Many factors necessitate the modernization of industrial control systems. These factors include performance limitations, loss of expertise with the system, component obsolescence, and overall system reliability. While the transition has historically been challenging, tools and applications have been developed to simplify and reduce risk in these modernization projects.
In the Emerson Automation Experts podcast, I speak with Emerson’s Aaron Crews who leads our modernization team. Aaron discusses a new application, REVAMP which is a groundbreaking, cloud-based, advanced software solution that helps manage the transition of legacy applications to optimal control.
It streamlines workflows, minimizes risks and errors, and reduces overall project costs. Embedded artificial intelligence helps ensure continuous improvement in efficiency and quality as plants transition to DeltaV™ distributed control and safety instrumented systems.
Listen to the podcast and visit the DeltaV REVAMP section on Emerson.com for more information on how REVAMP can help with your control system modernization efforts.
Transcript
Jim: Hi, everyone. I’m Jim Cahill, and welcome to this edition of the “Emerson Automation Experts podcast.” A while back, we had Emerson’s Aaron Crews join us in a modernization and migration podcast to discuss how DeltaV IO.CONNECT simplifies the process of control system modernization. I’ll link to that in the transcript that will accompany this podcast. If you’re searching around for the podcast, it’s titled “Simplifying Control System Modernization With DeltaV IO.CONNECT Podcast.”
Today we’ll continue on the modernization and migration theme, and talk about another offering within the portfolio, REVAMP. Welcome, Aaron.
Aaron: Hey, Jim. Thanks for having me.
Jim: Well, it’s great that you’re joining us today. Can you spend just a bit of time today grounding us in what you do and how the modernization offerings you’re talking about benefit our customers?
Aaron: Sure. So, I work for Emerson, obviously, and I lead the modernization team, which our job is basically to help customers move from older DCS, SIS, and PLC-based systems to DeltaV and DeltaV SIS. So, these are, obviously, pretty challenging technically, these projects, pretty complicated, but we do a lot of them, which is kind of good and bad. We do over 400 of these projects a year, so there’s a lot of complexity to deal with there. At the same time, we get a lot of experience through executing all of those projects. The most important thing is that we’re learning from them. So, we’re learning from that experience and we’re not inventing things along the way.
So, my team’s job is to help make sure those go smoothly. And we do that in some ways by just developing standards and new technologies that address the most common challenges. DeltaV IO.CONNECT, you mentioned, we talked about last time, that was one of those technologies that was all about a superfast cut-over DeltaV control, and also saving a lot of cost on those projects.
Jim: Yeah, it sounds like DeltaV IO.CONNECT was about making a seamless hardware transition to DeltaV, but what about the software? What are some of the challenges customers may have?
Aaron: It’s a good point. The hardware transition, it’s important, obviously, but it’s just part of the challenge. There’s also the fact that the new system actually needs to be able to control the plant and hopefully control it a lot better than before. So, you do have to deal with that software piece, and a lot of customers are really concerned about that. They don’t want to break something that’s running the plant today. The worst thing that happens is you do this project and things are somehow worse or broken because of something you didn’t know was happening in the old control system. And that’s usually the case.
Many users don’t know everything that’s happening in that old control system. They’re put in usually by other people, usually decades ago, and there’s a lot of complexity inside of that. So, you kind of have this challenge where you could spend a whole lot of time figuring out what exactly is happening in the old control system, or you can just implement something that’s brand new and potentially miss something that was really important that’s happening today. So, that’s kind of the situation a lot of folks are in. A lot of customers just say, “You know what, the least risky thing is to do a like-for-like functionality of what’s happening today is what’s going to happen tomorrow, and so therefore, we want some kind of conversion tools that are going to save time and ensure that like-for-like replacement happens.”
Jim: I can definitely see the appeal of converting configurations from one system to another. But won’t that limit the value you get with a modern control system like DeltaV?
Aaron: Absolutely. I think I’ve never been a fan of these types of conversion tools for exactly that reason. If you just do everything the way you were doing it before, then you’re not doing anything better. So, here at Emerson and with my team, we decided that tools are important, we have to deal with these challenges, but let’s take a new approach. Let’s take a fresh approach so that we don’t have the same, you know, value-limiting aspect of it if we’re going to use tools on projects. And so we developed a technology that we call REVAMP, and we think that kind of gives you the best of both worlds. Creating a lot of efficiency through automation and automated sort of transition of that old code to DeltaV, but also enabling the value that you’d get from a new DeltaV DCS.
You know, technology’s come a long way over the years, come a long way from scripts and macros and things like that that engineers typically use. And now we can take advantage of some cutting-edge technology and put it out to work, automatically document the old system, automatically functionally replace the old code with new standards in DeltaV, and enable all those modern features of a new DCS, and of course all the value that that brings.
Jim: REVAMP sounds very interesting. What are some of the cutting-edge technologies in REVAMP, and how does it work?
Aaron: Well, the first part of it is that REVAMP uses the cloud. So, cloud’s been around for a long time, but, you know, maybe not so much in the control systems world. But in these projects, it’s a great opportunity to take advantage of the cloud and centralize everything. And then from there, we’ve got a lot of computing power. So now we can start to use newer technologies like artificial intelligence, which is going to automate the analysis and help streamline the work, as well as machine learning. So, there’ll be parts of this code that we don’t recognize and we need to understand and identify and deal with, and the ML algorithms help us with that.
Jim: Seems using the cloud makes a lot of sense. So, what does that mean for everyday projects, you know, the team benefits from the other projects that are going on around the world?
Aaron: That’s exactly right. Since everything’s centralized in the cloud and since we have this machine learning concept that’s baked in, now if you’re a project team that’s in Singapore, and you do work, then that helps REVAMP get smarter. And then we do a project in Houston and that takes advantage of the knowledge that we learned through our experience in Singapore. So that centralization, not only does the typical cloud thing of letting every project get the new features as we continue to enhance REVAMP, but it also allows the tools to get smarter, and therefore every project gets more efficient and less risky.
Jim: It seems like with that centralization in the cloud, and you mentioned artificial intelligence, AI there, and that’s just a huge topic lately. So, how do we use AI with REVAMP?
Aaron: Yeah. You know, calling that a huge topic is like an understatement. I feel like my Twitter feed is just pure AI content these days, but…
Jim: You and me both.
Aaron: Absolutely. So, yeah. We use it, I would say in three main ways. One is to do the analysis. So, when the legacy control system files get run through REVAMP, there’s AI there to break all of that apart and analyze it and to find the clusters of reusable code and identify their functions and automatically assign things, and all of that kind of stuff. Then there’s this piece of what we don’t recognize, what’s not part of our experience and how do we learn from that? And so, we use ML on that piece of it. That’s interesting because on an individual project, we might see one or two cases like that, but across all 400 projects we’re doing that year, we might see hundreds of cases. So, we can see what’s common across those and make the tool smarter from there and continue to address more and more of the work that’s required.
And then the last piece, which is the newer piece, which is probably the reason our Twitter feeds are blowing up, has to do with the use of large language models. So, the old control systems, a lot of them had structured text type capabilities where you could write your own scripts to do features and functionality that weren’t part of the other parts of the control system. And REVAMP uses large language models, you know, very similar to something like ChatGPT to read through that old structured text and to automatically develop control narratives that are easy for engineers to implement the new control system.
Jim: Now that sounds really cool. One that, you know, it’s all being gathered here in the cloud and then these different areas of artificial intelligence, machine learning, the large language, you know, working at getting smarter and smarter on it. My only hope is it doesn’t find any of my PLC code on offshore platforms from back in the ’80s. I think that might send it in the wrong direction. So, what kind of impact does all this have on our projects?
Aaron: Well, the biggest is that it de-risks the project. Modernization projects are all about managing risk. As we said before, customers, they don’t want to break anything. So, this gives you the ability to fully account for everything that’s happening today, but also get the modern features that you’re looking for. So, it de-risks in the way that customers are looking for from a conversion tools point of view, but without the same drawbacks really. The other aspect of this is project mistakes. So, you know, you were talking about your old PLC code and, like, I’m not ashamed to admit, maybe a little ashamed to admit, I’ve maybe made some mistakes in my engineering back in the day. Obviously, we put all the checks in. We double-checked, triple-checked the work to make sure that none of those mistakes hopefully make it into the actual plant, but that’s a lot of work to make sure that there are no human errors along the way.
And so, if you can have a fully digital workflow where you didn’t get hands-on and accidentally fat-finger something or put in the wrong thing, make the wrong decision, then you reduce or eliminate the possibility for those mistakes and you reduce the amount of testing and double checking, triple checking that maybe is required for some of those aspects of the project. And so, those mistakes, they tend to go away. And that helps improve schedules. So, we used REVAMP on a recent project and we took like 11 weeks out of the schedule. And a lot of that was because we had the base configuration done in an automated and accurate way much earlier than we would have on a normal project.
And so, that allows for engineering efficiency. So, you got the schedule improvement, but then you have engineering efficiency as well, which allows engineers to spend the time on the work that’s most important, which really is the application of the modern technology, what do you want to do? How do you want to get more out of the facility? And focus on those efforts instead of digging through old code.
Jim: Wow, 11 weeks on a project schedule. That’s really impressive. And yeah, whatever the tools can do to eliminate human error because unfortunately, we’re not all perfect, even those of us engineers that sometimes think we are. Who can and will benefit from the use of these tools?
Aaron: Well, you know, as an engineer and techie geek person, I like to just use these tools because they’re fun, they make work easier, more exciting, more fun, more powerful. But ultimately, it is the customers that are benefiting. They’ve got these projects to do, they’re, you know, normally full of risk. And so, you know, taking that out is great. Improving the capital efficiency of these projects is great. And then of course, being able to deliver modern control, the modern features of DeltaV, and also to help standardize.
So, one of the things that we always hear about is that the way these old systems are programmed depended on who programmed them, you know. And so, you might have dozens of ways of doing interlocks and motor control or dozens of ways of doing PID control in the system, just depending on the vintage and who did it. And so, in this case, you can standardize all of those things. And that helps make the new control system easier to maintain, easier to expand, and help you, you know, your engineers work across systems more easily.
Jim: Yeah, I think that’s an important point. It’s not just about the project efficiencies, but the ongoing through making it more standard, getting rid of, you know, the personal art that’s involved that may be difficult for somebody else to figure out. Yeah, that’s huge. Let’s wind it all down. So, how does a customer learn more about how REVAMP can help them with their projects?
Aaron: Anyone can go to Emerson.com/Modernization and check out all the offerings, including, there’s a page there just on REVAMP. And there’s a video on that page, as well as on YouTube if you just search for Emerson REVAMP that runs through all the details of how this works as well. And then you can either reach out just through the website, I think there’s a link there, or reach out to your local Emerson contact. And we’re always happy to run REVAMP, so we can run it on your configuration, check out your code, see how it can help, and go from there.
Jim: Well, Aaron, this has been eye-opening, fascinating, the technologies we’re employing here and the benefits that they can really deliver to the project teams and the customers in the end through their lifecycle. So, thank you so much for joining us today.
Aaron: Yeah, thanks so much for having me.
-End of Transcript-
