 I would like to start with some broad trends that will anchor the basis of my remarks around the role of information and communications technologies in enabling and helping create this utility of the future. And I'll begin with a couple of broad trends because innovating in the utility industry, I mean I work in a global company and I get the privilege of going around the world and engaging with different problems in different parts of the world is characterized by its diversity. Right? It's very hard to generalize what kinds of problems, what does it mean to create the utility of the future if you not anchor it to a particular geopolitical situation. But there's a couple of trends that are very interesting. First in the developing world there's the fact that well over a billion people don't have access to electricity at all. And in the same way the chairman was alluding to the foundational characteristics that a grid, a modern grid has in the nation and in the economy. This aspect of providing electricity to that over a billion people is a key priority, right? And the choices that get made of how those grid gets constructed, the choices of generation, how consumption happens will impact not only those economies but will impact also the rest of us, right, in developed economies. But I'll highlight the very interesting trend on the left-hand side, the electricity connection with growth, with GDP growth. In this case you're seeing it for the United States. And there's two things that come into view when you look at that graph. The first thing is that obviously there was a very close and continues to be a very close correlation between electricity usage and GDP, right? So we see that the trends of what goes up and goes down is very interconnected. So whether it's a leading or a follower indicator, but they're clearly connected. But what's fascinating is the other trend that gets overlaid on top. And that is the fact that even in the United States when there were GDP growths of four or five percent, electricity consumption was growing by double digits. But there is this very clear trend as time goes by, and something very important happened, right, in the last 10 years or so, which is they crossed. This is a measure of efficiency, of utilization of electricity, it's a good thing. But it presents very, very fundamental implications to the utilities and their business environment and their ability to aggregate capital and invest to do the transformation that is being demanded. So even for the foreseeable future, even in growth rates of two or three percent, the electricity consumption in the United States is only expected to grow moderately, if at all. There is a second trend that I want to allude to you on the traditional value chain of generation and transmission and distribution and customers. The flow of electrons is what determine the value chain and the information flow, basically also one in that direction, in the end resulting in a bill that a customer would have to pay. But compounding to these trends is two big things that are happening, and Arlen is a wonderful example of both the opportunity and the challenges are presented when you incorporate, for example, a very significant amount of renewable energy at the generation size. But it's clearly a source of uncertainty that gets injected into the system. But there's a second source of uncertainty that in the decade ahead we're going to see growing in spades. And that is the fact that the demand, which used to be very well understood and predictable, is going to be much more uncertain. And the uncertainty is going to be driven by the fact that increasing amount of local generation may happen at the home, more complex loads, whether it's electric vehicles, or demand response programs of which one has to carefully understand the implications of how people will respond to those incentives will happen. But the net result is now we have this uncertainty on both ends. The third trend is that there is a growing amount of instrumentation and interconnectedness of these devices. And it is in this context that I would like to discuss how information, information technology can contribute and deliver value to address some of these issues and also create some opportunities. And bringing these two worlds together is a very interesting challenge. Actually, not that many decades ago, information technology not only wasn't at level of maturity it is today, but it wasn't as prevalent in the utility world as it is today, and I say it's going to be in the decades ahead. It was a focus much more on machinery and equipment and reliability of that equipment and constructing the physical aspect of the grid. But the achievement that has been done with electrification has been nothing short of extraordinary. I always highlight this fact that despite the amazing achievements of technology in the 20th century, when the US National Academy of Engineering was asked to decide what was the greatest achievement of the century and engineering-wise, they said it was electrification. And I say this with the greatest respect as a result when we talk about creating the utility future on the basis of something extraordinary has been achieved already. Now, we understand that there's a transformation that is going to happen, that we're going to get a chance to discuss in the utility industry. But I also want to highlight that if I was in agreement about electrification, I would say information technology would have come a close second and it may be that in the 21st century, the contributions of information communications technologies that we'll have in the world may be the greatest impact of the century. And this graph, which was created by Ray Carswell, is a fascinating thing. So let me just describe what it says there. It says $1,000 of computing at constant currency. How much computation does it buy you in 100 years? And if you look at this, it says about 10 to the 12, 10 with 12 zeros of value at constant dollars. Nothing in the history of mankind from a technology point of view has produced that level of progress. And the result of that is that now we have this confluence of pervasive computing, mobile technology that everybody has, virtually free computing, so to speak, right? Cloud computing, infinite capacity in being able to process information. These internet of things, increasing instrumentation, pervasive networks as well, and social. The desire people to connect with one another and share information with one another. All of those things combined together are produced in more data than ever before. And the result of that is that we are the very beginning of this journey, right? Now we have to put in adjectives like zettabytes and so on that people don't know what they are and I have to look them up too. So they are like, you know, 10 to the 21s and 10 to the 23rds. Numbers that we were only using for like, you know, numbers of stars in the galaxies and so on, now are things that we actually have to think through influencing the business and technology choices that we make. So I'd like to give you two concrete examples of areas in which we are applying some of these ideas and the role of information technology in addressing how to optimize. And I choose that work carefully. Energy systems and create the optimized utility of the future. Why optimization? Because it's a business imperative. Those trends that I was alluding to at the beginning. To be able to invest, right, in creating this utility future and the future grids in a context of the product being created, not growing significantly. In the context of a high degree of uncertainty, there is no other avenue that to in a way sometimes do more with less, to optimize. And to be able to use some of that optimization extraction to invest back. So I'll give you two concrete examples. So the first one I'm going to start is what is some of the most interesting ideas at least in the United States around how do you optimize energy as a system? Many utilities across broad areas. And there's a fascinating project that we're involved in called the Pacific Northwest Smart Grid Regional Demonstration. It's part of the DOE projects that were created during the American Recovery and Reinvestment Act, you know, the time of the crisis. There was $5 billion allocated for smart grid projects. This is a set of demonstrations. This is the largest of that demonstration project. It's a $178 million project involving 11 utilities across five states in the Northwest of the United States. And the big idea here is something called transactive energy management. And here's the idea. The idea is we need to introduce two new signals into the grid. These two signals are going to allow us to create a very distributed overlay of information that is going to utilize a cost-based economic signal for distributed control. So it takes advantage of the following trend. The trend is that there is embedded intelligence on almost everything. Of course, from a fancy substation to an electric vehicle to even your fancy toaster. There is some level of microcontrollers or microprocessors embedded in there. And there's enough computing power distributed to the entire network that they could receive signals that do two things. One is an incentive signal. The incentive signal reflects the true cost of electricity at any given point in the network. So when you're in the generation unit, there is a cost of electricity that can be imputed at that cost, at that node. As it flows and goes through it, that cost signal can be modified. It's a time series. It spans two days or three days. It can be in eight-minute intervals or 10-minute intervals. But basically, it just says what's the cost at any given point in the network. The feedback signal, it's another time series that flows in the opposite direction. And this is being generated by any given node in the network. And it basically says how much am I expected to consume over the next two days or three days? So if I'm that electric vehicle, I may say I'm going to consume a lot right now, but not a lot in the next eight hours and so on. And those updates get generated so long as you're connected. The idea is with those two signals, you can do very sophisticated things. So let me give you just a concrete example on how this may work and then generalize it for a second. So imagine that you have this overhead line. There's a transformer. There is three houses, three electric vehicles. The red line that you're seeing over there at the top left of the graph indicates the capacity load factor of that transformer. That's the maximum amount of power there. So what we're going to do is on those charge stations and in the transformer, we're going to make those transactive nodes, meaning those nodes now receive this incentive and feedback signal. So the first EV comes in and he's a flexible load person, meaning I like to charge now, but if I have to charge later, I can do that too, as the preferences have set up. So at the beginning, he connects and below the red line, no problem. I'm connected. I'm charging my car. The next one is a Tesla driver. He says, I only want to charge immediately, no matter what it costs. Under any circumstances, I want to charge. So what happens is that in this case, the signals are being sent, he's charging and you notice that we've reduced the amount of charging going into the first one, the one that was a little more flexible. We keep going and a third one, which is a very frugal electric vehicle owner, comes in, charges and because he's extremely cost sensitive, no amount of charging can take place right now in this context. But as the day goes on, at some point, his needs are being met and so on. Now, there's many different ways to solve this problem and I'm sure a variation of this is being done with electric vehicle programs here. But what is interesting is that this is a method that it's addressable for electric vehicles, but it also is going to be used across entire networks of systems. So this is the range of the project in which these five states are involved and it's all sorts of experiments from renewable integration to electric vehicle to demand response to at home heaters and these signals are actually expanding many utilities within utility going through sub branches and connected across broad regions. And I'm not going to go into details of what is here, but I just wanted to be able to go into the message that it's a framework that goes from something as simple as doing demand response to something as complex as balancing supply and demand at the level of a system operator. And the key thing is as you have increasing uncertainty on the generation and the consumption, these new signals are going to become central to enable the use cases and to deliver optimization. So let me give you now switch gears and give you a second example and now address it in the context of a single utility. What does optimization may look like in there? And this brought up an interesting point around how do you actually do the innovation itself? How do you bring up these ideas? How do you test out these ideas? And we became convinced that we needed to connect or create a new vehicle for collaboration a number of years ago and that's what led us to launch the Smarter Energy Research Institute. I'm going to show you a very short introductory video of this and then I'll talk about some of the ideas we're pursuing. So let me share a bit the agenda and the mode that we work together. The idea was to bring power engineers and mathematicians and computer scientists working together side by side to solve some problems that were long-standing and some new problems that were novel and approach them in different ways, right? So we engage a lot of dialogue with different utilities around the world and the first thing we had to do was to construct an agenda. Where were the area where we thought, right, advancing this idea of the utility of the future that we're more ripe to create benefit and value? And we settle on five areas. We settle around minimizing increasing reliability, minimizing outages. Now, as far as I know, you have an excellent, you know, reliability metric here, but I can tell you in the area where I live, I don't have an excellent reliability metric. I've been out of power to say the last two years, once eight days. The year before was six or seven days. And, you know, that happens in many areas now also in the world that have overhead lines and, you know, is one of also the impacts that you have increasingly violent storms. They knock down power lines and that's a big problem, right? A second area was, okay, it's a very asset-intensive industry. How could we manage assets much better? Again, to generate some surplus capital that can be used to do drive investment back into things we care about, that society cares about. Integration of renewables and distributed energy resources. Wide area situational awareness, right? How do we have a real estate, a real-time assessment of the grid using technologies like synchrophasers and so on? And then the last one, the participatory network. How are we going to transform their relationship with the consumers, right? And what active role are they going to be able to take? So, our approach on this was to say, let's take data that we already have, not data that we could have, but the data we already have, and try to reimagine some of those problems by building actual algorithms and applications that we can test and validate and prove. So, the institute is constructed around not doing studies, but creating working applications that we can test in the context of the utility. So, as an example on the first one, on the outages one, so we're doing something that takes very high resolution weather forecasting that we have developed over a number of decades that, you know, do weather forecasting at the level of one kilometer resolution with very fine temporal resolution. And we couple that with a damage model, a statistical damage model of the impact that previous storm has caused on the utility. And from there, we link it to an optimization problem to determine how do you position crews to optimally respond to the damage that this storm is going to cause. And we do all of that three days in advance. And when you do that three days in advance, you can prepare far better to the damage that is going to occur. And that's what this application does. It basically tells you what is the optimal prepositioning plan that you can do with a lot of lead time. This is important when, you know, in the United States, for example, when you require mutual assistance, there may be crews that are not in your utility that you have to bring in and it's first come, first serve. So, it actually leaves a very significant value to them there. A second example, this is what hydrocobac we're doing. And this is, they are deploying these technological synchrofacers that allows them to measure the phase angles and the voltage across their transmission network. And this data comes at 100 millisecond latency. So, the data comes in and historically, they could only use it for when an event had occurred, some problem had occurred, they could then use it after the fact to say, what cost it? What analysis could we do? But the desire here is, could we use it in my control room? Could I use it in real time? Could I make decisions as the data comes in? So, this presents an interesting information problem because you don't have time, as all this data comes in, to put it into a database, analyze it. The data as it comes in, you need to analyze and create the analytics and the visualization in 100 milliseconds. So, again, an interesting problem from an electrical point of view combined with information technology to solve a very real problem, because in the end, these impacts their ability to export. And that's a huge business proposition for them because they're exporting electricity in the United States. I heard today, they might be interested here in Ireland to be able to export clean energy to the UK. So, the ability to be able to operate within safe margins and not, worst case, every scenario has enormous implications for the business value that one can do after the fact. In these cases, a project we're doing with Alie Anders, the largest utility in the Netherlands, and this is very interesting because they were saying, okay, we're going to instrument all our distribution network. But they do the analysis of instrumented distribution network, the secondary substations to have very visible in the system and it's very expensive. It's like 6,000 to 12,000 euros per secondary substation. So, the idea was, boy, I wonder if we could use analytics to instrument just a subset and have the equivalent of virtual sensors to have a similar level of visibility but with much lower level of instrumentation. And that's exactly the project we're doing there where all of a sudden we're providing visibility of the network with a fraction of the instrumentation cost in the absence of that. Huge savings from investment that they can then allocate to new things. This is a very interesting project. It's called WISE that we're doing again with HydroKabac and this is about hydropower that are incorporating increasing amount of wind. They're 98% hydro in Canada, in Quebec. And the problem we have is as they ramp up wind production and they want to increase their export capacity, they were worth casing their safety limits of how they operate their hydro plants. So here we introduce techniques of stochastic optimization to redo the unit commitment problem. And as a result of that, we don't have the worst case and we've demonstrated that under any scenario that we have analyzed, we're able to operate within margin and increase again the export capacity of their network. It has enormous financial implications for them. To be able to do optimization at the level of 3% gains and 5% gains. And lastly on the project on the participatory network, here we're asking in the Netherlands, we're trying to understand, and I'm sure you've done similar experiments, but we're trying to understand if you're going to do a marketing campaign or reach out to customers to do energy savings, how do you know who are the best targets? And so we've done very sophisticated analytics to first be able to understand who is sustainability conscious in the company, I mean in the country. From there, to formulate the problem as an optimization problem, you say, look, if I have a budget of 100,000 euros and I want to be able to achieve a set of expected savings and energy-wise, who is the optimal target that I should do, right? And to be able to dynamically adjust this as we go. So I'll conclude with a dialogue that I'd like to have as well, which in addition to the innovation agenda itself, of the projects we're doing, I think it's very interesting to think about the model of innovation as well. And we've lived this, I work in an organization that's been 60 years doing research, and through its history, we've had to reinvent ourselves many times. Sadly, in the United States as an example, many world leading research organizations have disappeared, Bell Labs or Xerox Parks, a shot of what it was, many of these interesting places. And that has been, we've had to evolve in our business model, and we have this model right now of the world is our lab. We're going to reimagine how to build smart grids or smart healthcare systems. We cannot do that within the conference or laboratories. That requires partnering with others. This institute that we created is an interesting example of doing that. And what I would like to be able to also have a discussion with you, these are all examples of active research projects we have around the world in collaboration. But I'd like to be able to discuss aid innovation climate also in Ireland in the field of energy. And I have some points of view, but I'd love to hear from you around what makes some places so much better at innovating in this space than others. And what would you envision a good model would be for innovating in the energy space, in the ecosystems that you have developed here. So with that, thank you very much for your attention.