Shaping the Future of Asset Performance Management: Insights for 2025 and Beyond

A Look Forward for 2025 & Beyond

Join GE Vernova’s experts from product management and product marketing as we share a vision of the future of Asset Performance Management (APM). This will be an open discussion, moderated by GE Vernova’s VP of Marketing and will cover topics such as:

• Embedded Hardware: Increased digitization for the enterprise.

• Connected Worker: Enhanced mobile and desktop connectivity across more workflows.

• Unified Platforms: Combine operational efficiency metrics with emissions data, enabling holistic decision-making.

• Sustainability as a Core KPI: APM tools increasingly incorporate emissions reductions as part of asset optimization.

• AI Synergy: Shared use of AI to drive predictive insights for both asset reliability and emissions control.




--
TRANSCRIPT

Good day everyone who is gracing us with your time and attention today. I'm Tracy Swartzendruber, and I'm lucky enough to lead the marketing team for GE Vernova’s power generation and energy resources software business. I'll be your moderator of today's panel. I don't want to waste time on housekeeping. I'm sure most of you are familiar with how webinars work and our format. The only thing that I'm going to ask of you is please engage with us by entering comments and any questions you may have into the platform. We will try to address as available. Now let's start with quick intros. Martha, why don't you lead us off?  
 
Sure. Thank you. Tracy. Welcome, everyone. My name is Martha Saker. I am the director of health and frontline productivity. I have 25 years of experience with GE, optimizing asset performance in the power gen and oil and gas spaces. Wonderful,  
 
Neha. Hello, everyone. Hi. My name is Neha Joshi and I'm the product manager at GE Vernova’s Power and Energy Resources Software. I'm really excited today to talk to you and share my perspective about how landscape is evolving for maintenance and monitoring in terms of asset intensive industries. So really looking forward to our discussions today. Thank you. Great.  
 
And Ryan, why don't you take us home? Yeah, definitely. Ryan Finger, I lead product marketing here. I'm on the APM side. I'm very fortunate to work with Neha and Martha every day. My background is more on the enterprise SaaS space. I come from the financial services background, and I've been with GE just over over three years. So looking forward to adding some perspective on some more software, general ecosystem type topics on where the space is heading in this year and beyond.  
 
So thanks for joining us. Fantastic. Now, I don't want to get too much into the legal stuff, but we do have to remind our audience, that and give you a disclaimer that what we're going to be talking about today is not really a roadmap discussion. We're simply highlighting some different trends that we're seeing in the APM space. And and giving some more color around why we think these trends are important and where we see them going. So we have five topics that we're going to cover today in terms of our trends. Let's just jump right in with topic number one, which is embedded hardware.  
 
Martha, I'm going to I'm going to give you the first opportunity on this one. How are you seeing advancements in embedded hardware influencing the evolution of asset performance management systems? Yes, Tracy. It's interesting because there are many pluses, right? More data might equal more quality insights. And it's great to see so many new instrumented devices and being able to use that information. However, like in everything else, it boils down to a financial conversation and a financial decision, because all that intelligence also has some cost of ownership and maintenance connected to that intelligence. So, being able to perform analysis whether or not that data serves your KPIs and your purposes and whether or not that is a more efficient, more economical way to obtain the same information. So again, all great positive, more data, more insights. But it needs to be that right cost proposition. Makes a lot of sense. Neha, do you want to help us understand how edge devices and embedded systems are improving real time monitoring and predictive maintenance in APM? Absolutely. Thank you. So from my perspective, I see, embedded hardware is like, you know, robots or drones or satellites and how those are affecting, hardware and software.  
 
They're embedding hardware and software on the edge on, on the on prem side. So in the three areas like autonomous automated inspection, maintenance and monitoring and data collection, all three areas get involved when you're talking about the embedded systems. So, in terms of robots or drones, there are sensors attached. You can attach the sensors on top of the robots that can collect the data from the hard to reach areas, like very remote or, you know, to increase the worker safety. Mainly, mainly the intention is like, you know, it takes zone one compatible devices.  
 
They go in the hazardous spaces where human cannot reach or that frequently. And we are also seeing the trend, that the combined cycle plants are seeing change in the operation plant. They want to do more like a night shift where the wind energy is more or they want to use utilize that. And that is the time when it is not very easy to send human. But then robots can go in in those areas. The way this works is you feed the machines automatically and the robot can go, through software remotely. The robot can go, and, get the get the inspection data with the AI and ML embedded into those, devices. It can actually do the sophisticated monitoring and predictive maintenance at this point. So generic analytics is already shipped with the with those devices and with our SME knowledge, we can actually collect the data and feed that back to our back end systems like think about like APM or our back end system.  
 
So very good way of doing making sure that you have single source of truth. And last but not the least is the data collection perspective. Because these robots are collecting various data and numerous amount of data, which is not possible with manual inspections at this point. So, you have, robotic as a service, where you can collect the data on prem and send it to the cloud. So now you have a single source of truth in the cloud. So this all in comes to like, you know, all this, real time monitoring and predictive maintenance on the on the on the cloud side. Amazing. Martha, with all of this being said about embedded hardware, security must play a role here. Can you give us some thoughts on how does security play here in regarding the integrity of data collection and system control?  
 
Yes. All these connected systems definitely opened the opportunity for vulnerabilities to be introduced. And we feel, in our business, as the consumer and main, processor of that data, we have a responsibility to ensure that our platform provides both the connectivity and the security to use that data, process it, and minimize the risk of those vulnerabilities. In the platform is where the logging activities happen, where the user management occurs. So we need to provide all the capabilities to allow people to do this in a simple, fast manner, but is still ensuring that only the right resources will have access to that information. Fantastic. And Ryan, I know, I know, I've I've put three questions out here to Martha and Neha. I just know right. You've got a lot of opinions on all kinds of areas within here. Anything that you'd like to share? Definitely. And I think part of this, we're going to talk in a few later topics, the emergence of some of this technology that we're seeing, like, like artificial intelligence, with embedded hardware, a lot of what's going to be happening is some of these hardware providers, whether it's pumps or motors, you name it, right?  
 
They're going to be working to actually provide that hardware with software of their own, whether that's to extend what they do as a business to collect more data. Right. You see introductions of cheaper, more easily placed sensors. That's going to really be an emergence of what I would call, an increased level of data silos in some organizations. So as you're shipping new parts to site, deploying new parts or components, that comes with its own software, it gets back to exactly what Neha and Martha were talking about, which is how do you engage and use that data? So a big prediction is, as some of these hardware providers continue to embed software sensors or closed systems. How do you take that data and put it in the context of the rest of your asset data, your operational data, your emissions data, your financial data? So we're going to see, I think, an increase in how can you manage different data from different vendors, whether that's hardware or software. And usually that takes a huge shift internally and how you actually work as an IT and OT organization.  
 
So my my prediction here is that continues to happen on prem or cloud. Right? It kind of duplicates those potential silos because of the data coming from some embedded system. So a lot of what can be done with what Neha Martha said is a great way to help limit that. But there's going to be, I think, some hurdles coming for a lot of asset intensive organizations when it comes to making the most of that data. So I would say expect more, software to be shipped with hardware and expect more in terms of the emergence of how that should be handled and can be handled. Fantastic. Well, it sounds to me like we've sufficiently covered topic one and let's move into topic two, connected worker. Now this one, I think it's a fun one for this group. A real fun one. I'm going to ask Martha first and then Ryan as a follow up. You know, connected worker. I already have in my mind what I think it is and what it means to me. But we're going to start here with just a basic definition. What does connected worker mean to you? Because again, everyone's got a definition.  
 
And let's level set here what what it means to us. So Martha first you. Yeah that's definitely the point Tracy because there are so many definitions. And one of the things that we are very conscious about doing pretty much in every one of our products and new product introductions, is we want to be very fit for purpose. And when we talk about connected worker and I'll give you a definition of it, I would like to be very specific in regards to the fact that we talking about connected worker for the PowerGen and oil and gas spaces, keeping in mind, the staffing, the KPIs and the complexities of these two pillars. And for us, connected worker in this space is about enabling those pressure-filled feel to have everything they need to perform the work, hat they are called to do in a safer and more productive manner.  
 
And notice I use safety first, and that includes, the ability to understand what your tasks are faster, having the information necessary to perform that work, and all that is driven by priority. There's never enough manpower on any of these environments to do everything that needs to be done. So the center, the driver, the sunlight here in this, space is your asset health index. What needs to be done first? What do I need to do that and how do I do it safely? Is the content of our connected worker strategy. Great. Ryan, do you want to add your, your context? Yes, I see, I see Martha smiling. Her and I, we chat about about this often, and I like the way she described it in terms of what our focus is as GE Vernova on the software side, which is the productivity of the feet on the ground.  
 
And I think where connected worker gets a little diluted is when providers try to get into I would call front office and back office digitization with connected worker. Right. There's been introductions of technologies like augmented reality and virtual reality. So AR, VR, right. Some folks are saying that needs to be a part of your connected worker strategy for training, retention, inspections. And and it could be. But I think a lot of what's been discussed in the market, especially part of my role, is looking outwards, right? Talking to analysts, talking to customers, talking to our experts. A lot of what's been promised and what was that forward vision of connected worker, which is inclusive, inclusive, a virtual reality. And your meta headset has fallen a little bit short in terms of productivity and outcome. So I like the way Martha frames it. Martha I agree on that, on that aspect. I think when look at it from an asset angle, it needs to start and end with employee productivity. And that includes not just getting digital tools into their hands.  
 
But how do all your applications and systems work together with that data to make those insights better? Right. You can put a bunch of tablets in people's hands, but if the right data isn't in the right place at the right time, those outcomes aren't going to happen. So there's all this promise, and I think Neha’s going to hit on some of this too, with what's coming into the ecosystem. There's all this promise that's out there. But from conversations that's that's not being delivered. So it's really important. When you say connected worker, where do you have to start? And do you have the right data set up to truly connect your workers? If you don't, that promise is going to fall short. So I agree with Martha on the asset side. I have a ton more opinions on the enterprise front office, back office, connected worker. But, that's kind of my perspective on what we're seeing in the market today.  
 
Awesome. And I love that we've level set now on on what our definition is and how we're going to frame the next, next few questions here. So Martha, you mentioned always safety is paramount right. That that is full stop. Must be safe operations. But can you tell me some of the key benefits of integrating connected worker technologies specifically into APM workflows? Yes. And to me the key one is reliability. And I think this is something that's probably misunderstood that the relationship between safety and reliability, a reliable operation is by default, a safer operation.  
 
So for us, everything is centered around how do we improve the reliability so that we can avoid many of these field excursions that then would incur or will expose the field force to additional, situations. So again, the main concern for us is bringing that reliability and in parallel then safety would be augmented. Then the second one of course is the insight. There's never enough resources. Productivity is paramount. We see through the use of one of our products, which is the rounds application is a field data gathering application. We see that there are very few companies that can actually meet or all that inspection and data gathering tools, because, again, there's not enough resources. And we fully understand that. And our guidance on that process is priority. Not everything has the same pressure or the same impact on that reliability. So again, being able to drive all these field processes towards what is the highest priority, where do I have the most impact if I intervene. And that again in parallel will increase our users and operators field personnel safety. Fantastic.  
 
Neha, I know you've you've got some thoughts you want to share here. So I'm going to ask you sort of a two part question. Regarding and Ryan sort of touched on it in his more expansive definition regarding wearables. So part one is how are wearables and mobile technologies changing the way that frontline workers interact with APM? And number two, regarding, this how what kind of challenges are organizations facing in terms of training and adoption of connected worker solutions? And how do you see those, those challenges being overcome? Sure. So before I answer these two questions, I would like to go back to the definition again, my very own definition. But my simplified version is like digitizing the task repeatedly, replacing the repeated manual work. And second, my second thought around connected worker is automating hands free documentation that is accessible from central location.  
 
So looking at these two aspects of connected worker, what I think the main idea here is like visual inspection. So visually, getting the data or getting exactly what the problem is and reduce the cycle time to turnaround time and making making the decision or making the impact on like what, what the resolution would be. So visual inspection could reduce that or reduce that time. And variables, variables actually give you near real time data. Hands-free operation. Drone satellite will give you plant level like on top of the plant. Related aspect. And then, the fun word is agentic AI nowadays. But agentic AI can it can potentially redesign or optimize the workflow or the round or the workflow that will automatically, take the, take the data, collect the data it can optimize and it can it can analyze the bottlenecks. It can do the, change the rigid process that it is already has. It can analyze the patterns.  
 
So using AI technologies, we can all, solve that so that these are the technologies that are changing, the frontline workers with our APM, how they interact with our APM system end to end. And the main challenges that I see, in for areas like, you know, technology integration, of course, all these come with the integration part of it. Workforce resistance is a big one that I'm hearing from customers as well, that how do I, how do I add these new technologies or new ways of doing inspection for the existing what so workforce would be resisting because they might be worrying about the jobs and all that skill set development and cost optimize cost constraint if, if it is cost effective or not. With the new technology, I would say to conquer the workforce, less resistance, engaging employees early enough in the whole process itself, will give training and awareness because this is mainly for the safety reasons. These technologies are built and, workers can focus more on the data analysis part than the data collection part. You know, that is that is a key here.  
 
So it can definitely, get the buy in. I hope I have answered, both of from my perspective, I feel that's pretty complete on my end as well. So before I go into, our third topic, I would ask and remind our audience, please, any questions or comments you may have. Feel free to enter them. We'd love to see those coming in so topic number three and this one is sort of I know Ryan's getting excited because, this is sort of his passion area, unified platforms. So I'm going to kick it off, Ryan, to you with a question and then maybe ask Martha to chime in. How does having a unified platform for APM simplify or enhance operational efficiency compared to perhaps a traditional, or rather siloed? Yeah. And I love this, this topic because the whole ecosystem is, is changing.  
 
So I'll start with unified platform as a whole. And my background, not just energy prior to this is in financial services. So very large, large banks, right. And doing thousands and thousands of tasks, trillions of dollars. And this whole shape that happened was I'm going to use one platform for everything and keep extending that platform to do more and more and more and more and more. And at some point, what that industry realized is having one mega platform to do everything is probably not the best route, just in terms of vendor lock in, new technology roadmap. So overall, we used to see this push for a unified platform being 1 or 2 providers. So now when I say unified platforms, we're talking 4, 5 or 6, right.  
 
Providers in an ecosystem. And that's a blend of internal software development, external partners. Right. So how do you bring this together? So for APM when we say unified platform I want to keep tying it back to the data on the ground and then the outcomes for the organization, right. So the data on the ground when you have inspectors, right. Working with let's say our mechanical integrity solution, when you have operators, which is Rounds Pro right, or APM health, when you have monitoring diagnostic folks using predictive and prescriptive analytics, when you have robots and drones. Right. That Neha introduced. This creates a really interesting dynamic. Back to the data. And so what what we're seeing as I say unified platforms, there's going to be a huge emergence of organizations going to, like I said, 4 to 6 providers that those providers that need to work together to help centralize that data for customers, whether that's a data lake, whether it's you name it, we're going to see that kind of shift from this two three provider type space into more dynamic, right.  
 
And that's inclusive of the introduction of technology and machine learning operations. Data operations. APM. EAM. Where APM fits is across this is of getting that asset data that Neha and Martha are talking so eloquently about into the right place and then using that data. So where APM fits is kind of along the lines of your emissions management focus, your supply chain focus, your spare parts focus. We're at the heart of orchestrating that. So my prediction and then the trends that we're seeing is APM being a more dynamic part of a larger ecosystem versus more of a 1 or 2 platform approach to do the maintenance risk. Yes, I definitely agree, Ryan. And I think, the core of this and it's it's one of our, imperatives when we introduce products is when we integrating, when we're using data from this bill of systems which will always exist in our environments, there will always be an enterprise asset management, there will always be a historian.  
 
And yes, we need that data. We consume that data. But the question I would invite the audience, to go through every time that they are, considering a mega platform would be is this a case of one plus one equals three and what I mean by that is if you are integrating the systems, they already exist, they're into one of these platforms. And to be specific into our APM are the result that you are getting exponentially improved by the addition of that data set. When you capture data from one of the systems, what you are effectively doing is improving the quality of all the advisories and all the data that is delivered to the users. Are the insights that the higher quality are you able to do more with less? Therefore, are you creating some form of productivity by using and connecting these systems together? If the answer to that is no, I'm just moving data from one place to another, maybe that is not the best use of that integration effort.  
 
And the dollars from that platform. Fantastic. Martha, I'm going to do a follow up question about what are some obstacles to pulling the data together, and what can businesses do to address that? And perhaps am I making sense with that sort of follow up there? Yeah. Yes. Most definitely. And it's kind of an interesting question because as I previously mentioned, I've been doing this for 25 years. And when I started doing control systems for power generation, you know, the biggest problem was system integration. And here I am, 25 years later. And and the biggest problem is system integration.  
 
So yes, there's definitely, the constraints, the obstacles around bringing data together from different OEMs with different protocols intended to be used in a totally different way, and transforming that into a cohesive set of data for user consumption. So yes, bringing that data together. And then of course, when we talk about bringing data together, connecting systems, then cybersecurity a step in, you need to do that in such a way that normal deniability is introduced, and also in a way that does not increase the cost of ownership. We don't want to pass along to our users a huge increase cost of ownership, maintainability issue. We need to address that at the integration point. And the integration point is the APM platform. So again, through our essentials in our foundation, core of the platform, there is a lot of manpower that has been allocated to designing these integrations in such a way that delivers, ease of use that delivers productivity that delivers, more agnostic connectivity, in such a way that we can help alleviate some of those concerns. That is fantastic.  
 
And, Neha, I'm going to ask, you know, what are some of the the ways in which this unification, can contribute to better decision making? And a more holistic view of asset health. So some of the things come to my mind are, bring your own, you know, in terms of like AI or analytics, having, already created data science techniques or bring your own models to support the central AI platform to adding this, this to the, to the existing, existing scenarios, reusing and reusing, everything that, you have and then adding, new models. But, you can bring your own as well. So that's, that's where I see that the holistic view would help. And the central, view would help in this case. Fantastic. And, Ryan. Yeah. That's wonderful answer. Ryan, I've got two questions I'm going to, throw at you.  
 
These are coming from our audience. So thank you for participating. I'm going to read the the one here. Considering the challenges of legacy IT systems and the transition to microservices, what strategies can be implemented to ensure seamless interoperability between a historian and the composable services offered by this product? Assuming APM or APM, especially when managing multiple systems with varying data types and formats. Yeah, so a lot to unpack. And I always, my product marketer, I mean, we talk about seamless, right? Even with the switch to microservices, we're seeing a lot I wouldn't call them problems, but a lot of new ways to manage data as a whole, right. So microservices are super important to the flexibility and scale of what we do. And Martha just talked about essentials, which is what we consider our cloud based APM platform. And we underwent in V5 probably two years ago, a complete re-architecture to microservices for some of these things that Martha mentioned, right. The connectivity, the scale. So when it comes to a system like a historian, as a reminder, GE Vernova has a historian arm of our business, Proficy historian.  
 
So we have a really good understanding of historians, even not third party outside of GE, right? We work very closely with that team here. So when you think about the data integration piece, it's never going to be seamless, especially when you look at historians that maybe haven't been updated. The biggest pieces are those current historians systems on the latest version, are they leveraging advanced architecture or do they have good data management practices? From an APM side, I think Martha started to talk about this is we really focus, whether it's an API or a data loader. We have these built connectors that help to actually build the hierarchy between your historian data and your APM to leverage that. And maybe Martha can go deeper into how it manifests itself. But when you think about the support on the microservices, you have a cloud historian, a cloud APM, way easier than having an on premise historian and a cloud APM, right. At some point with unified platforms just where the space is today, there will be some sort of services work, whether it's your own team, external, to help support this, because this is the final unified platforms. There's a great vision for it, but everyone manages the data differently. So the way we do it is through data loaders and APIs.  
 
We have a lot of experience doing those integration projects, especially in our SmartSignal product and others. But the microservices help. But that's only if you're other providers are also leveraging microservices, right? To allow for that connectivity. So on prem to cloud, it'll always be a little interesting, which is why I stay away from the word seamless, but cloud to cloud, obviously a little bit easier. So hopefully that answers it. So the way we bring it in is through those pre-built data loaders and the way we action it is within our application. So health, reliability, that that's all running for the composability piece. And connectors probably that's the only missing piece. So we have various specific connectors dedicated to those pieces of equipment that we know we will encounter.  
 
So yes, definitely for historians legacy system through SCADA. That's where we see some of the older, oldest technologies that went on there that we integrate. So we have very specific connectors dedicated to the real time data and others. So again, it's very fit for purpose. But as long as that deep understanding of what is included on that enterprise, what's on the power plant, what's on that oil and gas facility. We managed to include various specific connection connector capabilities for those assets and sometimes, yes, we encounter legacy, not updated systems. That's not ideal from your cybersecurity perspective, but we know they're there and we make every possible effort to enable those connectors to gather the data. Perfect. And, Ryan, I know that you can see the question with me. Do you feel you've also covered the other question as well? I would add a little bit to this one around the managing multiple data sources for critical dependencies and then unified platforms. What are the success factors? So what we're seeing, I'm gonna put this in a SaaS perspective, which is the way the world is going in a lot of areas. Right.  
 
Managing data through unified platforms. What we're seeing right now is the rise of systems like snowflake, Databricks and others, which are kind of integral both data operations and data lake type of technologies, right. So when you're trying to unify platforms, I think it plays into the last question, which is your data needs to be in the right hierarchy across systems. You need to understand what problem you want to solve with that data. So for APM, Martha can talk about the problems we're solving. If you're thinking about EAM and your spare parts or that's a whole other problem. So when you're trying to manage multiple data sources, it starts with the problem to be solved. And then there needs to be a system in place that helps with that. So APM does a great job at managing EAM historians CMS data when you start to bring in other business platforms like a... You name it right outside of your asset operations and you want to contextualize, let's say emissions next to your APM.  
 
We're seeing a rise in that kind of a data lake data ops approach. So it starts with understanding the data you have to get together. And if you're unifying how you deploy your platforms. Being SaaS, we're seeing a huge lean towards the data lake approach, which again is a big learning curve, for a lot of energy organizations. And it's what financial services went through 5 or 6 years ago. So hopefully that answers the critical success factors are understanding your core system that's going to manage that data, understanding what data you need to solve what problems, and then understanding the subset of that data that how do you move that data where it has to go to solve that problem. So there's three levers of success, I think from a data lens. I know Martha, you might have thoughts on that as well. But that's kind of where we're seeing this heading is more of a data lake data ops approach when trying to bring Good stuff. I'm going to I'm going to switch gears here. We look like we're tracking right where we should be timewise, which is fantastic. And we're going to talk we're going to, we're going to move on to topic number four and talk about sustainability as a KPI.  
 
Now, this one, just as you are passionate about, centralized and unified platforms, I am passionate about sustainability as a KPI. I find it fascinating that, you know, APM as a, as a practice, has been around for decades. And at its core, it's been, spoken about as operational excellence, which in my mind is synonymous, absolutely synonymous with sustainable, more sustainable, operations. Just as Martha said earlier that a reliable, process is a more safe process. And operation I feel, operational excellence and achieving that by definition, helps you with a more sustainable operation. So some of the things that I'd like to unpack here, and we're hearing it from even the analysts, whether it's Verdantix or Gartner, right. They're talking more about APM as a foundational driver to a more sustainable operation, to decarbonization. How how are we seeing some sort of tangible examples of how APM can help track and improve sustainability metrics along those traditional KPIs like uptime and utilization? Anyone want to take that one, Martha or Ryan perhaps?  
 
Sure. I'll go for it, Tracy. And you are exactly correct. There is a misconception that sustainability needs to be this totally separate and very costly effort, and that is not so. This is one of those cases where, you know, historically, we've seen conflicts between key KPIs, and that's just not the case here. Sustainability aligns beautifully with other organizational KPIs. And optimizing performance is one of those key parameters. So if you're reducing your hit rate, there we go. That's your sustainability. If you are embracing a condition based maintenance program versus a time based program, your waste is greatly reduced. There is no parts going into dump yards. So any and all of these key KPIs around a better performing asset, a more reliable asset, have great contributions into that carbon footprint and into compliance. So again, it all happens by default by aiming to be, more performant and definitely having the tools to get there. Awesome. Neha, any thoughts?  
 
I guess just to add like optimizing the asset utilization so you can use APM for that because it beautifully helps you, you know, managing that as well and balancing the performance with the environment, in fact. And load distribution, predictive maintenance is the key, where it can predict the failures. It can avoid or minimize the downtime and resource usage or resource allocation. So and in both cases, APM can help. Fantastic. Ryan, any thoughts on on where you're seeing, sustainability being more embedded into APM and vice versa? Yeah. The trend that that I'm seeing is the is the top down asks to be a more sustainable and reliable organization, right. So corporations have their sustainability goals. And now plants and sites are working through those tasks to help support that goal. So I think to what Martha and Neha just said, that the trend is really thinking about utilizing asset maintenance as a way to help impact those KPIs for your organization, because it's likely where it comes down that sustainability needs to be a part of how they go about their daily work.  
 
And APM, I think has everyone covered, is a great way to start that journey, of actually reporting out on your impact of sustainability. So I think it's a it's a nice organic condition for some plants and sites being tasked with helping to hit this overall sustainability number or goal for their company. It's a good step change, for them to keep an eye out on. Fantastic. I'm going to I'm going to close out our fifth topic here. I think everyone if anyone is surprised by this topic, I'll be surprised because it's, it's the topic that just everyone can't get enough of. And that is. Yes, folks, we're going to talk about AI, specifically AI synergy. So Neha, I'm going to kick it off with you and ask, How, how AI can be used to create more proactive APM systems, particularly in predictive and prescriptive maintenance. Sure.  
 
So using AI artificial intelligence, you can it's main. It's developed for identifying patterns to predict outcomes. And it can suggest or autonomously implement any actions that are needed. So you can. So earlier I talked about agentic AI that can actually suggest you what actions you take. So by by the name AI and ML machine learning. You can you can identify the patterns, predict the outcomes, analyze the results, and then you can implement those actions. So I believe that both prescriptive and predictive can be done using the, you know, proactively. You can you know, you can use AI for, for for these use cases. Fantastic. And what are the most exciting developments you believe are going to revolutionize APM in the near future relative to AI? Yeah, that's that's my favorite topic. So emergence of, you know, LLMs, large language models and gen AI that we hear about.  
 
So I see that, you know, there's some of the trends that I see, you know, using these, technologies are, historical, analyzing the historical maintenance records, those documents, you can use Gen AI and start asking questions to those documents. Using the LLMs. There is something called as large vision models. Those are similar to large language model, but mainly developed for the, visual inspection. Right now we see a trend in the manufacturing processes. However, we should keep an eye on oil and gas and chemical or, those industries as well. And then we also see the agentic AI So integrating the diverse sources of data, including embedded hardware and visual data.  
 
And then, you can provide the actionable insight for the subject matter expert. So the way these, large region models or large language models work, they will give you generic outcomes. But then we have to add our SME knowledge to that. On top of that, like our APM, some of our APM solutions to get that, get the actions in hands of our, maintenance program. So, and then another trend that we need to keep an eye is around the real time data analytics or on prem, analytics, which is like the turnaround, especially for the manufacturing side of the business, where we, we should keep an eye and. Yeah, that these are, like, very highly emerge of these, new AI technologies that we need to keep an eye on. Right. Martha? How can organizations ensure that AI driven insights from APM are explainable and actionable for all stakeholders? Yes. Well, that has to do with some misconceptions about AI and with the way that we ambition using AI inside our portfolio.  
 
And those decisions are always driven by how can I deliver productivity in the hands of the human? We never have in mind here any big use of AI that would replace the human. And that again, I feel that is the biggest misconception. AI is a technology, like many others that we apply and employ in our products, in the service of our users, and as such, being able to explain to operators and maintenance personnel the quality and accuracy of these insights that is now giving you direction towards a higher priority is derived from artificial intelligence. Now you make the decision, you take the action. So again, the trick here is always to place the technology in the service of the human and give them the productivity. Make their life easier, accurate and therefore safer. That would be my spiel to the workforce to let them know this is in support of their goals. Fantastic. Ryan, I know, I know, you've been doing a lot of work with multiple folks on AI, both within our own business and the broader GE Vernova ecosystem. You want to give us some thoughts on AI synergy? Yes. And I always like to start off with a view of the market because everybody gets so caught up in the hype.  
 
I think to Martha’s point is the reality of AI. And so when we're out talking to customers, to partners, internally, it has a lot of promise, right? But some of the rush to get, let's say, generative AI systems in particular into place, what's starting to happen is kind of the threshold of return, right. And I think it goes back to the question that was asked about managing all your data based data practices. AI starts there. You can't slap a AI on to bad data. You can't slap AI on a siloed data. You can't slap AI on to bad process. That's just going to lead to more bad process, more bad data hallucination, bad outcomes. Right? So thinking about AI from that lens, what we're starting to hear from customers is they want to start with a very particular use case. And I'll use outage planning as an example. This is a recent example talking about GE’s vision on how it can help it outage planning. And that impacts not only what APM can deliver in terms of the asset analysis, predictive maintenance, spare part planning. It also impacts I'll use a grid example dispatch and T&D.  
 
So we're starting to see on the AI front there was this big vision of transformation. A lot of conversations have been had and now is starting to turn more into GE. Where would you start and how can you help versus a big single blast? I'm going to change my organization with AI, so that's the fun part of AI. And I think I always tell people coming from the SaaS side particularly, you got to be really careful on how this stuff is branded and talked about, right? Like agenetic. Are we looking into it at GE? Yes we are, but is agentict a new version of RPA with a little bit more glitz and glamor. Absolutely, right. So understanding what AI can actually impact, what outcomes you want, I think Neha said it really, really well with large vision models and large language models.  
 
That's one subset of the stuff. So you also have to understand what part of AI is going to help meet your needs versus trying to leverage something like generative AI for everything. Because there's machine learning, there's neural networks, there's a whole world out there. I know Martha can talk about some of the closed loop stuff. There's a whole world out there that can probably help you solve your problem. So I find there was this big excitement to back down into where it was three years ago, which is we need the right technology for the right use case. It can't solve everything. So that's what we're hearing. And we're seeing that as well, with customers asking us for our input on how AI can be used internally for worker efficiency and then also, for their asset efficiency.  
 
So a lot happening. And my big thing on the trend is you got to be realistic. You got to understand how it works with the data you have. And I think to Neha’s point, keeping the SMEs in control is important. I think going towards an autonomous enterprise as an energy org, is it realistic? So it's how do you get the most out of your people and how do you make your people feel empowered with new technology? I, I love it and I love where we ended our discussion today. I think, you know, some of the very pragmatic thoughts from all of you. I want to thank you, Martha. Neha, Ryan, for your thoughts on these very important topics that no doubt we will continue to see unfold as we move through this year. And maybe some surprise, other topics creep in.  
 
I want to thank you, our audience, for joining us today and also promote the fact that registration is open for our annual Global user conference. Coming up the first week of April in Houston, we encourage you to join us there. You'll be able to interact face to face with, with our esteemed panel and many others from our organization. And more importantly, you'll get to dialog with your peers. And, talk shop and, find out, you know, where, where things are similar and where things are not, and learn and grow with each other. So I encourage everyone, just to check out registration, learn a little bit about our event, and reach out with any questions you may have. But thank you again for your time and attention. Until we see you again. Thank you. Thank you.