Listen: Next in Tech | Episode 117: Transformation in manufacturing

View Full Transcript

Eric Hanselman

Welcome to Next in Tech, an S&P Global Market Intelligence podcast where the world of emerging tech lives. I'm your host, Eric Hanselman, Chief Analyst for Technology, Media and Telecom at S&P Global Market Intelligence.

And today, we're going to be discussing digital transformation in manufacturing, a step beyond some of the things we've been doing around transformation more broadly. And joining me to discuss it is David Immerman, one of our consulting analysts. David, welcome to the podcast.

David Immerman

Hi, Eric. Thanks for having me, and I'm excited to talk about this very important topic.

Eric Hanselman

Well, very important on many different fronts. And before we get too far into the transformational aspects, it would be great to talk about really, what is manufacturing more broadly? What's the value chain? I mean manufacturing is such a broad term, but there are a lot of different pieces that really make it up, yes?

David Immerman

Absolutely. It's a massive industry and topic to unpack. And I typically like to start the conversation in terms of just the quick breakup between discrete process and continuous manufacturing. And I think for this conversation today, I'd like to focus more on discrete.

So when I take a look around at just our daily lives, how much we interact with products around us to get around, that we consume on a daily basis, just those are all created in a factory somewhere, right? And that's really the heart of discrete manufacturing.

So who specifically am I talking about? And this is aerospace and defense, automotive, electronics and high tech, whole array of industrial products out there, when you think about HVACs, equipment out in the field, whether that's oil pumps or in factories themselves as well as medical devices and many, many more. So that's generally who I like to think about in terms of discrete manufacturers, so folks that have these complex products if that layout makes sense to you.

Eric Hanselman

It does indeed. And I guess when we start thinking about what that whole world looks like, I mean the value chain in this environment is made up of a bunch of different participants. You've got the folks who, when you think about manufacturing, are actually doing the manufacturing themselves, the actual creating products through a manufacturing tool chain. But there are also other pieces on the front end of that. It's all of the design aspects. There are a lot of different parts that come into play in this whole sphere.

David Immerman

Absolutely. And I think you hit it head on in terms of generally, when folks think manufacturing, they think about what's going on in the factory.

I'd love to start the conversation in just thinking about the products point of origin and then as it traverses through factories and other points and ends up in customer environments and just a few areas in the value chain that I think are important to talk about for this conversation.

So starting as far upstream as possible, talking about the whole procurement and supply chain space. Folks who are really concerned with managing suppliers and any upstream purchasing activities, so these could be procurement folks that are looking to source parts, components, raw materials from an array of suppliers that they might be working with on a daily basis.

And these people are really concerned with managing costs, as you would imagine, as well as, and it shouldn't be a surprise, reducing their risk. I mean when you start to think about the latest supply chain disruptions, whether it's a pandemic or a chip shortage, a lot of the brunt of those disruptions fell to these procurement teams who worked quickly to unscramble and get their suppliers in check.

And I think just to try and size how big of a challenge this is, an automaker, for example, might have 1,000 Tier 1 suppliers who may have their own network of suppliers that could be in the tens of thousands. So when you think about a whole automotive supply chain for one automaker, it could be in the tens and tens of thousands and so on.

And when you think about disruption across this extended supply chain, there's a lot of instances where the manufacturer might not even know who else is in this supply chain past their Tier 1s. And then that's where a lot of the problems arise. So I think that's a very interesting space, and there's a whole digital transformation story as well.

Eric Hanselman

We've talked a lot about the extent to which the world at large really wasn't particularly concerned about supply chain until the pandemic, and then everything was wrapped around supply chain issues.

But to your point, the procurement piece is everything from raw materials, meaning if you're a car manufacturer, chunks of steel, all the way out to fully finished components, electronics if you're building the car, things of that nature, all these things. So it's a wide and varied environment when you think about what you're actually sourcing for some of these materials to actually produce the finished goods.

David Immerman

Absolutely. Just a ton of form factors and decisions that are arising in the day and we'll get to a little bit later in terms of sustainability, and that's a whole other equation to procurement. So it's definitely a very interesting space and a lot of challenges they have to face upfront.

But generally, the next after you have your supplies or components procured, it's, what does the creation of this product look like? And that's where I start to think about engineering or the product development process.

So the product designers or product engineers, mechanical, electrical or whatever it might be, that are creating these products, whether it be in sketch or paper form or more recently, I would say, product design platforms such as CAD, and they are tinkering with the features of these products that could be increasingly in digital form or physical form and looking for opportunities to differentiate their products and improve their customer experiences as well.

And like procurement, they have their own unique set of challenges where they are tasked to really do more with less in terms of create higher-performing products that have fewer parts and lower costs. And this is kind of conflicting in terms of a lot of products out there in these more industrial areas have very custom requirements. So this makes it more challenging because on the product development end, they become more complex and they require greater configurability.

So going back to the car example, an engine can have itself hundreds of parts, which can be configured in many different ways, whether it's different layouts of pistons, pumps or valves, where one customer may want it a certain way and another -- maybe one another one. So how do you handle this large amount of information in this process while keeping costs low, getting better performance targets and also improving your time to market or new product introduction or development rates, which is also a metric they go by.

So a lot of important product decisions are made here that expand throughout the life cycle, both upstream to procurement and the parts and components they have to get as well as downstream in terms of how the product is manufactured itself and then service when it's out in the field.

Eric Hanselman

That whole chase to better, faster, smaller, cheaper, whatever your particular differentiating piece happens to be, all that pressure to get better at whatever the aspects happen to be.

David Immerman

Exactly. It's, do all those better all at once and at lower cost as always. And then you get into the actual manufacturing of product. And that's where we start to think about factory itself, usually known as production or operations. And this is where we think about the assembling, fabricating or whatever might go into this manufacturing process.

And the folks there are -- really have always been concerned with driving efficiencies, and that's constantly squeezing the cost that I can get on my production facilities, and any techniques to go about that in terms of identifying and resolving any production bottlenecks.

But I think a common standard metric to go by is overall equipment effectiveness or OEE that's known throughout the industry quite well where simply said, it's an aggregated metric based on availability, performance and quality. And I think manufacturers are always constantly looking to improve this number in their facilities, but I found it interesting that our Voice of the Enterprise found that 75% of manufacturers have an OEE of less than 80%. And that's relevant because 85% is sort of the gold metric in terms of best-in-class. So there's still clearly efficiencies to be had in this.

And then I think about a little bit of past to these efficiencies, how do I get that remaining percent is both on the physical end with everything happening in Industry 4.0, automation, industrial robotics as well as the digital end and in innovations in IoT as well as the industrial software stack with PLC, SCADA, MES.

Eric Hanselman

And there's a lot that's going on. When we think about being able to improve efficiency, it's making sure that you've got the right materials, that you've got the manufacturing processes that work well, that the quality of the materials is good enough.

Yes, it's interesting, though, to see that the VotE data is identifying that most are saying that they're under that golden target of 85% really by a full 5 points, around 80-ish percent. It does identify that this is a complex environment, there are a lot -- literally, a lot of moving parts and that achieving high levels of efficiency is pretty hard.

David Immerman

It definitely is difficult, but it's something they're going to continue to strive through and get that every percent better they can as it can impact thousands and thousands of dollars. So it's...

Eric Hanselman

But getting the product built is not the end of the story...

David Immerman

Great segue. So -- and then there is this logistics aspect of getting the product to its final destination that I won't go deep into. But I think an interesting area is the field service space. So where I'm really talking about here are service technicians that are maintaining, repairing or just servicing these products that are in customer environments or being consumed by end users in some way.

And what's interesting about this field service space is that it historically has also been viewed as a bit of a cost center. But even more recently, it's been viewed as more of a revenue driver when you start to think about things like product as a service.

But on the cost side, you think about this challenge in terms of, I may have thousands of these industrial products across these very broad-ranging geographies. Maybe I only have 200 technicians.

How do I optimize them to not only ensure that the equipment doesn't fail for our customers, but the massive amount of costs that are on the line in terms of their fuel efficiencies, let's say, that come with a truck dispatch to send one of these technicians out to go service them -- it could be anywhere from -- I've seen $200 to $400 per dispatch, multiply that around 200 technicians a day, however many days a year. And that's a pretty large chunk of us. So again, something that manufacturers are looking at both to drive revenue as well as keep cost manageable.

Eric Hanselman

Well, I mean it's not even just the hands-on, hey, it broke, we got to fix it, kind of stuff, but even just the logistical management of, is it in warranty, contract management, all those kinds of things about, who pays for it?

And if you start sort of leaning towards the consumer side, all of the goods handling, returns, the channels that you're working with, the partners that you've got in place, again, quite literally, a lot of moving parts.

David Immerman

Yes, it's very sophisticated.

Eric Hanselman

So anytime we've got lots of complexity, we've got lots of different pieces to be able to bring together, it sounds right for digital transformation.

But what's different in manufacturing about the digital transformation aspect? And we've talked about what the typical enterprise IT transformation looks like. And that's gathering more data, being able to go integrate that into systems. How does it change in manufacturing?

David Immerman

Yes. It's definitely a category to unpack. And that's why I think teeing up the value chain is important for digital transformation and putting into context because sometimes digital transformation is across these different functions, right, across the product life cycle and these different functions.

And that's where the word digital thread starts to kind of appear, right, where it's connecting these different systems and making them talk better together for whatever your business outcome is. But I wouldn't always limit to that. I think you could also say that digital transformation is in a specific department, like procurement or it's a mix of them and they don't necessarily interrelate right now, but they could in the future.

But I think what's really different is that it's just where manufacturing has come from, especially from a technical perspective, span a lot of legacy systems across these different parts of the value chain, whether that be the technical systems themselves or spreadsheets, manual inputs on paper, a lot of custom systems out there that once you're prepared to try and integrate with and get valuable data from, it isn't available.

So this creates, I would say, pretty vast information silos. And in manufacturing where you can't afford any downtime literally because of the extensive cost, you're very risk-averse to adopting the latest technology. And you know you have these challenges for integration and whatnot. So that's caused some hesitancy in the past.

And I would say it's created a lot of situations where there's a lot of siloed projects from one site that has a different digital project and a different set of tools than the other one. Same goes for use case. And it's created really a patchwork, if you could say, across the manufacturing installed base.

Eric Hanselman

Well, that's something where when we think about what Eric Johnson who's on a previous episode from the Journal of Commerce was talking about, there is that tendency to hang on to the lowest common denominator. And he's talking about that low level mostly from a logistics perspective, that fax is still a great way to communicate because you can mostly count on people having fax capabilities.

And it sounds like in manufacturing, you've got that same kind of issue that there are all of these legacy mechanisms that have been working successfully that people don't necessarily want to touch with that concern that there's risk you might break something.

David Immerman

Yes. And I think it's a great point on that concern. And it's just, is it worth the risk to install this flashy new piece of AI, IoT or AR, whatever it might be, in a manner that isn't going to disrupt my operations, which at the end of the day is what I measure it on, right? I measure it on my cost and overall production. So once you start to toy with that with the manufacturer, you'll quickly lose them in a conversation.

Eric Hanselman

So what do you see some of the trends in manufacturing in terms of adoption? How rapidly are people moving forward?

David Immerman

I think we are more or less a tip away from the have-nots to the haves. There are more that are joining this lighthouse manufacturing category with the forward thinkers who have scaled use cases across many sites. And I think there's a couple of reasons of that, and what's changed really?

And I think for IoT specifically, maybe 5 or so years ago, we were in kind of a world of, let's say, pilot purgatories, right? These stalled projects that had no hope in sight of getting to an actual production environment. And there's a lot of reasons why this happened in terms of manufacturers were opting to build them themselves or known as DIY. They might have gotten stuck in terms of a stage of analysis paralysis, which kind of equates to, I have all these different options. Which one do I go with?

Point being that there wasn't a lot of structure around these digital transformation groups. But we've definitely gotten to a point where they've started to formalize, the right stakeholders are in the room. They're being led by CXOs or at least have their support.

So I think that's definitely important in that holistically, manufacturers are seeing it as more of a strategic differentiator in terms of, I can offer different products and services via digital. And then there are some that ultimately saw it as a necessity with things that have happened like COVID, where I need to bump up remote monitoring up my strategic initiative list because I frankly do not have the workforce to go service that machine, so to speak.

So those are a few. And I think just generally, technology has matured in the past however many years, whether it be IoT, AI, cloud or edge, in that these providers are also more geared towards solving actual manufacturing problems. So we see IoT providers really starting to talk about OEE and lowering their cost of manufacturing, also extend to things like procurement with demand forecasting via AI, lessening your inventory and then even AR, being able to alleviate the skills gap to increase your productivity.

So I think there's been some messaging as well as maturity in the tech end, which has helped, and then just the mindset of manufacturers has changed into a more positive mindset for digital transformation.

Eric Hanselman

Well, it seems like manufacturing is an area where there hasn't been that same sort of leap towards the new shiny thing that there has been on the IT side and that it really required that manufacturing technology got to the point at which they could actually speak to manufacturing concerns in order to actually really deliver some value, to actually communicate what that benefit was and really identify it in ways that the manufacturing community can understand and work with.

David Immerman

Yes, I would say that is a very accurate way to put it.

Eric Hanselman

Now you touched on something that I want to dig into before we get too far afield, which is sustainability concerns. And how is digital transformation playing a role on the sustainability side in manufacturing?

David Immerman

Yes. It's definitely the next big initiative on manufacturing executives' plates, let's say. And again, the conversation to the value chain because I think there are things you can do across it via digital transformation that can actually start to make an impact on it and report to stakeholders or customers that, hey, we are going more green than we were.

I mean that's a big one, right, in terms of being able to source more eco-friendly parts and not just the parts themselves but working with eco-friendly suppliers, people want to know, who is in your supply chain? Do you have a responsible and ethical list of suppliers that you are working with?

Engineering is definitely at, I would say, the crux of sustainability in that designing more efficient products can make a massive transformation of your carbon footprint. So looking at tools like generative design to create different products that have, I'll say, different structures that achieve the same performance.

There are techniques such as lightweighting, where I design this product with fewer parts but that has significant impacts downstream in terms of, hey, I design this engine with fewer parts. It's lighter. Think about the fuel economies the ending truck could have as part of that given that it's operating at a lighter weight.

Eric Hanselman

You said the generative AI word, so we got to dig into that there. But I think that's something that's worth digging into in terms of what the reality is for some of these pieces. And Johan Vermij has talked a little bit about it. He was doing -- moderating a panel on that at CERAWeek last month.

This is one of those things that I think is another interesting piece, and it's easy to sort of say, hey, generative AI is going to help make this better. But you touched on some of those aspects. And I think it's worth to sort of dig into, what are the different things that at least today we're considering might actually make a real difference?

And things that occur to me are things like being able to do -- you talked about lightweighting, being able to build parts that have the same structural integrity, functional performance, whatever their particular design criteria happened to be, and yet use less material and manufacturing process or are lighterweight or whatever the beneficial characteristics are. And that's something where we've actually got some sets of capabilities that we can leverage today, right?

David Immerman

Yes. The concept has been around for a while in terms of generative design, but the tools have only become more sophisticated. And that is both on the software side with, I would say, more CAD programs that have generative design support as well as what we start to see with additive manufacturing and 3D printing. So I think the menu of options has expanded, and we have started to see it have an impact.

I think a couple of areas that have been definitely impactful in the near term are a lot of manufacturers have opted to have their own 3D printers and do additive manufacturing in their own factories to replace a part if something were to happen unexpectedly.

So a machine breaks down, I might ask supplier to send a new part or a machine. But in the meantime, right now, I just need something that is good enough for a certain amount of days. Well, that would be great to actually spin up in software and construct the additive manufacturing instantaneously, and then I don't lose any downtime.

Eric Hanselman

And you might also have the additional benefit of if that additive printer is actually closer to your location, you save costs in terms of shipping, you save fuel, all those other good pieces around that.

David Immerman

Yes. It really extends the benefits once you start to look at the whole value chain. And going back to the sustainability part, I mean it is interesting because I think a lot are starting in what they can measure within that factory. And then I'm not an expert in this, but in terms of Scope 1 emissions as we start to think about -- and then expanding on that in terms of Scope 2 is the energy in terms of power or whatever other power source, electricity that these machines are generating in terms of Scope 2 emissions.

And then there's this whole other field or broader category known as Scope 3 emissions, which is everything across this value chain we've been talking as well as even including things such as customer usage of your product. And I've even seen S&P put out some numbers in terms of -- I think it's over 80% of carbon emissions for manufacturers are in the Scope 3 broader bucket.

So it's, how do I even start to measure that as well as monitor it, let alone report on it? But I think it starts with starting to get those near-term quick wins and have some measurable success, and then it builds out broader from there.

Eric Hanselman

And measurement is really the key, as you're pointing out. This is something where in order to know what those Scope 3 emissions are, you got to be managing them. You have to understand what they are. You've got to be able to bubble them up that supply chain so that you actually can have visibility into what they actually are.

And I guess there are pieces of some of this we've talked a little bit about in discussions around some of the industrial IoT kinds of things, industrial Internet of Things, in that we're now sort of getting into this world where it is becoming important enough that it is starting to drive some consolidation in the markets, it is starting to create some change in terms of overall information management. What do you see there?

David Immerman

It's definitely an interesting space, having looked at it for over 5 years now, probably warrants a podcast section in itself. But as you mentioned, consolidation definitely is something that has been very evident. I think maybe 6 or 7 years ago, we looked at 300 or so IoT platforms. Everyone said they had one, and now I would say there's probably a few dozen that are currently credible in the market.

And I think there's a lot of reasons why this was in terms of manufacturing had made investments in software in place such as MES, SCADA and PLC that technically, IoT was playing a role in that. So it's not to say you needed this industrial IoT platform to come in and change everything. It's that you needed to start thinking about it, wrapping -- or around it and adding additional value on top of it versus replacing it.

And the way I kind of think about the IoT or industrial IoT space goes back to the value chain in terms of, there's IoT that goes on in the factory and use cases there, such as production monitoring or, let's say, assembly line optimization. And then there's the whole service space in terms of products out in the field or what does IoT look like out in the field, which is also kind of interesting and emerging when you start to think about remote monitoring or remote service.

Being able to get greater predictability into that can save massive oncost. And we touched a little bit on maintenance. That's been a big one in terms of the shift from preventative maintenance to condition-based and increasingly predictive and prescriptive.

Eric Hanselman

Well, it gets back to what you mentioned early on, which is that digital thread that runs through the entire manufacturing chain and how you manage it, how you look at it, how you integrate it. I will bring up the thing that Ian Hughes often cites in terms of being sort of the front end to the industrial IoT, being able to view that in the metaverse as the digital realm in which we actually look at all of this because we can't get through a podcast episode without mentioning metaverse.

David Immerman

Absolutely. The industrial metaverse has definitely began its long journey, and the way I have seen, a very interesting area that's starting to gain a lot of momentum. And I don't know if you want to call it metaverse or digital twin or simulation, but this idea of being able to look at what a manufacturing plant could look like.

So there's both a version of this where it's greenfield, so I have a new factory, and I can customize my layout that will ultimately generate the greatest production amounts, right? Then there's another version of it where it's what I have or it's brownfield where it's configured this way. Should it be somewhat different based on the simulation or AI model? Or maybe it's even, hey, I should consider automation in this part versus this other aspect I need a worker to inspect the quality of the product. So it's definitely an interesting tool that I think in terms of maximizing production capacity has a lot of potential.

Eric Hanselman

Well, we'll have to see where this goes as we head down the road. Well, Dave, this has been great. I appreciate having you on the podcast. There is clearly so much more to discuss around this in the various aspects of being able to pull this together. But thank you for all the insights.

David Immerman

Yes. Thank you, Eric. Thank you for having me. This is great conversation.

Eric Hanselman

And that is it for this episode of Next in Tech. Thanks to our audience for staying with us, and thanks to our production team, including Caroline Wright and Ethan Zimman on the Marketing and Events teams, and our agency partner who makes us sound great, does all of our back end, the 199.

Join us for our next episode where we're going to be talking about digital divide and some of the issues about really getting all of the benefits of digital capabilities out to the broader world and into different communities. I hope you'll join us then because there is always something Next in Tech.