John S. Rinaldi's Ode to Automation

2 – What IoT and Industry 4.0 Means to Manufacturers Today

John S. Rinaldi Season 1 Episode 2

Industry 4.0 and IoT are some of the biggest buzzwords in technology, but what does that mean for manufacturers looking to merge their plant floor with that 21st century technology? 

Join host John S. Rinaldi and special guest David Schultz, a consultant and 25-year veteran in process control automation, as they discuss where manufacturers are at today with the adoption of IoT and Industry 4.0, including: 

  • The difference between Industry 3.0 and Industry 4.0
  • What MQTT means for manufacturing
  • Important Industry 4.0 considerations for getting the right strategy and partners in place

Connect with John:
Email | LinkedIn | Newsletter | YouTube | Blog | Books

For all things industrial automation, visit rtautomation.com. 

John Rinaldi:

[ Intro Music] H ello, this is J ohn Rinaldi and here with the digital connectivity podcast and today's session, we are going to be talking to an eminent consultant and expert, and, oh, he's got an incredible resume. His name is David Schultz, and we're really glad to have you on w elcome David.

David Schultz:

Thanks, John. Thanks for inviting me on.

John Rinaldi:

Well, David's a great friend of mine and he's a, he's been to our office. We've sat around and talked about IOT for a long time, and I thought he would make some, uh, great. So we can have, so we can have a great conversation and we should publish it so that other people can listen in on it. David, for people who don't know you why don't you tell, you know, give a little, uh, history of, of, of what you've been doing for the last, what is it now? 70 years you've been in industrial automation?

David Schultz:

Right? Uh, so just, just 25 actually. And yeah, so I've, I've been in, in, uh, process control automation for about 25 years now. I started off at the device level, so I was more in that instrumentation control valve area and then slowly journey my way up the technology stack, if you will, and was doing a lot more around the SCADA MES control systems, those types of things. And then of course you can't be in this, uh, an automation in, in 2020 without having some, um, without being involved in industry 4.0 and IOT. And so a couple of years ago, um, after having worked for several, I would call platform companies that did that focused on those same things, um, decided that, wow, there's a huge need because there's so many, so much coming at us all at once. Companies really need help understanding what's out there. What does that journey look like? Um, and because of my background within, uh, reliability, that I tend to focus on those types of applications as well. So if that's really the whole value of what it is that I'm trying to create.

John Rinaldi:

I think if you hit it on the head that there there's, there's a whole lot coming at us in terms of technologies, cultural changes, we've got a, you know, cultural change where the it and OT people are being integrated at some organizations. We've got a lot of the baby boomers. Who've who've who really have all that expertise over the last 30, 40 years that are retiring. We've got new people coming in who expect that, uh, just lift their cell phone and magically, they can make things happen. They can get data. And so it's a, it's a whole different world right now. Where are we in, in, in, in, in IOT and manufacturing is, is what's the state of it, or, and you know, I, to tell you the truth, I don't know because I have people telling me that everything is changing there's, you know, with all the way that the, the fact that PLCs are going away. And then I've talked to other people who were on the ground selling sensors, or, you know, doing projects and in, on the factory floor, they say, well, nothing really has changed. We're selling the same products and doing the same stuff. How do you view that?

David Schultz:

So I, I think it's both, um, absolutely thing is changing. I think how we're using technology is going to be different. I think how we go about delivering that technology is going to be very different. So, you know, in the past you would have this large system we're going to put in a SCADA system, we're going to do some kind of project, you know, exactly what that looks like. And at the end of the day, we're going to have, you know, this, this amount of IO, this number of screens, and here's how the operators are going to involve or to, to work with that system, that those types of things are still going to exist. But now what we're trying to do is unlock all of this data that's on the plant floor. And that, that PLC data is still going to be one piece of data that, that is going to get utilized. So I guess going back to when you originally were starting off the question of where are things well, we're just really at the very front end of things. And I think people have this idea that somehow just as quickly as the iPhone came out and apps got developed, that we're going to be able to do the exact same thing with manufacturing. And it's not quite that simple because there's a lot of things that have to be done from the underlying data in order to get it ready to just have your plant floor in your pocket, you know, on that cell phone.

John Rinaldi:

So is, is this stuff going on mostly with the bigger companies, the Proctor and Gambles and the DuPonts, or is it actually happened at the, at the smaller manufacturing companies? I mean, whose is the, is there a, is there a digital divide in, uh, in IOT?

David Schultz:

Well, I Think it's both, but I think they're going about it very differently in how they're doing it and what I'm seeing happening at some of the larger, uh, fortune 1000 companies is that they're continuing to utilize some of the automation partners that they've already always had or already have in wanting to do these digital transformations. But I think where they're getting stuck is that they do this first, um, they do the first part of it, or they, the, you know, the use case proof of concept, the pilot, whatever it is. And then they get bogged down in those pilot purgatory, because they're still trying to use industry 3.0 thinking in order to do that, there are smaller manufacturers are absolutely doing it, but it does come down to picking the right technology, picking the right strategy, picking the right partners. And so they've been successful because they haven't been involved with some of the larger automation companies. And they're able to take advantage of some of the I'll going to call them, the smaller players, if you will, that exist within this space. And so they're actually realizing those benefits and it's not, it's not the$1 million behemoth project. It just consumes a lot of resources.

John Rinaldi:

I, you know, I've, I've heard, uh, numbers from people like Cisco saying that only 24 to 20, some 20, some percent of IOT projects are successful. I've had other people from the, you know, the big Goliath Goliath companies that do all of this research projects. And they say that there's, there's, there's not that many projects that are there that are that roll out to scale that it's difficult to scale these things that tell me about that is this, you know, you, you mentioned industry 3.0 technologies. So what is the difference between, so I guess I'm asking you two questions, uh, the difference between the industry 3.0 and 3.4 technologies. And then what are some of the limitations that are, that are preventing these projects from being successful?

David Schultz:

So when you're looking at industry 4.0 technology, there are four different aspects that I take a look at. It's, it's an open architecture, meaning that all of my stuff plays well with each, with everything. It's a lightweight communication, it's a report by exception and it's all edge driven. And those are very different than be the ways that industry 3.0 systems were architected. And so some of the reason why this industry 3.0 thinking is not working, is it because they still like the manufacturers want to continue to architect where I'm not making this point. The point where it's my PLC talks to my SCADA, that then talks to my MES. It's sort of like the, you know, my leg bones connected to my five bone connected to my hip bone kind of thing. And that's not the way industry 4.0 technologies work. Everything should be talking amongst themselves specifically at that, all of this, these layers of the Stackery node within the overall ecosystem. And so that's why, um, there, there are these things aren't being successful. So in Cisco, um, publishes that, yeah, there's a reason for that is because all of a sudden they realize there's absolutely no way I can do this exact same one-to-one connectivity across thousands of devices. It's not scalable. And that's why you're having that limited success.

John Rinaldi:

Well, and that's where some of the things, some of the newer technologies that MQTT is designed for thousands of nodes and for four, anyway, lots of nodes in a, in a manufacturing environment, we generally don't have thousands of nodes and some of the, so that's why you, so you need some advanced technologies. Do, do, do most of companies have the expertise they need to, to, uh, implement industry 4.0 technologies and to do these kinds of projects or, or they need help. And, you know, from people like you and what kind of, what kind of help can you provide?

David Schultz:

So do they have the people to do it? They absolutely do. It's just getting again, it's that right strategy and the right partners to help them do that. Certainly it's something that can be rolled in house, but it does utilize some new technologies. So you've talked about MQTT. That's really not new within, uh, from technology standpoint, it's recognized Oasis standard. Um, but it is new within manufacturing. So it's taking a 20 year old technology that was built for getting data out of oil fields and in a very lightweight way, and then applying that in a manufacturing environment. So I really think some of the first, the first piece is understanding, okay, what does MQTT mean for manufacturing? Uh, me personally, I like using the spark plug specification along with that MQTT and then using that to move all of my data around. So using that, uh, that spark plug B, I will create something called a unified namespace to where all of my nodes. So again, your PLC node, your HMI node, your SCADA node, a story and node ERP note, NES note, et cetera, et cetera, all of those are not connecting in your publishing, the data that you want, that that should be made available in the unified namespace. And I'm also consuming that data. And so now you have this one to many scenario instead of a one-to-one and that's the, it's the application of the technology that makes that happen.

John Rinaldi:

That's what you're talking about. There is, there's nothing that's revolutionary about that, but it's a different kind of process than we've we've had with industry 3.0 and the former way of doing things. Is that right?

David Schultz:

Yes. Yeah. I mean, people are familiar with MQTT. Anytime you're using either Facebook messenger or Apple messenger, the little three dots that are familiar to everybody, that means somebody's typing, that's actually MQTT that's powering that.

John Rinaldi:

And it's an open technology, you know, when you, you, you described that was the first of the four attributes of an industry 4.0 technology, I think was the fact that it was open now, what do you mean, do you mean open source or what do you mean by open?

David Schultz:

It's an open architecture. So the, uh, the MQTT standard that is available to you, you can download that. You can develop your own devices that utilize the MQTT protocol and as long as well as that spark plug, uh, specifications as well, that is all available to you. So if you are a manufacturer, if you are a, uh, you know, of whether it's an OEM of the hardware, the equipment that is available to anybody that wants to utilize that standard for the communication, that's what I mean by...

John Rinaldi:

You know, what's interesting about that to me is that we've, we've had this debate for a long time about open standards, closed standards and semi-open standards. So for example, OPC OPC,UA, EtherNet/IP are they're they're they're open standards. You can go get the EtherNet/IP specification and implement it, but the places that you also are required to certify your device now with OPC UA, there's no certification required. You can just get the standard, you can implement a device and you could implement a crappy OPC UA server. So the problem is, is in, is, is that I think a manufacturer wants to know when they're going to work with a partner and, and buy a drive or, or buy a temperature controller or a temperature sensor that supports one of these technologies. They want to know that this thing has been developed and actually in properly, and that it works properly. And that, uh, and that there's, that there's going to be a, somebody, some backing it up and it has been tested thoroughly. And we don't have that in some of these. And when we, when we just go and get an open off the shelf protocol, so some people are saying, well, we really want, we really want those semi-closed semi-pro proprietary kind of standards where it's being controlled. Do you buy into that?

David Schultz:

Um, I don't, um, the reason why I like spark plug B is that there are several major automation companies that now support that, um, that specifications. So I don't think it's something that is, well, I'm not really sure. Maybe five years ago, that was probably more the case, but in that amount of time, those standards have become a lot more available. And again, MQTT, spark plug B that was really designed for automation. Um, that's in there.

John Rinaldi:

So I, I guess I still don't understand, is, would you consider EtherNet/IP an open standard?

David Schultz:

Um, well, because now it's available to anybody. Yeah. A lot of devices will use EtherNet/IP to make that, that it's, it becomes that, uh, that connection layer our app, uh, within your, your overall technology. And so now you're putting this application on, um, on top of that. So yes, but does it play well with others? That's really the question you're trying to answer.

John Rinaldi:

You want to, we want to get, I think, you know, where we want to get to long-term and I think this is good. This is going to be long after you and I are gone, unfortunately, is where, where machines can talk to one another and ask and say, here's what I want. And this is the kind of information I need to be configured. This is what I want to do. And they'll talk to the machines, we'll get together and make those, those, those kinds of connections. And they will share information and they'll just work together seamlessly without a lot of configuration. But I think we're a long way from that. And until we get there, we're going to have to, we're going to have to pick some standard communications technologies. Don't you think, in order to make a lot of this work well together? And I don't think we're there are we?

David Schultz:

So there's, there's a certain level of that, that when you plug something into the, the network that it's going to be, self-aware, it's going to publish that data and that anything else that is, um, part of that ecosystem is going to know exactly what just came online, what type of device it is. And it's also going to know what kind of information is available. So for instance, if I were to plug in a PLC that uses the, that publishes all of its in a uh, you know, using MQTT, spark plug B, it's not going to publish all of its tags are now going to be available, and it's going to exist inside a structure. That's going to say I'm a PLC, so then when it shows up anything else, it's now looking at that, it's like, oh, look, I have some more PLC data that's available. I like PLC data. I'm going to do something with it. And so now it's going to be a consumer of that. So it's a very simple use case, but there's, those types of things are already coming down the line now.

John Rinaldi:

Would you switch? You have a, I don't know, I think you, for in sake of brevity with you probably skipped over is that the normalization there, and there's two issues there, there's the normalization of the units, and there's also naming standards and location standards and all that in order for that to really be effective is you have to have common naming standards. So, you know, if the, that, that the motor drive has a specific tag, and it's going to be called something, that's going to be called by a name that implies some location, it's going to have some, um, some data metadata associated with it that tells you that it's coming in revolutions per minute and all of that stuff. And I don't see much of that other than OPC UA. There's almost nothing that carries that kind of context, contextual data along with it.

David Schultz:

So, one thing I guess I didn't, uh, speak to, so it's a great question. Is that, I'm sorry. I just lost my train of thought. Apologies. Um, so when you're building that, I referred to it earlier, the unified namespace, when you're building that out, you know, it's, that's not just this data connection where now I'm just telling it, okay, here's some, here's some tag data, you know, go do something with it. That's, it's providing that, that data it's providing that, that single source of events, but it's also providing you the structure of your overall data or overall business utilizing a plant model. So I tend to use an ISA 95 or the Purdue model for doing that. So I'm going to know this PLC is sitting out on in cell three of line two, and it's the PLC on this particular type of machine. And it does bring through a lot more of that metadata, um, if you've set it up properly to publish it. So I'm going to have units I'm going to, that are going to be normalized. I know exactly what that data looks like. So if I say this is 32 GPM, that pump is 32 GPM. That's what that text says.

John Rinaldi:

Uh let's uh, let's, let's turn to another subject. That's, that's, I've been thinking about quite a bit lately, lately, the, the influence that Amazon has had not only on the culture and everything else in the, in the outside manufacturing, but let's, let's, let's not talk about that. Let's talk about Amazon's influence in manufacturing. Now they're coming in with creating products for the factory floor. They're trying to get everybody connected to their cloud. Do, do, do, do small manufacturers need a connection with Microsoft Azure, an Amazon, or a Google in order to process their data and turn it into real actionable information, or can they do it themselves? And it uses other kinds of tools and stay local? Do you, where do you, where do you fall on the state local versus remote kind of timeline there?

David Schultz:

So I think what people have assumed as this industry 4.0 is somehow I'm going to have a cloud service, and I'm going to start plugging all this data into it. And then once I've done that, I'm now going to have all this really just great, insightful information. That's just going to magically show up in my dashboard. And that's, that's just not the reality. And I think in some ways it's gone back to the earlier part of this conversation is why some of these things are failing is that people just aren't really appreciating how much it takes to get that norm, that data normalized and contextualized.

John Rinaldi:

Well, I bet I've heard, I've heard that called IOT porn.

David Schultz:

Yeah, exactly. Yes. There's, there's a lot of eye candy out there for doing it and, you know, a dashboard you just does, it's not magic. Well, that's, that's some of this that, that one-to-one connectivity. So from my standpoint, there is a tremendous amount of untapped potential that already exists on the plant floor, whether you're a small manufacturer or a big manufacturer, and you don't need to push all that data to cloud, you can absolutely just start off with some very small start off with one piece of equipment, get it architected properly, getting that data all there to where every piece of information, anything that's there that, that it is utilized is now available to anything else. So get that structure built in so you can start doing it. And a lot of that is just local resources.

John Rinaldi:

They think they're going to save, save, save some data to a database, or is it, or is this some things that you do with post-processing or are you going to, or is this kinds of things that you would do in real time? What kind of, what are you talking about David?

David Schultz:

Both. So it's getting all that information available in real time. That's that's the beauty of the MQTT is because it now has the, all of that data. That's in real time, I just published my last value. So if I'm subscribing to it, I get that last value of data. There's also going to be a database, like say a data historian. Well, that's going to be a subscriber to that database or to that, um, that broker. And that's what the unified, a unified namespace allows for you is that now it all comes back and I can historize that data. And then start using that for say some basic trending. That's all going to take place locally as well. And then, so now you started off with just that one piece of equipment, and you now expand that to that, that line. If you will, to where now you have all that line equipment, um, ditch there, but yet you can, historize it, you have it available real time. You can make dashboards from it and you can create alerts and warming from it. It's there, there's a lot of things that can be done with it.

John Rinaldi:

So do you need, do you need more than Microsoft Excel? Can you just have the database and move data into Microsoft Excel in some fashion, and then Excel has a lot of basic functionality that I think you could use to, to analyze your data? Isn't that true?

David Schultz:

Well, yeah. So Microsoft Excel is the world's number one reporting tool. And it's not because it's better at doing data analysis. It's just, that's because that's what people know. And interestingly enough, when we start talking about process historians, almost every one of them has an Excel add-in why, so people can take that historical data, bring it into Excel and they can start manipulating it. But there's a lot of those historian client tools that are available. You can use those, uh, much like, um, Excel, but you can get a lot more rich data out of that, that you just can't do. And it's just because it's out of the box design fit form and function.

John Rinaldi:

So do we, we don't really need any fancy analytics package, I guess, is what you're telling me.

David Schultz:

Not at all. So Anna, there is a whole process for how you want to do it, or how you want to do your digital transformation. A lot of that machine learning, AI analytics, you know, wherever you want to throw it, that occurs later on in the process, you've got to start from the basics, what's the information we want, how are we going to utilize that? And then what's that going to mean for our overall business and making those decisions. And then

John Rinaldi:

I think you've just laid out a good scenario there. So what do we, you know, I guess the first question for anyone to ask is what do we want to accomplish? What where do we think we can, we can learn something that's going to add value where we can actually improve our process. Is that right?

David Schultz:

No, it is. I mean, it's always, I want to have a definition of what's the problem we're trying to solve. Let's just not do a 4 O effort just for the sake of doing it. It's what, and where do we think is there sometimes that's a very hard question to answer, because we just don't know, which is why I say start with something small, just to unlock a few pieces of data and see what that tells you.

John Rinaldi:

I, uh, I was visiting a, uh, a company that shall remain nameless here, but I went there and they said, John, we just been, we've just got our product or an IOT enabled. And it's, we're putting data up into the cloud and he says to me, so what do you think we should do with that data? How do you think we can make any money off of it? I said, well, that's the first, that's the question you should've asked to start with.

David Schultz:

Yeah, they've got their cart before the horse a little.

John Rinaldi:

Yeah. He, uh, he went on to say that, well, you know, there was one of the, those big consulting firms was talking to the board of directors and it came down through the, through the, from the board to the president, to the, to the vice president of engineering and down to me. And they said, we got put data in the cloud because there's$4 trillion of data of value and getting data in the cloud. So here we are. But, you know, I think one of the problems today, David, is that we've got, we've got all these tools that make it so simple Amazon and of course, Microsoft make things so simple because they'll just grab, there's so many ways you can send your data to the cloud that it's easy. And, and it doesn't cost a whole lot on the front end at the back end is another story, but it doesn't cost so much on the front end. So we just start sending data to the cloud without really sitting down to think about why we're doing that. If it was, if it was more expensive, it's like sending emails. If sending an email costs a dollar, we would all send a lot fewer emails and sending, sending data to the cloud cost real money right away, people would be, would think through at first about whether they should or not. So, um, you know, that that's good. Where do you think we're going to be in and, you know, give, tell me your crystal ball, where are we going to be in five years? Are things going to be radically different five years so now than they are today?

David Schultz:

Uh, absolutely. They are. And I think you're going to see some of the, the companies that were the industry 3.0 leaders. I think we're going to start having some industry 4.0 leaders. And there, there may even be companies that we haven't heard of that are going to be, wow, that's the latest and greatest thing. You've got to go have this. This is the, this is the key to your success. Um, so from now, I think you're going to see a lot more real time data. That's going to be available. I think you're going to have a lot more connectivity and devices that are, that are plugged in. I think you're going to have, uh, and even just the skill sets of the people that are down on your plant floor, it's going to be dramatically improved. So you were talking earlier about some of the people that are retiring out of the market. I think that new group of people coming in, I, I think you're going to see just a very different usage of that data on how we're going about doing that manufacturing. And you're going to see a lot of really great ideas coming from that new group as well.

John Rinaldi:

So, so that's, that's interesting. So what about the guy who's been in manufacturing for the last 25 to 30 years? He's been a control engineer and there's this new era coming out. What, is there anything that he should do? And mostly I'll say he, because it's mostly men or, you know, 99% of them, what should he do in order to prepare himself and be able to use industry 4.0 technologies? What does he have to learn? Where does he go? What kind of resources are available to a person like that who wants to update their skill set?

David Schultz:

Yeah. So for me personally, it's just learning what some of these say programming language looks like. So it's, it's understanding, okay. Where's Python, what's Python got used for, how do I take advantage of say Python, um, within it? So again, that's a programming language, uh, that, that people can take advantage of, or it's understanding, okay, I've heard this MQTT. What is that? No understanding what it, what is a MQTT broker? What is that payload structure? Uh, it's paying attention to those types of things. Um, you know, it's, and I guess it, what I was thinking when you started asking the question and I'm going to deviate and come back, is that when I was growing up, I learned how to type, and that was sort of that skill set that, that created a fundamental difference. And then now my, you know, my kids generation, all of them, you know, all, they, all, they all know how to type, but what they all learn is you got to know how to program. And so it's taking a look at what are the kids that are coming out of the collar, you know, out of college right now, what are some of the skill sets that they have in order to be successful in what's going to become industry 4.0, and then paying attention to that. And I think that's going to be fundamentally. One of them is just, again, that, that ability to program. And then you're also just going to have to be familiar with, um, just how you go about that, the way you think about how we do industry 4.0. And again, it's not just the one-to-one connection. Um, so I would say those are some of the skill sets you'd have to have.

John Rinaldi:

Well, that's, that's all, that's all wonderful here. We're kind of getting to the, uh, the end of our, of our time. So David, if somebody wants to get a hold of you and talk more about maybe an IOT project or learn more about the kinds of services that you have, we're gonna, where can they find you?

David Schultz:

Sure. So I'm on LinkedIn. Um, it's just David Schultz. I'm in the Milwaukee area. Uh, the company is G5 Consulting, so you can go to the website, uh, G5CES. Um, that's so, uh, G the number five and then consulting engineering services, uh, dot com. And of course, um, you know, phone number it's(262) 771-0123.

John Rinaldi:

Thank you very much. So this is John Rinaldi. We've been talking to David Schultz. Thank you very much, David, for the time it's been, been a pleasure talking to you, is it always is likewise. So, and I just like to remind folks that are listening, that Real Time Automation, we have a, uh, digital tech digital software toolkit for helping people with moving data and doing the kinds of things that David and I were talking about today. And you're welcome to go to our website, rtautomation.com, or actually you can even, uh, email me at jrinaldi@rtautomation.com. So thank you very much. You have a great day. We'll talk to you next time.[ End Music].