrated decision-making 16 process really is a place for critical thinking.
17 What's going on in each one of those decision-making 18 boxes in terms of defense-in-depth?
19 The plants are not static. They have 20 changed. Many capital improvements have been made.
21 Where are the margins, both in physical margins and 22 analytical margins? So, it's important that we have 23 a culture that drives for having those conversations.
24 And I apologize for whatever the technical 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
49 difficulties are. But we're not -- I'm not able to 1
see the slides either on my end so I will be speaking 2
verbally. I hope you all can still hear me. I will 3
just speak to describing my -- for sake of time.
4 There's another process. Oh, here we go.
5 Thank you very much. Looks like the technical problem 6
has been resolved. If we could go to the next slide.
7 And these integrated decision-making 8
principles out of Reg. Guide 1174 -- we're on Rev. 3, 9
by the way, even though this Reg. Guide came out in 10 the 1990s -- we do use it to influence and help guide 11 a number of other processes we have and one of those 12 processes I'm going to talk to you is about LIC-504.
13 That is an agency or NRR process.
14 If you'll recall, the Davis-Bessie event 15 from 2002 -- we're actually, I believe, close to the 16 20th anniversary -- important lessons learned. You 17 need to have a process for not only making integrated 18 decisions but also documenting the basis for your 19 decisions.
20 So we had a number of corrective actions 21 in the agency we took to create such guidance. This 22 particular guidance is used very much in the agency 23 today. It's a mature process.
24 We're currently on Revision 5. In the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
50 latest revision, we did add other considerations. For 1
example, if there are actions that are warranted what 2
would be some of the risk offsets -- for example, any 3
additional exposure to radiation workers. Those 4
should be active considerations and what we are 5
looking at in terms of options.
6 And also, we included a section on risk 7
decision or, rather, decision-making biases. There 8
are some areas where you can get into groupthink, for 9
example, and we need to be conscious about those when 10 you're in that decision-making box or curve that 11 you're not susceptible to those biases or you try to 12 minimize them.
13 It is a two-step process where we look at 14 immediate safety issues, first, if we need to take 15 prompt regulatory action, and then if not we'll look 16 at the longer-term actions in any second step process.
17 And we are using the Be riskSMART 18 framework, which is in a NUREG that came out, 19 NUREG/KM-0016. I would encourage you all to read it.
20 It's a very easy read. Again, it's not unique to PRA 21 but it's promoting more of the risk triplet across the 22 work that we do in the agency.
23 Can we go to the next slide?
24 All right. The first example I want to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
51 give is back in August 10th of 2020, we had a Derecho 1
event that hit the Duane Arnold plant, a very powerful 2
line of wind, a storm that came through the plant and 3
the plant performed as expected, and so to the 4
operators this was very good news.
5 But we did identify a few unexpected 6
conditions that happened. Even though there was loss 7
of outside power, which is not unexpected, there was 8
an issue of potential combined event and that the 9
central service water system did have some degradation 10 from debris that was on the river that came into the 11 intake system.
12 So we did use the LIC-504 process. A few 13 important takeaways came out of that. Looking at 14 similarly situated plants and those similarly situated 15 plants had done some upgrades, for example, alternate 16 cooling for diesel generators. If they're dependent 17 on a central service water they have alternate means.
18 We also found now with greater capability 19 to quantify the benefits of flex that that is actually 20 quite a difference maker in terms of risk impact.
21 Depending on the nature of the plant and its location 22 and its site-specific hazards, it is quite influential 23 on the risk results.
24 And we applied the Be riskSMART framework, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
52 and the big takeaway there was even though we 1
recommended a generic communication in terms of 2
information notice, we also used the T in the teaching 3
elements to get the word out and conduct a webinar and 4
a multi-organizational panel to discuss the insights 5
that came out of this study of various sites that we 6
did take a look at that have some susceptibility to 7
Derecho.
8 We did also find that the risks were not 9
trivial but nor were they in a matter of a level where 10 we needed to do some type of mandatory backfit 11 analysis. So they were -- they did have import and we 12 thought it was important to share that information 13 with industry and the rest of the NRC.
14 We do have significant turnover, a 15 turnover in the industry as well as in the NRC. So 16 the T can also be viewed as knowledge transfer.
17 If we could go to the next example.
18 Doug had mentioned a number of capital 19 improvements that plants have made, particularly for 20 NFP 805 implementation with alternate seal injection, 21 backup diesel generators, and so forth. One of the 22 hazards the plants did do additional measures for is 23 to provide additional protection for high-energy 24 arcing faults.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
53 We more recently have come to discover 1
there's a little bit more aluminum in these electrical 2
components than originally envisioned and that 3
aluminum might be a little bit more reactive than 4
5 And so what does that actually mean if the 6
hazards are slightly different in terms of risks? So 7
we're currently undergoing a LIC-504 evaluation and in 8
that first step, which we did complete, we used more 9
of a defense in depth qualitative risk thought with 10 the defense in depth in mind, looking at the plants 11 already have a level of protection for heat but they 12 also have protection from post-9/11 measures for 13 dealing with large fires and explosions.
14 So these are other qualitative 15 considerations we need to bring into our decisions.
16 The work is ongoing and there's a massive effort, 17 actually, been going on between the NRC and EPRI for 18 years to build more consensus models to help bring in 19 the more advanced fire modeling capability as well as 20 the advances in the PRA technology.
21 And then my last example -- if we can go 22 to that slide -- is on new reactors. I will not go 23 through all this on NuScale specific. You can read it 24 at your leisure.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
54 We had a very important takeaway and a 1
commission direction, which had to do with the single 2
failure criterion and the treatment and use of risk 3
and those kind of decisions.
4 But more importantly, there was an insight 5
or direction from the commission as a gentle reminder 6
that we are to apply risk-informed principles in our 7
decisions and it actually looped right back to the 8
1990s policies that I mentioned that are still active.
9 So we were kindly reminded as a staff you 10 need to continue to be applying these principles in 11 your work and so that's what we're doing. We're 12 continuing on that journey.
13 And if we can go to my last slide, I'll 14 wrap it up.
15 A couple takeaways. I want to leave you 16 with a couple of thoughts. We continue to support 17 risk-informed decision-making through our programs.
18 Risk is permeating a lot of our work. It touches a 19 lot of areas in the agency.
20 We are also using it in a manner that 21 still complements defense in depth and safety margins 22 and we're also including consideration for enterprise 23 risk management, which is a federal government wide 24 requirement on how we manage our projects and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
55 activities.
1 We do also take that into consideration as 2
well, and we're also leveraging this new Be riskSMART 3
framework with PRA technology and also looking at 4
other business lines to see how we can help out.
5 And with that, I will turn it back to 6
Andrea.
7 MS. KOCK: Great, Mike. That was a really 8
great summary of the history of risk-informed 9
decision-making at the NRC and some other things we 10 have going on. So thanks for that.
11 So I'm running a little bit short on time.
12 So we're going to quickly go to the fourth polling 13 questions, 4A and 4B, and then we're going to jump 14 right into questions so we have plenty of time for the 15 discussion section.
16 So 4A and B have to do with the use of PRA 17 and risk-informed decision-making in our work, and the 18 first question is PRA and RIDM create work 19 efficiencies and aid in correctly focusing priorities.
20 True or false?
21 That looks pretty steady. It looks like 22 the vast majority of you feel that PRA and risk-23 informed decision-making do create work efficiencies 24 and help us focus our priority.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
56 It'd be really interesting to hear from 1
some of the folks who are saying maybe and depends.
2 I would be interested to hear about that. Maybe we 3
can get to that during the Q&A.
4 Okay, and 4B. 4B is PRA and risk-informed 5
decision-making create efficiency benefits when 6
interacting with regulators.
7 So it looks like we're easing out there.
8 A majority of you say sometimes. It would be 9
interesting to hear when those sometimes are and what 10 the considerations are, and whether PRA and risk-11 informed decision-making create efficiency when 12 interacting with your regulator.
13 That's interesting. And then there's a 14 smaller percentage of people say always and a very 15 small percentage say never.
16 Okay. So with that, I think we're going 17 to go right into the discussion section of the panel 18 session. We have a little over 15 minutes left.
19 And so the first question I have is 20 actually for -- I heard it was Dr. True. So first 21 question for Dr. True.
22 MR. TRUE: No Dr. here.
23 MS. KOCK: No Dr. Okay. I thought 24 somebody called you Dr. earlier. So just Doug.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
57 MR. TRUE: Smain was just being very nice.
1 MS. KOCK: Oh, okay. Great.
2 Okay. So first question for Doug. In 3
recent years, there has been an impression that PRA 4
and risk-informed decision-making are too complicated 5
and costly.
6 Do you believe the benefits of risk-7 informed decision-making have been exhausted in the 8
current framework and, if so, what needs to be fixed?
9 And while I start with Doug, I'd also 10 invite other panel members to jump in after Doug 11 finishes. Thanks.
12 MR. TRUE: Thanks, Andrea.
13 I got a number of thoughts on this and I 14 think it actually ties somewhat back to some of the 15 answers to the questions we were asking.
16 So I think people generally felt like 17 there was more that could be done and I think that's, 18 certainly -- I, personally, believe that's certainly 19 true.
20 But, I think, also the last question, I 21 think it was, that talked about efficiencies, 22 sometimes we do get kind of bogged down and I have a 23 few thoughts about that.
24 I think part of this is we're on a journey 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
58 here as the NRC tries to become a more modern risk-1 informed regulator.
2 As Mike pointed out, there's a lot of work 3
going on to try and bring people up to the same 4
understanding of what risk informing really means.
5 This is true on the industry side as well as the NRC.
6 It's not unique to the NRC. That this industry was 7
founded on sort of deterministic approaches to things 8
but bringing risk into this is a foreign concept to a 9
lot of people that have never been exposed to it.
10 I think there's more work to be done to 11 educate people to understand what risk is and what it 12 isn't, understand what its limitations are and what 13 its effective uses are, and that's something that we 14 all need to undertake.
15 I do think, secondly, on this subject, 16 that there's been a tendency sometimes to focus way 17 too much on the numbers.
18 We get all bogged down on decision 19 thresholds and I think that it really should be more 20 about how we understand the plant from the PRA 21 perspective. Numbers are important but there are 22 uncertainties in the numbers. So getting too bogged 23 down in the numbers is not helpful and can often 24 stretch things out unnecessarily.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
59 And third, you know, do it on some of 1
these applications that are more ambitious. For 2
example, risk-informed completion times require a more 3
extensive and expansive PRA because you're asking a 4
more difficult, broad question. That makes it a 5
little bit more costly, and I think that's been a 6
challenge, too.
7 But we still -- we still continue to see 8
places where we think there's opportunity for more 9
risk-informed thinking. We're in the midst of a 10 discussion about tornado impacts on cast loadings that 11 we think has a low safety significance and merits 12 being dispositioned that way.
13 We also have, you know, other issues going 14 on in the industry where we have successfully used the 15 low safety significance process to decision-making.
16 So I think there's more opportunity and I think 17 education will go a long way to help us moving forward 18 on that.
19 MS. KOCK: I just invite any of the other 20 panelists to weigh in on that question.
21 MR. FRANOVICH: I would say I agree with 22 Doug completely in his assessment. I would add that 23 I think you see a little bit of a lagging effect going 24 on because there's an ongoing shift to move to the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
60 more advanced risk management programs.
1 What I'm speaking of, more specifically, 2
it started with the surveillance frequency control 3
program.
We also have 5069 and the risk 4
categorization of SSCs and then the risk-informed 5
completion times.
6 For the surveillance frequency control 7
program, I think we're right now at almost 100 8
percent. The plants have been authorized to use that 9
program, which started with the first plan, I think --
10 I believe it was Limerick in 2008 as a pilot.
11 But we are making significant progress in 12 approving the 5069 applications that are coming in --
13 I think we're well over 20 now -- as well as the 14 completion time program. And as those reviews go on, 15 eventually they get authorized and they'll get 16 implemented. But it's in a staggered way.
17 And so when they get authorized and we see 18 more broad use of them, I think you're going to feel 19 more of the work that's going on now that doesn't 20 maybe resonate or feel like we're doing enough in the 21 way of risk informing.
22 But there is a lot more room that can be 23 done in a lot of other domains. One of them that 24 we're actually trying to do a little bit is in the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
61 space of treatment of the alternate source term.
1 Can we use some graded approaches and 2
thinking in that space since we have collected a lot 3
of engineering experience, in particular, from seismic 4
and seismic PRAs?
5 That's just one small area. But there's 6
a lot of other domains where, I think, we can actually 7
do much more, and then there's the whole area of the 8
new and advanced reactors, which is a whole another 9
topic area.
10 MS. KOCK: Okay. I'm going to move on to 11 the second question. The second question was for 12 Smain but, again, I'll invite other panelists to jump 13 in on this one.
14 First of all, a question -- a statement.
15 Great presentation, Smain. So thanks.
16 MR. YALAOUI: Thank you.
17 MR. KOCK: And then the question. In your 18 experience, what is the fundamental difference between 19 defense in depth and safety margins in traditional 20 engineering and risk-informed processes? Is it the 21 same or is it balanced in a different way?
22 MR. YALAOUI: Thank you very much. I 23 think this is a very good question.
24 I think we all know that defense in depth 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
62 and safety margins are part of the traditional 1
deterministic approach and these are fundamental 2
concepts.
3 But is this -- how this is balanced within 4
that idea? Once again, this is a very tricky 5
question. I would say no, a straightforward answer to 6
that. I think it's very difficult, I'd like to say.
7 As in one of my slides, I say there is no waning 8
factors to balance the different key principles of the 9
RIDM.
10 But once again, how this -- how this is 11 balanced in RIDM to think that defense in depth is 12 something that decision-makers need to consider. It's 13 not always easy to say, for example, to which extent 14 or how much safety margins erosion is acceptable.
15 These are not easy questions.
16 I think we go with different case -- on a 17 case by case basis and it needs to have all the 18 specialists from safety -- deterministic safety PSA, 19 engineering judgment and other specialists to have a 20 good picture for the risk decision-makers.
21 So I leave it to other panelists if they 22 have any other thoughts to provide.
23 MS. KOCK: Okay. If not, I'm going to 24 move on to the third question. This question was for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
63 Mark but, again, others can jump in.
1 How did the trends that you described, 2
Mark, compare with overall consideration of defense in 3
depth of the safety margin? Even if some trends go 4
up, is defense in depth and safety margins being 5
impacted?
6 MR. HAGGARD: Well, so during the training 7
assessment, we didn't consider the defense in depth 8
and safety margin. I think one of the things I 9
started -- when I started the presentation, I mean, I 10 made a comparison to risk-informed decision-making 11 primarily because the defense in depth and safety 12 margins are kind of like fundamental.
13 They're always there. So the changes that 14 we have enacted over the years have been primarily in 15 instituting risk-informed decision-making.
16 So I don't think the trends -- I don't 17 think that -- if I'm understanding the question 18 correctly, I don't think the training assessment that 19 we looked at -- I don't think it had -- took any 20 consideration in terms of safety margin and defense in 21 depth because those are kind of like baseline.
22 They're always there.
23 Obviously, if some of the trends go in the 24 wrong direction, it's going to impact the amount of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
64 margin we have, you know. But we didn't fundamentally 1
look at that with the exception of looking at risk.
2 We did a comparison looking at the amount of margin in 3
terms of the risk numbers that we were looking at.
4 And, obviously, if that risk number -- if that goes up 5
then you would see an impact on the safety margin, if 6
I'm understanding the question correctly.
7 MS. KOCK: I'd just ask if any other panel 8
members have anything else to add.
9 MR. TRUE: Yeah. I'd just add that I 10 think, Mark, in your presentation, you talked about 11 how the B.5.b and FLEX were considered.
12 Those are new levels of defense in depth 13 that we didn't have back 10 years ago or 20 years ago, 14 and are a reflection of actually increases in defense 15 in depth, and safety margin is a little bit harder to 16 measure. You know, a well done PRA can often do that 17 but sometimes even has difficulty doing that as well.
18 But I think we haven't seen significant 19 reductions in defense in depth in applications we have 20 done. So I think it's increased net or where we were 21 10 or 20 years ago. My take.
22 MS. KOCK: Okay. Anything else on that 23 question? If not, we're going to move on to the 24 fourth question.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
65 Mike, this one was for you. But I really 1
think it applies -- I think any of the panelists may 2
have reflections on this one. So I'll start with you 3
and then just let everybody else add.
4 So the fourth question is, it is true that 5
risk-informed decision-making has expanded. However, 6
in recent years, it seems that either NRC or industry 7
have pulled off in this expansion on areas like 8
physical security and others.
9 Do you think there's a slowdown due to a 10 blockage that needs to be overcome or is the low-11 hanging fruit no longer available?
12 MR.
FRANOVICH:
So that's a
very 13 interesting question. I think the -- I would say 14 there's still active work going on in deploying --
15 reviewing and deploying, approving the programs that 16 are coming in that I mentioned already to three 17 advanced risk management programs.
18 And so it may not look like there's a lot 19 of work going on but I would say there's a lot of 20 production work, and when you look at what's probably 21 taking the oxygen out of the room, on a lot of the 22 operating reactors work it's that of the work of the 23 advanced reactors and the licensing modernization 24 project, which is really strong, more PRA technology 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
66 dependent framework.
1 There's been some shift in that area. But 2
on the operating fleet, you see more of the 3
production. We have had interactions quite a bit with 4
the owners groups, in particular, in a few projects 5
that are striking some interest.
6 One of them is, in particular, is there a 7
possibility of looking at how a
licensing 8
modernization project framework could be applied to 9
the operating reactors -- the LMP to operating 10 reactors -- and leveraging some of the insights that 11 are coming out of our Level Three PRA project, which 12 is ongoing work.
13 There is some interest there how that 14 might shape and level review and work in Chapter 15 15 space. That's still sort of in its infancy as a new 16 concept.
17 We do have other work going on, for 18 example, the risk-informed process for evaluations --
19 RIPE. We do have the first submittal in house.
20 That's being actively worked on right now.
21 You know, I think when you just start a 22 program like that you see where it goes. But that, 23 too, is sort of at the beginning phases. So there's 24 still a zeal. I definitely see a zeal there.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
67 And if I listen to my colleagues over in 1
the security arena, they have some -- done some recent 2
work in terms of off-site response capability, 3
leveraging FLEX and risk insights. There's still work 4
there.
5 But I would say it may not garner the 6
visibility that it once had. I mean, our attentions 7
have moved. I'm not judging this one way or another.
8 I just observe these kind of dynamics and the shift in 9
a lot of work with the advanced new reactor designs so 10 it maybe just appears that things aren't as active in 11 the risk side.
12 But I think they're -- in my perspective, 13 they're very active.
14 MS. KOCK: Okay. Other perspectives on 15 that one? And I think this will probably be our last 16 question. So if others have perspectives on that --
17 MR. THAGGARD: Yeah, I'd just like to add 18 one thing in terms of the insights about the security.
19 I think the comment about the low-hanging 20 fruits is probably a good analogy. If you -- in order 21 to quantify the risk of security, I do think it's a 22 bit more challenging because trying to quantify the 23 initiating event is, you know, difficult.
24 And so I think the idea that maybe it's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
68 more difficult to look at risk from security -- from 1
that perspective, you know, and so that's probably a 2
good point. I would agree -- I would agree with the 3
question -- the question that -- what they're raising.
4 MS. KOCK: Further perspectives on that 5
last question?
6 MR. TRUE: Yeah, I'll jump in. I, 7
largely, agree with Mike. There is opportunity.
8 There is still more -- a lot more activities going on.
9 I think that the implementation of the very low safety 10 significance issue resolution process and the RIPE 11 process to try and help us dispense with issues of low 12 safety significance quickly and allow our resources to 13 be put back on the things that are more significant 14 are really important and still in the early stages of 15 implementation.
16 And there are a number of utilities that 17 are continuing to actively pursue some of the more 18 ambitious applications like 5069 and the risk-informed 19 completion times, and we'll see those coming over the 20 coming years.
21 On the physical security side, I agree in 22 principle with Mark that it's very difficult and may 23 actually be beyond the state of the art to quantify 24 safety or security risks and -- but that doesn't mean 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
69 we can't use risk insights and a lot of the things we 1
have learned on the risk-informed completion or the 2
risk-informed activities to inform our security 3
practices, and I think we have started that with 4
looking at offsite response.
5 We have started it with looking at other 6
dimensions of the security response that, I think, can 7
be made more realistic with a complete understanding 8
of the overall plant and its capabilities. So I think 9
there's some places to get some benefit there.
10 And then even outside of the reactors, I 11 think there's -- that risk-informed thinking can be 12 beneficial.
13 Oh, and one last thing. We're starting a 14 new initiative on using risk insights -- not 15 necessarily PRA but risk insights in the aging 16 management programs to focus on the things that are of 17 safety significance and put less focus on those that 18 are less safety significant in the overall aging 19 management program, and working with the NRC on that 20 and expect to see significant benefits there in the 21 long-term operations.
22 MS. KOCK: Okay. I'm going to move into 23 some closing remarks. We have only one minute left.
24 Just want to take the opportunity to thank everybody 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
70 who supported today's panel session.
1 I really appreciate all the panelists' 2
time. I appreciate those of you who are in the 3
audience and participated in the discussion. It was 4
rich. The questions were great. So thank you for 5
that.
6 Just some highlights that I picked up from 7
the presentations and some common themes. I think we 8
got out of today's session that risk is created if the 9
created concept is multifaceted, and that can bring 10 challenges.
11 And I think we heard that risk-informed 12 decision-making does allow us to focus on the issues 13 that are most important to safety. I heard today 14 about a lot of tools that we can use to help us make 15 these decisions.
16 Doug talked about PRA. Smain talked about 17 processes and procedures and Mike talked about 18 integrated decision-making. To me, those are all 19 tools that we can use to help us make risk-informed 20 decisions and use the tools that we have in the best 21 way possible.
22 I heard from your response to the polling 23 questions and the discussion that we have made 24 progress but there's more to do, and how can we move 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
71 forward to address those issues where there's still 1
challenges -- you know, I think building understanding 2
through discussions and understanding different 3
perspectives. I heard about issuing guidance to help 4
folks understand how to make those decisions in areas 5
that are challenging like defense in depth.
6 And I think we heard that overall plant 7
safety has improved, and PRA and risk-informed 8
decision-making have been a part of that. But, 9
however, there are still uncertainties that exist and 10 we need to kind of step back and look at the big 11 picture.
12 So in closing, if you can bring up the 13 contacts slide. Oh, you did. Thank you very much.
14 There were a few unanswered questions. So 15 the contact information for our session coordinators 16 is up on the slide. So any unanswered questions will 17 be sent to these coordinators. I encourage you to 18 follow up with them to get the answers to any 19 questions that you have that were not answered today.
20 And with that, I'm going to close this 21 session and tell you to have a wonderful evening and 22 we'll see you tomorrow.
23 (Whereupon, the above-entitled matter went 24 off the record.)
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com