Presentation: Brewing Innovation: Inside New Belgium’s Use of AVEVA Operations Control

2024 - AVEVA World - Paris - HMI/SCADA

Brewing Innovation: Inside New Belgium’s Use of AVEVA Operations Control

Alright. Good morning, everyone. Thank you so much for being here. My name is Rachel Dwinal. I am the HMISQiva technical community development lead here at Aviva. I'm also the owner of Heroes HQ, our HMI SCADA community platform. If you're not a part of it, strongly encourage you to sign up. It's our technical community platform that is available for customers, partners, students, you name it, anyone that has anything to do with Aviva HMI, a place for you to come together, ask questions, get to know one another, and network with our community. So super excited to be here today with Frank Lalibera from New Belgium Brewing. So, Frank, thanks so much for being here today. Oh, my pleasure. Thanks for having me, and thank you everyone for coming to my talk. I have a captive audience now. My name is Frank Lalibera from New Belgium Brewing, and I'd like to talk today about three things. I'd like to talk about software architecture and planning for a very, very ever changing world with new cybersecurity and new products from Aviva, new customer demands, new processes, all that stuff. I'd like to talk about the, well, I'm gonna call it outdated nature of the crawl, walk, run model of very linear thinking and how we can think outside the box with that. And I wanna talk about how Aviva operations control has helped New Belgium make some great strides in those couple things. But before I do that, I'd like to talk about New Belgium Brewing, give you a little backstory. So we were founded in nineteen ninety one by an electrical engineer named Jeff, and his wife, Kim. We have been very, very passionate about engineering, very nerdy, very, very invested in our own engineering, much more so than many breweries from the very beginning, and that persists today. We've used Wonderware products for a lot longer than I've been involved with them, which is great. So that plant, of course, started very small, and we expanded and we expanded and we expanded, and every expansion comes new hardware, some new software, new pieces, new people. So the the plant grew very organically. And then in twenty sixteen, we built a plant in Asheville, North Carolina. That plant we built from the ground up. So we got to design everything. We got to make really good network diagrams and electrical drawings, very standardized control system, which is great. Very, very satisfying to have that and to build that. In two thousand twenty one, we acquired a brewery in Comstock, Michigan called Bells. They had some standards and a little bit of drawings and some similar automation, no Aviva products. So a bit of a learning curve. Right? Acquiring a business that is different than yours. And then in twenty twenty three, we acquired a, a brewery in Daleville, Virginia, and that brewery has very old siloed automation, lots of controllers from the from the nineteen nineties. Every single standard you could ever imagine, every cabinet looks different, every single piece of hardware is different, every piece of software is different. So very little for standards. So, again, another very different situation. So you talk about the many different kinds of technology at across the different breweries and whatnot. Talk to us about a little bit about the different stages of life that are at these different breweries. Yeah. Absolutely. As I alluded to, if you build a brewery from scratch, it can be a lot further along. If you design everything intentionally, it can be a lot further along than if you acquire something that didn't have the intentionality. And, of course, it's where we find ourselves. So we find ourselves with our everything from a very immature plant in, from an automation perspective in Virginia to a very mature plant in terms of automation in North Carolina. So they're in these different stages, and very typically, we call them, you know, the crawling stage, the walking stage, the running stage. And at least from an automation perspective, our plants were in very different stages with that. And, yeah, we're we're very, we're very aware of of how how challenging it can be to make production look the same, feel the same, get the same numbers out of it in very, very different places. So how do you bridge that gap then between the lesser mature operations to the more mature ones? What does that look like? Yeah. So classically and no. Your eyes aren't blurry. They're wavy because it's not very clear paths. Typically, you take that those you know, the old school way of thinking this is you take something that is like those very style of automation things, and you put in your standard piece of hardware and your standard code and your standard, you know, HMI, your standard or Veeva products, and you build all of your stuff up slowly one piece at a time, and then everything is correct. And believe me, I would love nothing more than to get five years and you say, okay. Make it right. And all the network diagrams will be correct, and all the electrical diagrams will be correct, and all the pipes will be in a straight line following each other instead of all over the place. It would be very, very satisfying to do. It would lead to a better product. But the reality of the situation is that you well, a lot of times when you acquire a plant or something, it needs to run it, like, a hundred percent percent right now, and there's not enough money to do that stuff, and there's not enough time to do that stuff. And you you have to make concessions and pick the most important parts of doing that. So so in, in this example, of our Virginia plant, we had a, of our Virginia plant, we had a, for the PLC guys in the room, a slick five zero five. So an old sixteen bit PLC, from nineteen ninety nine on an old cam line. And the cam lines are very, very important to our business, and they, thank you. They we, of course, want to look at how it's running, and I can't stress this enough. The HMI that's on this thing just emulates a panel. There's a start button. There's a stop button, and there's how fast are you going. That's it. So there's no data. There's no performance. There's no OEE. There's nothing like that. It's very, very old school controls. So, again, the old school way of thinking of this would be rip that out, put in a new control system, build a new screen for it, get the performance data you need out of it. But, again, that takes time, that takes money, that takes engineering hours, that takes a lot of talks around standardization, maybe network changes. Instead, what we did was we took the data out of that via Modbus. It was a little painful. It's sixteen bit, so we were doing a lot of bitwise math to count numbers as high as our can counts higher than thirty two thousand seven hundred fifty five. And we brought that into our Galaxy, historized it, pushed it up to Insight, got Insight OEE on it. So we've got this piece of twenty five year old technology that has OEE on your phone on it without having to redo every single part of it, without having to spend a lot of money and a lot of time doing this. It takes very little time. So you talk a lot about these challenges and and having to overcome them. And so how do you skip over these difficult steps that were traditionally necessary? Right. So I alluded to this, but because we have such a robust plant in our North Carolina plant, since it's so far along in automation, we get to play with the cutting edge tools. We get to go to these conferences and say, oh, that might help. That's actually the one. And this plant has engaged people and very good data and very robust automation and networking. So it's very relatively low barriers to go do this. It's very easy to go do this in comparison. So you can get those solutions dialed in in our in North Carolina plant. And then once you I'm good at implementing them, and we, as a company, are good at looking at them and knowing what we want, having a good definition for what we want, it's much easier to take that and apply it to an old piece of technology and and not do the traditional route. Awesome. Talk to us a little bit about you know, we talk about flexibility, and scaling. Talk to us about scaling projects over time. Yeah. So, the other thing I alluded to at the beginning of this talk is that I'm going to talk to you about architecture. Architecture is a really tricky one. Everyone at every company has to think about what their server architecture looks like, what their license counts looks like, all of that stuff. And the again, the traditional way of doing this is you take what your future state you think is going to be. You say I'm gonna have a hundred thousand tags and a hundred users, and we're gonna use these products. And you go to your Aviva partners, and you build a license plan, and you go to management, and you go to the financial portion of your company, and you you you, you know, play some numbers games, and you figure it out, and you come up with a solution of where you want to be someday. You buy that. You build those servers out, and then you grow into them. And I think that's a medium or a bad way of doing things. I think that it's very, very hard to plan for a future when the future can change so much as we all know. The cybersecurity can change. The process can change. Your mergers and acquisitions might change. The world changes a lot around you at Medeviva. Aviva changes a lot too. They have new products and new capabilities all the time, and it's really hard to plan for something well that is that far away. You don't know future you, so you can't plan for future you. You don't know your future company. You cannot plan for your future company. I think it's better to do very small bites, and I'm bear with me for one minute on this. I know that this is not an easy way to do this now, and I'm here to pitch why Aviva operations control helped. I think it's better to say, well, we only have a hundred points today and one guy who uses the screens. Let's just stand up a super small galaxy. And then as we grow into it, we'll make it bigger. We'll add another server here. We'll add a redundant pair of AOS servers, and we'll add a redundant pair of RDS servers for the HMI app. Once you actually grow into them because, again, your needs might change, the business demands might change, everything around that might evolve before you actually grow into it. And if you have a flexible license model, like operations control offers, where it says, use whatever products you want, use whatever architecture you want, however many servers you want, we'll just count the number of users. And if you have a flexible license model, like operations control offers, where it says, use whatever products you want, we'll just count the number of users, then you get to actually right size everything you're doing on the fly. You get to right size everything every step of the way, which means you have a much better chance because they're small little steps of having a correct architecture. Right. And I appreciate you speaking about that because I think that there's a lot of people that are probably in this crowd too that are smaller smaller manufacturers, come from smaller businesses. So what advice can you give the smaller manufacturing companies, that are looking to scale their operations over time? And and what advice can you give them to be successful in doing so? Yeah. It's a little bit of what I've already said. It's it's making sure that you're staying kind of nimble with it. As a small company, your advantage is you can move faster than a big company. So you can you can adapt to some of these changes much quicker. And what that means is having the flexible sense. I will we don't have to have this exact structure. The structure can change if it makes sense to change. We don't have to have this set of servers. That set of servers can change if it makes sense to change. And if your licensing model says, it doesn't matter what you do, what new things you try, how many servers you try. If you say, maybe this will fix the bug if we have another set in here and there's less load. Maybe you say there's a new product and it's easy enough to try if it doesn't cost money, and I don't have to go to my financial people, and I don't have to go to the corporate, you know, projects thing and and make a sales pitch. You say, I could just install the software and see if it works. And then I have a business case, and then I have, here's how it really works. Here, look at my phone. Wouldn't this be cool? That's a lot easier to to do over time in small bites. And the the other advantage you get with kind of this constant rearchitecting is you you are very familiar with your own process this way. And sometimes, of course, at large companies, if someone one person stands up every server and everything and then you have to maintain it or you're the person who stands up and someone after you has to maintain it, they might not even know how to do all of that. So it helps the company be more flexible. It helps you right size things more, and Aviva operations control is a great shortcut to being able to take those steps. Awesome. Well, thank you so much, Frank, for being with us and for sharing your story. The new Belgium story is always awe inspiring. So thank you so much. So we wanna open it up to you all for questions. Definitely, we have extra time to be able to answer any of those that you have, and Matt will run a mic to you if you have a question.

Hear the remarkable journey of New Belgium Brewing, one of the largest craft breweries in the United States, as they harness the power of AVEVA's comprehensive software solutions to revolutionize their brewing operations. Learn how New Belgium Brewing integrated AVEVA’s MES, Insight (cloud), and other Operations Control tools to gain real-time insights and control over their brewing process. Operational Efficiency: The strategies employed to optimize production workflows, reduce downtime, and maintain consistency in beer quality. Sustainability and Innovation through minimizing waste and energy consumption. Future Prospects: The ongoing benefits and future plans for further leveraging AVEVA's technology to stay at the forefront of the brewing industry.

Download Presentation

Industry

Consumer Products

Company

New Belgium Brewing

Speaker

Ray Matthews

Company

New Belgium Brewing

Speaker

Frank LaBarbera

Session Code

AW24-HMI-D2-SESS-484

Transcript

Alright. Good morning, everyone. Thank you so much for being here. My name is Rachel Dwinal. I am the HMISQiva technical community development lead here at Aviva.

I'm also the owner of Heroes HQ, our HMI SCADA community platform. If you're not a part of it, strongly encourage you to sign up. It's our technical community platform that is available for customers, partners, students, you name it, anyone that has anything to do with Aviva HMI, a place for you to come together, ask questions, get to know one another, and network with our community. So super excited to be here today with Frank Lalibera from New Belgium Brewing. So, Frank, thanks so much for being here today.

Oh, my pleasure. Thanks for having me, and thank you everyone for coming to my talk. I have a captive audience now.

My name is Frank Lalibera from New Belgium Brewing, and I'd like to talk today about three things.

I'd like to talk about software architecture and planning for a very, very ever changing world with new cybersecurity and new products from Aviva, new customer demands, new processes, all that stuff.

I'd like to talk about the, well, I'm gonna call it outdated nature of the crawl, walk, run model of very linear thinking and how we can think outside the box with that. And I wanna talk about how Aviva operations control has helped New Belgium make some great strides in those couple things. But before I do that, I'd like to talk about New Belgium Brewing, give you a little backstory.

So we were founded in nineteen ninety one by an electrical engineer named Jeff, and his wife, Kim.

We have been very, very passionate about engineering, very nerdy, very, very invested in our own engineering, much more so than many breweries from the very beginning, and that persists today. We've used Wonderware products for a lot longer than I've been involved with them, which is great.

So that plant, of course, started very small, and we expanded and we expanded and we expanded, and every expansion comes new hardware, some new software, new pieces, new people. So the the plant grew very organically.

And then in twenty sixteen, we built a plant in Asheville, North Carolina.

That plant we built from the ground up. So we got to design everything. We got to make really good network diagrams and electrical drawings, very standardized control system, which is great. Very, very satisfying to have that and to build that.

In two thousand twenty one, we acquired a brewery in Comstock, Michigan called Bells.

They had some standards and a little bit of drawings and some similar automation, no Aviva products. So a bit of a learning curve. Right?

Acquiring a business that is different than yours. And then in twenty twenty three, we acquired a, a brewery in Daleville, Virginia, and that brewery has very old siloed automation, lots of controllers from the from the nineteen nineties.

Every single standard you could ever imagine, every cabinet looks different, every single piece of hardware is different, every piece of software is different. So very little for standards. So, again, another very different situation.

So you talk about the many different kinds of technology at across the different breweries and whatnot. Talk to us about a little bit about the different stages of life that are at these different breweries.

Yeah. Absolutely.

As I alluded to, if you build a brewery from scratch, it can be a lot further along. If you design everything intentionally, it can be a lot further along than if you acquire something that didn't have the intentionality. And, of course, it's where we find ourselves. So we find ourselves with our everything from a very immature plant in, from an automation perspective in Virginia to a very mature plant in terms of automation in North Carolina. So they're in these different stages, and very typically, we call them, you know, the crawling stage, the walking stage, the running stage.

And at least from an automation perspective, our plants were in very different stages with that.

And, yeah, we're we're very, we're very aware of of how how challenging it can be to make production look the same, feel the same, get the same numbers out of it in very, very different places.

So how do you bridge that gap then between the lesser mature operations to the more mature ones? What does that look like?

Yeah. So classically and no. Your eyes aren't blurry. They're wavy because it's not very clear paths.

Typically, you take that those you know, the old school way of thinking this is you take something that is like those very style of automation things, and you put in your standard piece of hardware and your standard code and your standard, you know, HMI, your standard or Veeva products, and you build all of your stuff up slowly one piece at a time, and then everything is correct. And believe me, I would love nothing more than to get five years and you say, okay. Make it right. And all the network diagrams will be correct, and all the electrical diagrams will be correct, and all the pipes will be in a straight line following each other instead of all over the place.

It would be very, very satisfying to do. It would lead to a better product. But the reality of the situation is that you well, a lot of times when you acquire a plant or something, it needs to run it, like, a hundred percent percent right now, and there's not enough money to do that stuff, and there's not enough time to do that stuff. And you you have to make concessions and pick the most important parts of doing that.

So so in, in this example, of our Virginia plant, we had a, of our Virginia plant, we had a, for the PLC guys in the room, a slick five zero five.

So an old sixteen bit PLC, from nineteen ninety nine on an old cam line. And the cam lines are very, very important to our business, and they, thank you.

They we, of course, want to look at how it's running, and I can't stress this enough. The HMI that's on this thing just emulates a panel. There's a start button. There's a stop button, and there's how fast are you going.

That's it. So there's no data. There's no performance. There's no OEE. There's nothing like that. It's very, very old school controls.

So, again, the old school way of thinking of this would be rip that out, put in a new control system, build a new screen for it, get the performance data you need out of it. But, again, that takes time, that takes money, that takes engineering hours, that takes a lot of talks around standardization, maybe network changes.

Instead, what we did was we took the data out of that via Modbus. It was a little painful. It's sixteen bit, so we were doing a lot of bitwise math to count numbers as high as our can counts higher than thirty two thousand seven hundred fifty five. And we brought that into our Galaxy, historized it, pushed it up to Insight, got Insight OEE on it. So we've got this piece of twenty five year old technology that has OEE on your phone on it without having to redo every single part of it, without having to spend a lot of money and a lot of time doing this. It takes very little time.

So you talk a lot about these challenges and and having to overcome them. And so how do you skip over these difficult steps that were traditionally necessary?

Right. So I alluded to this, but because we have such a robust plant in our North Carolina plant, since it's so far along in automation, we get to play with the cutting edge tools. We get to go to these conferences and say, oh, that might help. That's actually the one. And this plant has engaged people and very good data and very robust automation and networking. So it's very relatively low barriers to go do this. It's very easy to go do this in comparison.

So you can get those solutions dialed in in our in North Carolina plant. And then once you I'm good at implementing them, and we, as a company, are good at looking at them and knowing what we want, having a good definition for what we want, it's much easier to take that and apply it to an old piece of technology and and not do the traditional route.

Awesome.

Talk to us a little bit about you know, we talk about flexibility, and scaling. Talk to us about scaling projects over time.

Yeah. So, the other thing I alluded to at the beginning of this talk is that I'm going to talk to you about architecture.

Architecture is a really tricky one. Everyone at every company has to think about what their server architecture looks like, what their license counts looks like, all of that stuff. And the again, the traditional way of doing this is you take what your future state you think is going to be. You say I'm gonna have a hundred thousand tags and a hundred users, and we're gonna use these products.

And you go to your Aviva partners, and you build a license plan, and you go to management, and you go to the financial portion of your company, and you you you, you know, play some numbers games, and you figure it out, and you come up with a solution of where you want to be someday. You buy that. You build those servers out, and then you grow into them. And I think that's a medium or a bad way of doing things.

I think that it's very, very hard to plan for a future when the future can change so much as we all know. The cybersecurity can change. The process can change. Your mergers and acquisitions might change.

The world changes a lot around you at Medeviva. Aviva changes a lot too. They have new products and new capabilities all the time, and it's really hard to plan for something well that is that far away. You don't know future you, so you can't plan for future you.

You don't know your future company. You cannot plan for your future company. I think it's better to do very small bites, and I'm bear with me for one minute on this.

I know that this is not an easy way to do this now, and I'm here to pitch why Aviva operations control helped.

I think it's better to say, well, we only have a hundred points today and one guy who uses the screens. Let's just stand up a super small galaxy. And then as we grow into it, we'll make it bigger. We'll add another server here. We'll add a redundant pair of AOS servers, and we'll add a redundant pair of RDS servers for the HMI app.

Once you actually grow into them because, again, your needs might change, the business demands might change, everything around that might evolve before you actually grow into it. And if you have a flexible license model, like operations control offers, where it says, use whatever products you want, use whatever architecture you want, however many servers you want, we'll just count the number of users. And if you have a flexible license model, like operations control offers, where it says, use whatever products you want, we'll just count the number of users, then you get to actually right size everything you're doing on the fly. You get to right size everything every step of the way, which means you have a much better chance because they're small little steps of having a correct architecture.

Right. And I appreciate you speaking about that because I think that there's a lot of people that are probably in this crowd too that are smaller smaller manufacturers, come from smaller businesses. So what advice can you give the smaller manufacturing companies, that are looking to scale their operations over time? And and what advice can you give them to be successful in doing so?

Yeah.

It's a little bit of what I've already said. It's it's making sure that you're staying kind of nimble with it. As a small company, your advantage is you can move faster than a big company. So you can you can adapt to some of these changes much quicker.

And what that means is having the flexible sense. I will we don't have to have this exact structure. The structure can change if it makes sense to change. We don't have to have this set of servers.

That set of servers can change if it makes sense to change. And if your licensing model says, it doesn't matter what you do, what new things you try, how many servers you try. If you say, maybe this will fix the bug if we have another set in here and there's less load. Maybe you say there's a new product and it's easy enough to try if it doesn't cost money, and I don't have to go to my financial people, and I don't have to go to the corporate, you know, projects thing and and make a sales pitch.

You say, I could just install the software and see if it works. And then I have a business case, and then I have, here's how it really works. Here, look at my phone. Wouldn't this be cool?

That's a lot easier to to do over time in small bites. And the the other advantage you get with kind of this constant rearchitecting is you you are very familiar with your own process this way. And sometimes, of course, at large companies, if someone one person stands up every server and everything and then you have to maintain it or you're the person who stands up and someone after you has to maintain it, they might not even know how to do all of that. So it helps the company be more flexible. It helps you right size things more, and Aviva operations control is a great shortcut to being able to take those steps.

Awesome. Well, thank you so much, Frank, for being with us and for sharing your story. The new Belgium story is always awe inspiring. So thank you so much. So we wanna open it up to you all for questions.

Definitely, we have extra time to be able to answer any of those that you have, and Matt will run a mic to you if you have a question.