Industry 4.0 and manufacturing ecosystems has been added to your bookmarks.
The fourth industrial revolution, or Industry 4.0, is upon us. Brenna Sniderman spoke with Tanya Ott on how systems—connected via the Internet—can now analyze data, and learn from and adjust to their environment to inform future activity.
You know, it is happening very, very quickly, and I think that’s part of the reason that many people are afraid of it.
TANYA OTT: We’re talking the fourth industrial revolution today on the Pressroom—what you need to know and what you don’t need to do.
TANYA OTT: I’m Tanya Ott, and this is Deloitte University Press’s podcast on the issues and ideas that matter to your business today. I don’t have to tell you this: Things move quickly in today’s world. I’m a Gen Xer—I didn’t have a computer until college. A few years later, I got my first beeper—remember those? By the way, I hated mine: The newsroom I was working in assigned it to me, and people kept paging me in the middle of the night about emergencies, like an employee got his finger stuck in the fast food restaurant drain. True story! I’d turn it off, and when asked if I missed a page, I’d shake my head and say, “Oh darn—must have run out of juice.”
In 1997, the first computer chess-playing system beat reigning world chess champion Garry Kasparov. Ten years later, the first iPhone came out. Fitness trackers followed. And today, a type of artificial intelligence called machine learning enables computers to teach themselves to grow and adapt as they’re exposed to new data.
Brenna Sniderman has an eye on it all. She’s a senior manager with Deloitte’s Center for Integrated Research.
BRENNA SNIDERMAN: I focus on advanced technologies and what they mean for business strategy and businesses.
TANYA OTT: She studies things like artificial intelligence, robotics, the Internet of Things, and 3D printing. She’s written an article titled Industry 4.0 and manufacturing ecosystems: Exploring the world of connected enterprises. I asked her—4.0, huh? What were 1.0 and 2.0 and 3.0?
BRENNA SNIDERMAN: This is a question I get all the time. Why 4.0? Seems that we’re never happy unless we’re putting “.0”s on things.
TANYA OTT: Brenna explains that 1.0 was the first industrial revolution of the late 18th and early 19th centuries.
BRENNA SNIDERMAN: Began with steam and the power loom. So all of a sudden items that were once made by a single artisan in a single shop could be produced a little bit more on a mass scale. And here we had the first instance of capital or machinery replacing human labor.
TANYA OTT: The second industrial revolution took place in the early 20th century.
BRENNA SNIDERMAN: And this is the one a lot of people are very familiar with. This is the assembly line, Henry Ford—what you think of when you think of, you know, the traditional factory—and here we saw that capital increasingly replaced labor, at least on the shop floor, in terms of machinery. But we also started to find all types of other new jobs were created to sort of manage that process.
TANYA OTT: The third industrial revolution took place from the 1970s into the early 2000s. This was the introduction of information technology—IT, computers, spreadsheets . . .
BRENNA SNIDERMAN: The dreaded spreadsheet.
TANYA OTT: There was the Internet and simple robots that could do the same repetitive task over and over again.
BRENNA SNIDERMAN: And this was incredibly revolutionary in an industrial sense, because computers truly and the Internet truly changed the way that business has worked.
TANYA OTT: And now—it’s Industry 4.0. We now have the ability to collect lots of data from connected systems—say, the Internet of Things sensors in your car or in a manufacturing plant—and analyze that data and then use to it to inform future activity.
BRENNA SNIDERMAN: I guess the long and the short of that is: Rather than having a robot that just does the same repetitive task over and over again, we now have systems that can be flexible, learn, and adjust to their environment, adjust to previous experiences to do things that are a little bit more conducive to the situation that they’re in.
KELLY MARCHESE: For me, Industry 4.0 is all about capitalizing on the great data and technologies that exist out there.
TANYA OTT: That’s Kelly Marchese. She leads Deloitte’s Supply Chain practice.
TANYA OTT: So some other phrases that folks might be familiar with would be, like, connected enterprise, smart manufacturing, smart factory, and Internet of Everything. Does all that mean kind of the same thing, but you’re just calling it 4.0?
BRENNA SNIDERMAN: Yeah. That they all mean the same thing. I mean, you hit the nail on the head.
TANYA OTT: Got it. So essentially what you’re talking about is this is a loop that goes physical to digital—like sensors and things like that—and then all that data then back to physical.
BRENNA SNIDERMAN: Yes. We have what we call the physical to digital to physical loop, where essentially a connected physical object will have a sensor or some other connectivity to it. It does something. And that sensor or the connectivity that’s part of it will create some sort of digital information that has been transmitted back to some sort of system where it’s analyzed. We figure out what it means. Or the computers, more accurately, figure out what it means. And then translate that learning or that data back into some sort of activity or physical movement that the physical object then does. Essentially what it means is a physical object does something. Information about it is created digitally, that then impacts the physical world in some way. So when we’re talking about physical objects, we mean robots, we mean drones, we mean 3D printers that create an object. When we talk about the digital side, which I think that’s big data analytics—sensor technology and things like that.
KELLY MARCHESE: You know, we’ve gotten very good at taking physical information and turning it into digital information. And many companies are getting really good an analyzing that data. But what’s really exciting is how you now not just look at that analyzed data and say what happened but to say what can happen, and actually taking action on those analytics to change the behavior of the future. And that’s really the game-changer. It leads you into more predictive things, taking some of the human element out of making decisions and being able to handle so much more complexity than any human could.
TANYA OTT: I find it’s really useful for people when we take things out of the broad theoretical and into the practical and the applicable. So what you just explained there—could you give me an example of how that would play in an actual setting?
KELLY MARCHESE: Sure. So one of the areas that I’m fascinated with is additive manufacturing. I think it’s one that a lot of our Department of Defense clients are very focused on because of the ability to manufacture parts where suppliers don’t actually exist.
TANYA OTT: I want to pop in for just a second. For those who don’t know of this technology by the name additive manufacturing, we might more commonly know it as 3D printing.
KELLY MARCHESE: Exactly. So with additive manufacturing or 3D printing, the way that [this] actually is an Industry 4.0 revelation is that you take a part that maybe doesn’t exist in your supply system, the supplier doesn’t exist, and you can scan it. So you’re taking physical information and creating it digitally through a 3D scan. You can turn it into the digital world, and maybe it’s a broken part, so you can work on what’s now a 3D scan and adapt the design and then translate that into software where you can simulate a 3D build, send that information over to a 3D printer, and then print it. So now you’ve taken the digital information and turned it into a physical part.
TANYA OTT: So if you’re going to take this into the manufacturing realm, and you were talking about a 4.0 technology that makes that physical to digital to physical link—can you give us an example?
BRENNA SNIDERMAN: Well, I’m going to take one particularly good example would be General Motors, which has plants for control network. What they’ll do is measure humidity levels in the plants, and what they discovered was they’re painting the chassis of cars, and when humidity levels reached a certain point they couldn’t paint them anymore—they have to repaint them—and it really damaged their productivity to have to go back and repaint a lot of these things. So we’ve got these sensors that are measuring the physical environment, humidity on the plant floor. It comes back. It’s set into some sort of reader that reads all of that information, analyzes it, and analyzes what else is going on in the plant at the same time. And when the humidity and the other environmental factors hit a certain level, everything is automatically rerouted on the floor to another spot so that we don’t have to deal with repainting or sitting, you know, with machines idle for some time.
The benefit of that is then we get into things that are what we call predictive maintenance—or, you know, some sort of predictive activity where when you gather enough data about a specific object or a specific behavior or a specific plant or facility and all of the aggregated information around it, you can start to make predictions on what might happen next, based on all the patterns that you’ve seen in the past and start to proactively or preemptively address them. And we have seen some examples of that. There was an electric company that was collecting maintenance data and historical data over the course of one year for a turbine in its factory that, for whatever reason, kept breaking and kept requiring maintenance, and the breakdowns kept escalating and getting worse and worse. And what they found after analyzing that data was that technicians have been addressing the symptom of the problem rather than the root cause. And by analyzing all of the information—not only about the way at times in which that turbine was maintained but also historical data about what was going on in the plant at the time, what was going on with that piece of machinery at the time and what it was doing—they were able to actually understand what the root cause of the problem was . . . which was a thermal expansion problem, and that led to the symptom which was a vibration that was making the bearings come out that caused equipment shutdown. So what this did was, they estimate that it saved them several million dollars along with many fewer days of equipment downtime, because rather than repeatedly addressing the problem of the vibrating bearing, they were able to look at all of that data and understand exactly what was going on within the piece of machinery. So it not only enabled them to fix that piece of machinery but to predict when it might happen again and apply that information to other pieces of machinery in other plants.
KELLY MARCHESE: And that can allow you to do a lot of different things. Not just plan when your maintenance is but determine what parts you need to order in advance. It manages your inventory levels. If it requires very sophisticated maintenance personnel, you can make sure they’re at the place at the right time. You’re minimizing risk because if that piece of equipment breaks down it ends up creating a line shutdown. A lot of times when the equipment goes down, it could cost hundreds of thousands of dollars an hour. And so to be more predictive drives huge savings.
BRENNA SNIDERMAN: You could extrapolate that predictive capability, which is an extremely powerful one, to all kinds of areas. If you look at shipping and logistics for trucking companies or any type of machinery that you know that you buy a great deal of workers. Thinking about shipping and logistics or even sales of automobiles—if you can gather enough data from connected cars to understand what’s going on with them and when things might be going wrong with them, you can help predictively maintain those cars. Let’s just say you’re a shipping company that has to carry a lot of packages around. You can start to predict when trucks are going to need to be taken off the road to be repaired and repair them beforehand before things get too bad, which helps to make sure that your fleet stays maintained. But not only that—we have examples of shipping companies that are also aggregating data from sensors on their vehicles that not only look at weather but are also pulling in data from what’s going on with traffic patterns, what’s going on with the types of deliveries that they have, and also pulling in how often they have to go back to one specific location to pick up packages. And all of that information is helping them to dynamically route that route.
Another notion, which is I think a little bit more esoteric and perhaps difficult to understand, is the notion of the “digital twin,” which is also a component of Industry 4.0. It’s the idea of having sort of a digital simulation of a physical object occurring or being updated in real time. Now, there are some organizations that are using digital twinning alongside of their plants—you know, wind plants or electrical plants—to take data from what’s going on in that plant in real time into a digital plan that’s running alongside it and use that twin to start modeling out different scenarios and variations and what might happen, so that they can start to adequately prepare their plant for things like: What happens if this machine goes down? What happens if there’s a flood? What happens if there’s a hurricane? What are the things that might happen to this plant, and how can we proactively prepare for that? A lot of that sounds a little bit science fiction, that we can predict what’s happening in the future. But you start to look at data and start to recognize a lot of patterns that are happening.
TANYA OTT: What are the biggest challenges facing companies that are looking at implementing this?
KELLY MARCHESE: Well, I mean, it’s really hard. This goes against a lot of conventional wisdom, right? So in the past, equipment was just supposed to perform, and now we’re actually asking it to help us drive decisions and change behavior. A big part of the challenge is talent to determine how you might use it.
BRENNA SNIDERMAN: Finding individuals that can plan, execute, and maintain these new systems—engineers that are trained in handling the unstructured data and big data tools that can deal with the information that are coming out of connected machinery and individuals on the shop floor that can maintain and work with this machinery. This is a real challenge. I think the other challenge in this particular regards the unwillingness to work with these new technologies, because quite frankly they’re unfamiliar. So that sort of goes in hand with finding and training people, getting them skills to work with these things. And also, you know, finding the capital to invest in a lot of this.
TANYA OTT: When you’re talking about the lack of workforce, you know, people just aren’t educated to deal with this yet: Is industry responding to that? Is the education system responding to that? What sort of change needs to happen in order for there to be the kinds of people, the kinds of engineers, who can take all of this data and do something meaningful with it?
BRENNA SNIDERMAN: So that’s a very complicated question. In both academia and organizations, I think, the answer is time. Because on the educational side, one of the challenges when you have a lot of these new technologies, the teachers have to understand it before they can teach it. Some of these technologies are new enough that they actually may not have the experience with them to be able to teach them. We’ve run into that particular challenge with additive manufacturing and 3D printing. Know that on an industrial level you have to find professors that can teach it and understand it.
On the enterprise or company side, you know, the truth of it is these things are changing so quickly that there has to be some level of training within the organization as well. Even if individuals are coming out of the education system with a strong background in, say, engineering or manufacturing, you’re still going to have to get them trained in what specifically, what specific technologies you’re using. So I think it’s a combination of the two. But you know, I think it’s one of those things that honestly one of the biggest sort of salves or salvations for it will just be time.
Another challenge that we found is a major one—it’s standards and interoperability, which is something we’re also seeing with the Internet of Things. When you’re dealing with Industry 4.0, there’s so many different components of it, so many different types of both physical and digital technologies, and making sure that they can all work together, speak to each other, communicate with each other. Making sure they can integrate with each other because there’s going to be so many different systems. When you’re talking about working across a massive supply chain, how can you make sure that all the connected systems across all the different vendors in logistics and shipping and all of these different components can speak to each other and work together?
KELLY MARCHESE: The standards aren’t there yet. Ecosystems are more important than ever because no single organization can make that call and without stalling things out. So for organizations to be working together in these ecosystems, to work through what are the standards that we’ll commonly leverage is really important.
TANYA OTT: Does that eventually, you think, shake out in the marketplace, or is that something where government might come in and intervene and say it’s got to be done this way or that way to provide a framework for better communication?
BRENNA SNIDERMAN: You know, I think it’s a combination of both. And this has precedent too. This has happened in the past when we look at VHS and Betamax. Those were two competing standards for watching videos, and we see it a little bit with Blu-Ray and DVDs. It’s not unheard of. And you know, I think it is something that will work its way out in time, because there’s incentives in the marketplace for it to do so, because as an organization if you’re trying to build these connected devices, you’re going to want to make sure they’re interoperable. So I don’t think that it’s in anyone's best interest for them not to interoperate with each other at least to some degree. I can’t speak to who or how that will shake out—just that it has precedent in the past.
TANYA OTT: Obviously, there are a lot of players involved in this, and you may have a company that’s working with lots of different other companies, suppliers, and such.
KELLY MARCHESE: Right.
TANYA OTT: So data ownership is going to be a big issue.
BRENNA SNIDERMAN: Who owns that data? Who controls that data? How can you keep that data private if you had, say, two vendors who might be competitors with each other? How can we make sure that that data is secured in such a way that we know who owns it and what they’re going to do with it? And following on that is just the notion of, you know, data security—not just in terms of privacy across the supply chain but cyber security. When you’re dealing with all types of connective machinery, connected objects, connected devices—you know, 3D printers that are printing objects—how can you secure all of that to make sure that it can’t be infiltrated either for somebody to steal your IP or to bring a factory?
KELLY MARCHESE: That’s why, as we think about Industry 4.0, it not only changes what happens in an individual company—it changes, probably, business models. The way ecosystems are leveraged and contractual vehicles—how organizations’ supplier and customer relationships work and what the expectation is around how data is being moved back and forth and who owns it and how they might even pay for it.
TANYA OTT: That’s a complicated thing. How does that get sussed out in the marketplace?
KELLY MARCHESE: It’s going to take some, probably, trial and error.
TANYA OTT: And maybe some actual trial?
KELLY MARCHESE: Exactly, right?
TANYA OTT: You know, there’s a word that you’ve used over and over in our conversation today, and that’s the word “quickly.” And I would imagine that an organization looking at all of this could either go: Look, I want to get there first. I want to figure out who is the VHS, not the Betamax. I want to be in that space, and they’re racing to that space. Or they could go: It’s changing so quickly it’s almost paralytic. How do I know where to start? When to start? How to deploy? Where to invest?
BRENNA SNIDERMAN: You know, it is happening very, very quickly and I think that’s part of the reason that many people are afraid of it is the fact that it’s happening that quickly. I mean, even if you look at the progression of these four industrial revolutions: There were, you know, 150–200 years before the first and the second; there were about 70 years before the second and the third; and it’s only about 20, 30 years between the third and the fourth. So you need to look at how quickly progress is speeding up. You know, it’s a challenge to know how to react to something when you can’t fully wrap your arms around what it is just yet.
TANYA OTT: So what advice do you have to people in the industry who are struggling with that challenge?
KELLY MARCHESE: Well, I think it’s important that business leaders sit down and really understand what are their business objectives, first and foremost, and then decide what kind of technology.
BRENNA SNIDERMAN: You don’t need to adopt everything all at once just because you can. I know I’m using the term “quickly” throughout, but I wouldn’t say start slowly. I'd say start judiciously, which is not necessary the same thing as slowly. Think about the things you have to deal with first and tackle those first, and then move on to the next.
TANYA OTT: Brenna Sniderman and Kelly Marchese agree: Companies need to realize they’re not going to need to use every single one of the new technologies. There’s no one-size-fits-all approach. It’s more like an a la carte menu where you select the things your organization needs and then go from there.
There’s a lot more advice on how to order up your technology deployment in the article Industry 4.0 and manufacturing ecosystems. It’s at dupress.com, along with a huge archive of podcasts, including a recent one of the top three tech trends disrupting business right now. Here’s a tease.
BILL BRIGGS: I get it. Potential. Great. A lot of things the business would love. How do we make that real? How do we shift that from an irresponsible whiteboarding moment of what would be possible to what’s the actual road map and things we need?
TANYA OTT: That’s all today, folks. Check out the archives at dupress.com and let us know you’re out there! Tweet us at DU_underscore_press or email us at email@example.com And hey, do me a favor: If you enjoy listening to this podcast, if you learn something when you tune, tell a colleague. Share it on Facebook or Twitter. We’ve got a lot of dedicated listeners, but we’d love to reach more. I’m Tanya Ott for the Pressroom. Thanks for listening, and have a great day!
This podcast is provided by Deloitte LLP and its subsidiaries and is intended to provide general information only. This podcast is not intended to constitute advice or services of any kind. For additional information about Deloitte LLP and its subsidiaries, go to Deloitte.com/about.