Quantum computing for energy: Smarter grids, better supply chains

The evolving landscape of quantum computing

In a recent episode of the Energy Transition Talks podcast, Maida Zahid sat down with quantum computing expert Curtis Nybo to explore the challenges, benefits and future of this emerging technology. In this second instalment of the conversation, Curtis moves from theory to practice, focusing on energy optimization, logistics, cybersecurity and business applications for quantum computing across industries.

Overcoming challenges: The race to stable qubits

Quantum computing faces several hurdles, with coherence times being one of the biggest. Qubits, the building blocks of quantum computers, remain stable only for a limited time before randomly collapsing into a 0 or 1 state. This leads to unreliable results, making error correction and shielding from radiation critical in quantum hardware development.

Another challenge is the limited number of qubits available today. While large-scale quantum solutions require thousands of qubits, the most advanced machines currently operate with only a few hundred, restricting their problem-solving capabilities.

Optimizing supply chains with quantum computing

Quantum computing is revolutionizing supply chain logistics and optimization. Curtis highlights that quantum annealing computers are being used to optimize complex logistical processes, such as:

·        Route optimization – Determining the most efficient way for delivery trucks to distribute fuel or goods while factoring in weather conditions and constraints.

·        Demand forecasting – Predicting where materials need to be and at what time to prevent supply chain disruptions.

·        Electricity distribution – Although not yet implemented, quantum computing could optimize energy distribution as effectively as logistics routing.

"From optimizing supply chains to predicting energy demand, quantum computing gives us the ability to solve problems we never could before." – Curtis Nybo

Visit our Energy Transition Talks page

Chapters

• 0:00 Introduction to Quantum Computing
• 0:28 Understanding Quantum Computing Challenges
• 2:13 Hardware Limitations and Scalability
• 2:33 Quantum Computing Applications in Supply Chain & Logistics
• 4:21 Cybersecurity and Quantum Threats
• 7:48 Communicating Quantum Benefits to Business Leaders
• 10:06 The Future of Quantum Computing
• 12:12 How to Get Started with Quantum Computing
• 12:39 Closing Remarks & Podcast Sign-Off

Transcript

Maida Zahid: 0:04

Hi everybody. Welcome back to another episode of the Energy Transition Talks podcast. My name is Maida Zahid and I'm part of the marketing team here at CGI in Canada. Today we will be revisiting our conversation in quantum computing, and I'm joined by our Canadian expert, curtis Naibo, who is our lead for quantum computing here in Canada and also leads some of our AI and data analytics practices. So over to you, curtis.

Curtis Nybo: 0:28

Hi everybody. There's a lot of challenges that come with quantum computing, one of which is there's what's called coherence, and so we have set coherence times that a qubit can remain stable before it automatically on its own collapses to a state of zero or one where we want to be able to induce, we want to be able to measure that state on our own, kind of a chord, and not have it just collapse on its own, and that's called coherence times. And so we're right now we're seeing short coherence times where qubits only remain stable for a certain amount of time before they just randomly collapse a 50, 50 chance of becoming a zero or a one which leads to pretty not great results overall. And so when it comes to trying to handle that, handle that decoherence, that's one of the big areas that there's challenges, that they're trying to be able to increase that error correction capability, be able to essentially have quantum computers produce more accurate results. Because right now it's pretty prone to noise.

Curtis Nybo: 1:28

Any bit of radiation that is able to get through to the quantum chip can actually flip those bits randomly, so they need to have a lot of shielding to protect against radiation. There's a lot of different aspects around the actual physical qubit itself, but also there's only so many qubits available right now on quantum computing chips. So to be able to solve a lot of problems you need a thousand plus qubits, where right now some of the top quantum computers, depending on what architecture they are, have, you know, 50 to maybe hundreds of qubits, while you know quantum annealing computers essentially have a little bit more, but it's kind of it's limited by how many qubits are available for each hardware and processing.

Maida Zahid: 2:13

Before we move on to a little bit more challenges, I kind of wanted to understand some more benefits that you know we talked about earlier. So going beyond just energy utilities, let's say supply chain. So you talked about optimization earlier. So how can we kind of dive into the benefits from that perspective in, let's say, supply chain logistics?

Curtis Nybo: 2:33

Yeah, so that's actually the area that I spend most of my time is a optimization route. So we're usually using a dealing quantum computer to be able to optimize processes for different clients and solve different problems. And when it comes to optimization problems, which is usually around logistics and supply chain, in a lot of cases that includes different areas like fuel procurement, electricity distribution, infrastructure maintenance, anywhere where you're actually having to plan. So a common one is route optimization. So the traveling salesman problem if we have so many fuel trucks that need to make deliveries around the world say we have a fleet of 300 trucks and there's potentially 10,000 or say 2,000 fuel depots that need fuel delivered to, what's the optimal way to be able to schedule that fleet of trucks to those those uh depots? And you can take into account different uh different variables as well, like weather, and so that becomes a challenging problem for classical computers to be able to run efficiently.

Curtis Nybo: 3:41

And that's a common optimization problem that quantum annealing computers especially are really good for solving. Another one's demand forecasting. We've done quite a bit of work around predictive analytics, so trying to plan where certain materials should be at a certain time. It's very similar to the route optimization problem, but overall when it comes to energy and utilities. Electricity distribution is an interesting one that I've thought about, but we haven't had a chance to implement it. But it would be a similar optimization problem to that route optimization.

Maida Zahid: 4:21

And I think it definitely goes beyond energy utilities too, but a very common challenge, absolutely just like cybersecurity. So we know cybersecurity is probably one of the biggest things, just along with supply chain, that we hear all sorts of organizations talk about or deal with. So how can we use quantum computing for that, or the benefits per se?

Curtis Nybo: 4:42

Yeah. So cybersecurity is an interesting one, and that's probably where most people hear about quantum computing, because right now, the current encryption method for pretty much everything on the internet is using RSA cryptography, which is essentially trying to generate huge prime numbers, multiply them quickly together and then to be able to crack that encryption you need to be able to factor what prime numbers were multiplied to create the larger number, which is a very hard problem to solve, and right now it can be. A good thing about it is if you think of it as a password, that password can be. If computers got more powerful, say twice as more powerful tomorrow, all we'd have to do with classical computers to make it harder is add another digit to it, make it, make that a little bit longer and it becomes in like exponentially more powerful, whereas quantum computers they're. It's now with quantum computers that factoring of large numbers is is very possible using some algorithms like shore's algorithm, which has been around for a long time. They've never been able to really test it until quantum computers actually started to become feasible, and so an algorithm like Shor's algorithm now allows essentially for the factoring of those large numbers. So you can essentially take an RSA key, find out what those keys were that were multiplied together to create it, and then you've essentially destroyed a lot of the RSA cryptography that exists today, and so this poses a serious threat to the usual classical cryptographic systems. The good news is that it doesn't exist yet. No one's been able to. A quantum computer, essentially with enough qubits to be able to run that algorithm doesn't exist yet.

Curtis Nybo: 6:24

But the scary part is and I think what most organizations are worried about is that data can be stolen today that's encrypted, and then thieves just have to hang on to that data and say, maybe five years, 10 years down the road this becomes possible to be able to crack that RSA encryption. Now they can get that data. They can then unlock that data later on. Even though the data might be out of date, it may still have, you know, quite a bit of personal information. People only have social security numbers, for you know their whole lives. So it doesn't matter when that data is stolen in the future. That data might be able to be exploited in the future.

Curtis Nybo: 7:01

And so the other side of it is quantum computing does provide a bunch of different ways to create quantum resistant algorithms. So there's things like quantum key distribution, where we're using quantum mechanics to create essentially an unbreakable encryption. That's quantum proof, but there's also some kind of standard classical ways to create quantum resistant encryption as well, so things like lattice-based cryptography. There's a few common metrics, there are methods that are being explored today, and so I think in the future what we'll see is a bit of a change and a move away from the current framework that we use for encryption into a more quantum, safe encryption going forward, but I think that's still quite a few years away a few years away.

Maida Zahid: 7:48

For when you're, let's say, when you're talking to clients and trying to put these benefits into layman's terms, what are some of the key points that you come across and you talk to, like, let's say, business focused clients that quantum computing can deliver? So what are the benefits that you are most commonly talking about among these industries?

Curtis Nybo: 8:00

The most common benefits we're talking about when it comes to energy and utilities is usually that optimization type of that are currently running models that are taking in a lot of information. It's taking maybe hours or days to run a single run of that computation. That's a good candidate for quantum computing, especially for quantum annealing, and so we run into that quite often and that's usually where we spend a lot of our time. It also there's also a lot of use cases around, like we talked about cryptography. But overall I think what we're seeing is mostly around the optimization front with quantum computing, and so lots of, lots of organizations are finding just that current computation can take a long time.

Curtis Nybo: 8:48

Quantum can potentially speed that up, and it's important to note too that a good problem for quantum optimization isn't one that has just a huge amount of data. That's not necessarily what we're looking for. We're looking for a lot of problems that have a lot of variables. So going back to, like a scheduling problem, a large fleet of trucks and also a lot of constraints. Those trucks have workers that can only work between eight to five. They can't run at night. They have to use a certain specified amount of fuel.

Curtis Nybo: 9:21

So we're looking for problems that have a large set of variables and a large set of constraints, and then we can basically find that minimum energy state of that problem by minimizing that objective function, which is, you know, all those variables and constraints, and try and find the most optimal solution to that, to that result. And so what that would look like is we'd end up with a somewhat of a schedule that would tell you know which trucks to go where, at what time and what the best routes would be to follow. So it's not necessarily that we want a lot of data, like I said, um, lots of times, it's the problem itself. Uh, can needs to be able to be formulated into a quantum problem, which is usually that Lots of variables and lots of constraints.

Maida Zahid: 10:06

I think we covered a lot. You know sample use cases, challenges and looking ahead. How do you see this kind of scaling for energy utilities, or how do you see this coming? Where do we stand now? It seems like we're pretty in the early stages. How do you see this coming? Where do we stand now? Like it seems like we're pretty in the early stages. How do you see this moving forward?

Curtis Nybo: 10:23

So, moving forward, there's still a lot of work to do. Like I said, I think I talked about a few of those challenges before, and mainly around things like error correction, increasing coherence times, basically stabilizing those qubits for longer so we can get better measurements when we measure them, which results in better accuracy for our solved problems. But overall I think, looking into the future, we'll see advancements in the quantum hardware. So I talked a little bit about how a lot of problems today, like the cybersecurity problem with Shor's algorithm, can't really be used today because there's just not enough qubits within the quantum hardware to be able to solve that problem. And I think we'll start to see advancements in quantum hardware. So Google released their new quantum computer. We're starting to see more players into the game. We're seeing IonQ, regetti. They're big quantum providers and overall they're racing to try and see who can basically achieve quantum supremacy the fastest. So which computers, which computer architecture can solve problems faster than a classical computer is what everybody's going for, and so we're starting to see a lot more advancements and I think those advancements will start to really pick up.

Curtis Nybo: 11:36

Another area that we're seeing expand is the access to quantum computing. So for those wondering how to get started with quantum computing. It seems like it might be far out there, but it's actually not. Quantum computing is available through most cloud platforms. So Azure, quantum services, aws, bracket, even a lot of the quantum providers, like D-Wave, have their own cloud provider cloud platforms and so you can get in and actually utilize these quantum computers to help solve those problems or even just to experiment. You can start using that today, so the barriers to entry, I think, will also drop quite a bit. Hopefully.

Curtis Nybo: 12:12

It gave a bit of an overview of quantum computing and some of the capabilities. My advice would be to kind of just dive right in. Like I said, a lot of capabilities are available. A lot of the places we spend a lot of our time is just experimenting with, trying to shape different problems into quantum capable problems and then we try and solve them using quantum computers. So the technology is here, despite what you might be hearing, and I said, anybody can access it today. Thanks so much, curtis.

Maida Zahid: 12:39

Thanks might be hearing and like I said, anybody can access it today. Thanks so much, curtis. Thanks for joining us and thanks for listening everybody. Please subscribe to our podcast on Apple Podcasts and Spotify or wherever you get your podcasts from. Thanks for joining us.