Interview: CNSC's Velshi on Achieving Regulatory Harmonization

Since taking over as president and CEO Canadian Nuclear Safety Commission (CNSC) in August 2018, Rumina Velshi has helped Canada’s nuclear regulator prepare for a flood of small modular reactor (SMR) and advanced reactor design reviews as Canada advances ambitious plans to become a market leader in such these reactors. Velshi has also been an advocate for regulatory harmonization in such efforts, both through an innovative partnership with the US Nuclear Regulatory Commission (NRC) and via her chairmanship since 2020 of the International Atomic Energy Agency's Commission on Safety Standards. This week Velshi sat down with Energy Intelligence's Phil Chaffee on the sidelines of the NRC's annual Regulatory Information Conference on the outskirts of Washington, DC. A shortened and edited version of that conversation is below.

Q: You have been one of the biggest advocates globally of regulatory harmonization, both multilaterally and operationally. When I saw you speak at the World Nuclear Exhibition in Paris in December 2021, when the Darlington SMR competition results were announced by Ontario Power Generation (OPG). At a subsequent panel, you said that "Ultimately my vision is that we get to something like a [multi-regulator] ‘certification of a design'." Do you think that such a certification is in the works? Or even just a possibility?

A: I still believe that that's what we want to get to. And I still believe that that is possible. And if I look at what we are doing with the technology that has been picked for the Darlington site, the GE-Hitachi BWRX-300, I think we are really going along that trajectory, with us and the US NRC, and with OPG and the Tennessee Valley Authority [TVA], which will likely pick that same technology for the Clinch River site.

The commitment that the five CEOs [from the CNSC, NRC, OPG, TVA and GE-Hitachi] have made is that it will be a standard design and that only if there are site-specific unique needs that necessitate deviation from that design would that happen. And if that were to happen, then it would be brought up to the CEOs to then make sure that there is strong rationale for doing so. The work that we're doing with the US NRC is setting the groundwork so that at the end of the day you actually have a standard design. Whether it's certified, it is really one design that will be licensed by a couple of regulators.

There are other regulators knocking on our door. And we've said, ‘Look, if you are regulating the same technology, and if you make that same kind of commitment to a standard design, then you're welcome to join.’

All of us have regulatory requirements that are based on our existing fleet, our own experiences, or wherever those requirements have come from. And so if you do a one-on-one comparison, there won't be perfect harmonization. But what this is forcing us to do is even where there are differences, what impact does that have on the end product? And if we can accommodate that, certainly in our space we've got equivalency requirements, because they're more performance-based requirements. If you can demonstrate that you're going to meet those safety objectives, or whatever the criteria are, then that's good enough.

I still believe that to get to the kind of global deployment of hundreds, perhaps more than a thousand SMRs in dozens of countries, it's the only model that's going to work.

Q: In this instance with the US NRC and CNSC, it sounds like you're saying the final product might not be a jointly-issued certification, but is instead a design that is very, very close in both countries.

A: Right. And that it is because our licensing schedules are a little different. We're a little ahead, and they could be a year to 18 months behind. But at the end of the day, it will be that same design that has been licensed by more than one regulator.

Q: How would this work with other regulators? I know you just signed this deal with the Polish regulator. Would you start a new process with them or they somehow slot them into the existing work you've been doing with the NRC?

A: So we've got the UK nuclear regulator that we've also signed an agreement with, but that's waiting for them to make a decision on their technology. We've got the Swedish regulator who is also interested — again depending on their technology — and then with Poland, it's a new nuclear country. They have Synthos, as you know, which has already picked GE-Hitachi’s BWRX-300.

The way we envisage our model working is that they will join us, working jointly with the US NRC. They'll have staff seconded to us and seconded with the US NRC. They will be at the table reviewing this so that they get comfortable with how the joint reviews are going on. Because at the end of the day, it is up to the national regulator to approve the licensing.

But we are not slowing down for them. They have to catch up with the work that has already happened, and then continue along with us. And so we're going to see how that works. We're hoping that this model can accommodate new nuclear countries who can take confidence that ‘Have I got these experienced mature regulators? Yes. I have confidence in the process. And I have been a part of it as well. So it's not that I'm just accepting it, I am myself getting comfortable with how this is unfolding.’

Q: How would that would it work for a program where there's no real overlap, and it’s after you've issued the construction or even operation license? I mean, let's assume you issue a construction license for the Darlington SMR, and then subsequently a country comes along — let's say an Estonia. How do you bring them in if they can't then sit in on any of the review process?

A: Right. The work that we're doing with the US NRC, it's half a dozen or so joint reports that we have issued, and those are all publicly available. With countries like the UK or Poland, they would have to review those reports. And then they could question and ask and so on to get comfortable with it.

I suspect that three years down the road, if another country decides to follow the same path, they would look at our whole licensing process, look at our assessments or reports, the Commission's reviews and decisions, and make sure that they are comfortable with it [the process].

Now please remember, these are the first-of-a-kind ones [licensing reviews] for the next little while. But how do I see things happening 30 years from now, in 2050 or so? I see it quite differently then, because it will be in a more mature state. And your question about a [multi-regulator] certification of a design, I see that actually in place then.

We are in discussions with the IAEA on this. We have a pre-licensing process, a design review process. The IAEA currently provides an Integrated Regulatory Review Service: every few years they come and assess the different regulators and see how well the regulators are carrying out their responsibilities, and issue a report. And we're hoping that they can come and do a peer review. They'll be bringing experts from around the world of the pre-licensing process. We have one, the UK has the GDA [generic design assessment] process, the US NRC has a certification process. Have a look at those processes, review them, and then give your Stamp of the Champ on those processes. So that a new country, if they are looking at our vendor design review process results, they can take comfort that an independent objective assessment has shown that if something has gone through this, what's coming out of it is of good quality.

And so we're in discussions with the IAEA. I know there's interest in that. Can we make that work? We look at the aviation industry as the model that we could look at it. We also do that currently already in the nuclear industry with our shipping containers: if one country has certified a flask, then in Canada we would say 'Well, will this meet our needs because we've got colder temperatures?' Or whatever it is. And if it's within that [Canada-specific issue], we don't redo the tests. We don't do the reassessments. We accept that certification. And, and so I think that's kind of the model we want to go with, that we should be striving for.

If I look at what the UAE did with the Barakah nuclear power plant, it's a technology that was deployed in Korea, and they pretty much took it lock, stock and barrel, with very minor modifications for their site. And the success speaks for itself: deployed safely on time and on schedule.

The principle around here — don't customize your plant, accept what others have done after you've done your quality checks — is the only way we can realize the full potential of SMRs and advanced technologies.

Q: So would the CNSC be comfortable on the other side of this equation? I'll give you a hypothetical because I know Ontario is now talking about large reactor newbuilds. Let's say an AP1000 is selected by Bruce Power or OPG. That design has already been fully licensed in the US, China, and the UK. Would you be comfortable relying on those detailed reviews to do a less comprehensive one of your own?

A: Absolutely. The work we're doing with the US NRC is also on that very same premise. The US NRC had certified a predecessor of the BWRX- 300, the ESBWR. Those topical reports from that time, we're using those and not recreating those. So long as we know that we have done our own due diligence of 'Has that gone through a rigorous process?' We don't need to reinvent any of that. And so if it came to large reactors, whether it's the AP1000, whether it's the APR1400 from Korea, we as a regulator would feel absolutely comfortable reviewing and accepting what other regulators have done

Q: How do any of these regulator-to-regulator efforts slot into what IAEA Director-General Rafael Grossi told the conference about this morning regarding the IAEA's regulatory harmonization efforts?

A: The Nuclear Harmonization Standardization Initiative [NHSI] has an industry stream and a regulator stream. The regulator stream has three different working groups. I've got staff participating in each one of those, as does the US NRC, as do many other regulators.

We jointly work on those, and it complements what we're doing. It doesn't address all the needs. What we're doing with the US NRC or what France, the Czech Republic and Finland are doing with Nuward [an EDF SMR] is more technology-based, more results-focused for some specific technologies.

What the work that the IAEA with the NHSI initiative is building blocks. How does information get shared? Putting a technical document on certain things. How do we make sure that our international standards are appropriate for these new technologies? And do they need revisions? And so on. I think it's looking at things like that.

Where does NHSI go next, after it's done this? I've got some thoughts, and I'll share some with you. Particularly for these new emerging countries, and the support they need for infrastructure, perhaps an international technical support organization [TSO] is a model that we should be looking at. All these countries cannot afford — or it doesn't make sense to build — their own expertise. Either on a regional basis or international basis, you could go get that support when you need it. You still have to be the smart buyer and know what it is you get, but you don't need to have that enhanced capability.

Q: Sort of like an international IRSN [France's Institute of Radioprotection and Nuclear Safety].

A: Absolutely. You need exactly that — an international IRSN, an international TSO.

There are a lot of things we're looking at, but it's with that overall end objectives of how do we not duplicate efforts? How do we get to standard designs? And how do we help these newly emerging countries get up the curve very quickly, and in a safe way?

Q: Since the invasion of Ukraine last year we've seen an ever-growing partition within the global nuclear sector, with Russia and China on one side, and many OECD nuclear countries on the other. Do you think there's a way to keep that partition from growing in the regulatory world? Is there an effort to reach across this divide, to have transparent lines of communication with regulators in Russia and China and other countries when the politics are difficult?

A: Difficult question. Very difficult question. I'll talk about Russia — China a bit less so, partly because I just haven't had that many dealings with the Chinese regulator in the past. With Russia, things have changed in the last year or so. Because at many international fora, the Russians used to be a very visible, active player. So, we've talked about these discussions around harmonization, for instance, or setting up an international TSO. There would be a concern among countries if Russia was at the table in those kinds of conversations. And yet, when it comes to safety, it is in everyone's best interest that we hear from everyone. But I think the reality makes it quite challenging.

Q: In Canada, the CNSC is shifting from overseeing a nuclear industry with a couple fuel cycle facilities, a small fleet of Candu reactors and some medical isotopes to reviewing who knows how many innovative reactor designs, with more every day. With Canada such a first mover on SMRs and advanced reactors, how has the CNSC prepared to regulate such an incredible breadth of different new technologies?

A: This work started long before I joined the CNSC. There was a recognition that this was coming. And the mantra we have in my organization is to be in a state of readiness.

You know, one of the biggest areas of challenges people see with the deployment of these advanced technologies is the regulator: that there isn't regulatory certainty, predictability or efficiency. And we want to make sure that we are not an impediment — an unnecessary impediment — to the deployment of these technologies. So how can we be in a state of readiness? We've tried to be a lot more anticipatory in our thinking. I'll give you a few examples of how we've done that.

One is to engage early and frequently and on an ongoing basis. For SMRs, we've been very fortunate that the country has taken a pan-Canadian approach, with an SMR Roadmap, an SMR Action Plan, and now we have an SMR leadership table as well. And the regulator has been at the table, a key player, right from day one. In other jurisdictions, the regulator likes to keep their distance because they want to make sure that their independence is not compromised, or the optics around that. We feel quite differently about that. Policymakers need to hear our perspectives. We need to know what's coming so we can be prepared, and so we can actually influence the outcomes. This has no negative impact on regulatory independence, which is key.

The second one is that we've been very fortunate. We have received funding from government in anticipation of what's coming so that we could build up our capacity. We could start hiring long before an application came across our desks. The building of capacity and expertise, having relationships with universities and research organizations, has been the second one.

The third one has been our regulatory framework. As you said, it's all been devised and set up for Candu reactors. How do we make it technology neutral? How do we make it more performance-based, but also more risk-informed? These are new creatures, these SMRs, and you don't want to take a cannon to something where a slingshot may do just fine. And so how do we make sure it [regulatory oversight] is commensurate with the risk. So we've refined our regulatory framework. We've had a peer review of that, to endorse it and identify where we can improve. So our regulatory framework is up to speed.

We still have processes. We have an impact assessment process — it's like our environmental assessment process. And if you were to ask the Impact Assessment Agency 'Well, how long would it take?' They can say 'Oh it could take up to seven years, and whatever hundreds of millions of dollars.' And we said 'That just doesn't work.' Given the urgency of the situation, we need to approach this very differently. Kind of like the COVID vaccine, right? For something that would normally have taken 10 years, getting it out in 10 months required a paradigm shift, and we need to do that with this. So we've started conversations on that.

If I take the Bruce site [the site in northern Ontario home to Bruce Power's multi-unit Candu plant] — the largest nuclear site in the world — they may be considering adding other large reactors on site. It would have to go through an impact assessment process — that is the most studied site in the country. We know that. The community engagement has been fabulous. And it's a standard setter. So with this impact assessment, where you're trying to see what the social, economic, and environmental impacts are, a lot of that work has already been done. Why would you want to redo that? And so how do you come up with more customized terms of reference — the scope of work for this impact assessment? So something that could take up to seven years, maybe we can do it in two years. We're reviewing our Canada impact assessment process, as well. That work is underway. And I'm actually really pleased with what is going on.

Then there is the international collaboration. Because we know that none of us have all the answers. We haven't licensed nuclear plants in over 30 years. And how can we join forces, learn from each other, and lead to better safety outcomes? So we started with the US NRC. And as you've heard, we've worked with others. I happen to chair the IAEA Commission on Safety Standards. That was my top priority: you've got to review the standards to see what are the gaps for these new technologies.

The last one, which is just so important, is around building public trust. Not only in the regulator, but in these technologies. And for these different communities that are being considered for the deployment of nuclear, we as the regulator have been out there to let them know who the regulator is, to hear about their concerns, and to build relationships. We've been doing a whole lot to be in that state of readiness. And I'm really pleased with the progress that we have made.

Q: You said you don't want to be an impediment to the deployment of SMRs and advanced reactors, and then qualified that with 'an unnecessary impediment.' What is a necessary impediment, and what is an unnecessary one?

A: Well, look. We'll likely always be on the critical path. Then it will always be in the eyes of the beholder — 'You could have done this faster' and so on. The unnecessary part is 'Should I have anticipated this better?' And that is what I mean by the impediment. It is also being more risk-informed in our decisions.

I'm not naïve. I know some people will always say the regulator is just this unnecessary evil we have to deal with. But I think in Canada, I would hazard that most of our licensees would say that having a strong regulator, who provides certainty and clarity and predictability, is exactly what is needed. And we're very fortunate in having that.

Q: During your tenure at the CNSC you've been one of the most prominent women in the global nuclear sector. To what extent do you think that sector has built out a place for women, and to what extent are changes still needed?

A: I love that question. You know, I've been in the sector for over 40 years. I was one of the first, if not the first, [female] nuclear energy workers in Canada. And from that time to now things have changed rather significantly. Whereas before you were maybe the only woman, or you could count on one hand the number of women, but look at the RIC [NRC conference] today. Women don't make up 50%, but there there are a significant number.

So I think there has been visible change. Has it been adequate? Absolutely not. Women still make up less than 25% of the workforce in nuclear. And then if you look at the more technical [roles] — the science, technical engineering, mathematics — or if you look at the leadership roles, it's a lot less.

What has changed? Just before I took on this role, but after I had been appointed to it, I had been invited to speak at a conference in Ottawa. It was kind of like my maiden speech. I made the gender equity piece a big part of my remarks, and it's not even that the eyes were glazing over, but it was like, 'What the heck?' like it was not spoken about. And that was five years ago.

Now the emphasis given around diversity and having different voices, a lot of that has just changed in the last three to four years. A lot of it happened during the pandemic. A lot of it happened with Black Lives Matter, when it became so evident that you need to be more inclusive. Now you don't have to make the case for why you need to do it. It's now more on 'help me to get to where I where I need to be.'

It's going to take time. The pipeline is really weak. The last couple of years have actually been a setback, because many of the women [in the sector] are leaving the workforce for a multitude of reasons. So I think you need an even greater concerted effort. In my organization, we're very fortunate: we're pretty much at gender parity at all levels, in all fields. Not so in the Canadian sector. And even if I look at the Western countries, we're all pretty bad. A lot more work needs to be done.