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