AWS for Industries
Mini Series with Deloitte (Part 1 & 2)
Adopting smart factory concepts and components can bring great value to your business but understanding how to approach the technology upgrades, and what integration looks like for your facilities, can be a challenging endeavor.
In this podcast, we’ll highlight how smart factories, also known as smart manufacturing, smart refining, or digital operations, leverage rapidly advancing technology to create an impact in the manufacturing environment by connecting data from across the enterprise, systems, and supply chain.
Listen in as we chat with Jason Bergstrom, Principal Smart Factory Go-To-Market Leader, and Stephen Laaper, Principal, Smart Factory Leader, about available tips and tools, and some of the most common concerns about smart factory upgrades, that you might be facing yourself. Whether you’re wanting to strategize and road-map your smart factory upgrades, get the most value out of your proof of concepts, or scale your already connected smart factory, you won’t want to miss this.
You will also learn how Deloitte’s fully functional smart factory experience center provides a real-life example to help customers comprehend what a working smart factory looks like as you start to think about your modernization.
We’ll explore areas where technology has recently enabled growth and improved processes, the positive impact of delivering data to people at the right time, and how to account for the shifting skillset need in the manufacturing industry.
Listen now to Part 1: Evolution of the Smart Factory (Featuring Deloitte) Apple Podcasts, Spotify, Stitcher, TuneIn
Duration :00:23:58
Listen now to Part 2: The Art of Possible: Find Balance Between People and Automation Tools (Featuring Deloitte) Apple Podcasts, Spotify, Stitcher, TuneIn
Duration :00:19:31
On resilience:
“We’ve seen the shift from organizations that were looking at their manufacturing operations purely from a cost basis, to now actually adding variables around their ability to supply in the face of challenges. And that’s this notion of resilience.” – Jason
“Moving from a lot of fixed manufacturing capacity to variable capacity is really critical and is completely different today than it was even five years ago.” – Jason
On people:
“There’s another really important aspect, though, to automation, and that is the skill sets that are required in this automation, because even in facilities that are truly lights out, that doesn’t mean that they’re devoid of people.” – Stephen
On how to approach smart factory automation:
“It really is that sort of mantra of think big, start small, scale fast. We would encourage organizations to think about that because the last thing you want to do is find yourself doing a lot of “random acts of digital” around the organization that aren’t really connected.” – Stephen
On the Deloitte Smart Factory at Wichita State University
“When we built the facility, we intentionally chose to not do a lights out facility, Doug. And as part of that, what we did is a combination of what we call fully automated, semi-automated, and manual steps in order to be able to sort of highlight that human machine interface” – Jason
Featured Resources
AWS Delivers on the Promise of Industry 4.0 at the Smart Factory Wichita
The Smart Factory at Wichita
Ask A Question
Send us your questions at industrialpodcast@amazon.com. You can also post your question below in the comment section. We will reply to all questions within 1 business day.
What is AWS Industrial Insights?
Welcome to AWS Industrial Insights. In every episode, we interview visionary leaders from industrial companies to share their insights on technology, innovation, and leadership. This podcast is for industrial business leaders who are looking to make data-driven decisions and learn from those who’ve experienced similar challenges. By interviewing leading executives, we’ll uncover their insights and learn exactly how their organization found a solution. You can find all episodes of AWS Industrial Insights on your favorite streaming platform or listen below.
Episode 1 Transcript:
Jason
“Having lots of SKUs, being, you know, more open to the idea of kind of smaller asks of production environments rather than large high volume production runs, we’ve seen a complete shift in the technology landscape.”
Stephen
“…across different industries, across different sectors, at the end of the day, Smart Factory is really all about the combination of these rapidly advancing technologies combined with creating impact in the manufacturing environment.”
Caroline
Growing skills gap, increasing cyber threats, supply chain disruption. Do these sound familiar?
Doug
It’s a tough industry to be in and we’re here to help.
Caroline
I’m your host, Caroline.
Doug
And I’m your host, Doug.
Caroline
And you’re listening to AWS Industrial Insights, the podcast for manufacturing and industrial business leaders who aren’t afraid to think big.
Doug
We interview executives from well-known companies to share the disruptive ideas and topics like leadership, technology, and innovation.
Caroline
So let’s get started.
Thank you, Jason and Stephen, for joining us today on AWS Industrial Insights. We’re really excited to have you here. And to start off, you know, it would be great to kind of get to know you, Jason, could you introduce yourself and share with us your title, where you work and your current role?
Jason
Sure. Thanks for having us today, Caroline. I appreciate it. My name is Jason Bergstrom. I am a principal at Deloitte Consulting in our smart factory business. I’ve been serving the manufacturing space for the last 25 years, and I lead all of our efforts around go-to-market, which encompasses all of our sales, all of our marketing and our sales motion with our ecosystem sponsors.
Caroline
Awesome. Well, Jason, welcome to the show. Thank you so much.
Jason
Thank you. Thanks for having me.
Caroline
Yeah, of course. So now would also Stephen, can you please introduce yourself for us?
Stephen
Hi there. Great to be here today. My name is Stephen Laaper. I am the smart factory business leader for Deloitte, and we are in the business of helping organizations improve their manufacturing operations, not only by using some of the traditional techniques that they have come very accustomed to over the last two- or three-decades, things like Lean and Six Sigma, other, you know, managerial techniques.
But now also combining those with some pretty rapid advancements in technology. It’s an exciting time to be in the smart factory space.
Caroline
Awesome. Well, Doug and I are so excited to have you both here. I think you’re going to be some great subject matter experts for what we’re going to talk about today. And to jump right in, you know, we’re really focused today talking about the smart factory. And I know that AWS and Deloitte have a pretty unique relationship around this topic.
To get started, Stephen, can you tell me a little bit about like when we’re talking about what is a smart factory, what does that mean? And also, can you define like what is THE smart factory, that specific relationship with AWS?
Stephen
Yeah, absolutely. And maybe if we just start with that broader smart factory definition. It really goes by a number of different names. You know, we chose to call it Smart Factory, but many of the customers, many of the clients that we work with, they, you know, they have a slightly different vernacular that they put around that.
So that could be smart refining, it could be smart manufacturing, it could be digital operations, for example. While the name is, you know, perhaps a bit diverse across different industries, across different sectors, at the end of the day, Smart Factory is really all about the combination of these rapidly advancing technologies combined with creating impact in the manufacturing environment.
And what we’re finding increasingly is, you know, the impacts that are being created in the manufacturing environment really require data and information from all across the supply chain and across the enterprise. I think a really great example of that is when we start talking about the digital thread and how all of that data moves through the supply chain to have a positive impact on the operations environment.
Caroline
And quick question, when you say the digital thread, what do you mean by that? I kind of heard that term used a few different times in different contexts. Like can you kind of clarify how that applies here?
Stephen
The digital thread really is about building the connections between different parts of your organization in a way that allows information to be transmitted digitally and in as part of a common data model. When we think a really, you know, a really specific example of this might be digital product development and how the tools and the systems that a product engineer might use can speak directly to perhaps a manufacturing execution system.
So that way there’s a tight linkage between what’s happening in the design environment and what’s being executed in the manufacturing environment, but that’s one very specific example, but representative of where organizations are going with the digital thread.
Caroline
That’s interesting, and I’m curious too. Do you think that that has changed significantly versus like from ten years ago? Why should people care now?
Stephen
There’s a dramatic improvement in the connectivity of these different systems, connectivity of different manufacturing environments, connectivity across the organizations that are adopting these types of technologies. That interconnectivity in itself doesn’t create the value, but the creation of, of these connections between those different parts of the organization and these systems that they use is a really important impact.
One of the things that that we recognized is that sometimes these types of connections in these types of tools are hard to see. One of the requests that we often at Deloitte would get is, you know, could you take us to see a smart factory implementation? And of course, you know, we’ve got a lot of different opportunities to do that. During COVID that, of course, became a little bit more difficult to bring external parties into, you know, into their manufacturing environment, to be able to see some of these demonstrations.
And for a variety of reasons, we chose to embark on a smart factory of our own, and that smart factory is located in Wichita. It’s on the innovation campus of Wichita State University. And that is done in conjunction with an ecosystem of which AWS is a very important component. In that smart factory, we call it the Smart Factory at Wichita, it’s part of a global smart factory network, and inside that smart factory, we have built a fully functioning kind of manufacturing environment.
We make a pretty cool product. It’s actually a STEM education kit. The best part is it serves as a real product for us to demonstrate the power of the smart factory. And it also serves a greater purpose in our ability to then take those STEM education kits that we produce and actually donate them to middle schools in underserved communities across the country.
Doug
So, Stephen, or Jason, whoever wants to answer this next part is. We’ve actually built a concept using the concept of a smart factory to bring customers to touch, feel, and see. Is that correct?
Stephen
Yeah, that’s exactly it, Doug, and one of the you know, one of the one of the challenges with most manufacturing operations is that, you know, their mission, of course, is manufacturing. It’s usually not hosting guests that are there to see their smart factory in action. And when we built the Smart Factory in Wichita, it was very much done with both of those purposes in mind, that it’s a fully functioning manufacturing environment to see these you know, to see these different types of capabilities at work.
But it’s also an experience center where people can get their hands dirty. They can participate in smart manufacturing to really understand what these types of capabilities really look like in the manufacturing environment.
Doug
If we look back over the past ten years in the manufacturing environment, we’ve kind of been talking about the smart factory or a smart factory concept for a long time. Talked a little bit about this, but what changed and what are we bringing to life within THE [Wichita] Smart Factory to show that to the customers?
Jason
It’s a great question Doug. So, I think there’s a number of factors at play. The first being a number of macroeconomic issues that we’ve seen bubble up over the course of the last five years or so, dating back even before the pandemic. But certainly over that time, we’ve seen a lot of changes going on. Examples of these include things like clients and customers becoming much more focused on being, you know, having lots of SKUs, being, you know, more open to the idea of kind of smaller asks of production environments rather than large high volume production runs.
We’ve seen a complete shift in the technology landscape, although Stephen and I have been in the space for a long time and technology has evolved, it really hasn’t gotten to a place where it could fully sustain digital operations until just recently, and a lot of that was unlocked during the pandemic. And then on top of that, you have the pandemic effects as well.
The idea of being able to do things in a more automated fashion, we’re seeing labor challenges that I think most manufacturers are feeling, especially over the course of the last two or three years, that make it a little bit tougher to run operations. All of that combined with just the general realization that most organizations need to digitize their operations to compete, has really lit a fire within the marketplace.
And as a result, the shift to how do we really think differently about how we manufacture, leveraging these new technologies, solutions like those from AWS, is really the focus of almost all of our clients in the marketplace.
The smart factory at Wichita, as Stephen alluded to, is really a fantastic environment where we’re able to bring together a lot of different world class providers of those technologies and showcase them in an end to end fully functioning production line, which allows them to see and proves to them that it can be done, that you can really think differently about how you manufacture products from inputs coming into a warehouse to final products going into an AWS truck on the other side of the facility. It’s really a magical experience and one that we’re extremely proud of.
Doug
If I’m a CFO, a CEO, a CDO, keep going with all the different C’s, right? Just to make fun, CTO, keep going… Why would I want to go to the Smart Factory? I mean, I’ve got my factory out back. What am I going to see that is going to perk my interest and say, wow, this is completely different?
Jason
I think historically the interest and the focus has been with the COO or perhaps your most senior manufacturing leaders, and the reason for that is this is in their backyard, right? Improving things, making step change improvements, improvements to productivity, to production, output to capacity, those are all see COO goals and objectives. And what’s happened as organizations have learned how to address their end-to-end production systems and really move to digitizing from the front of the line to the back of the line, is they’ve now unlocked capabilities that are much more fundamental to how CEOs, CFOs, CTOs run the business.
As an example, and I alluded to this a moment ago, the idea of being more agile and more flexible in how you manufacture at a lower cost basis, could completely change what you sell in the marketplace. It could fundamentally change the markets you serve. And those are strategic questions that are impacted by digital operations that are of great interest to CEOs and CFOs.
And we’ve found in a number of studies that we’ve done that, you know, things like time to market and new product development, those are critical components to how you compete. And digital factories, smart manufacturing is able to make significant reductions and how you bring products to market around that digital thread that Stephen alluded to. We’ve seen reductions in those of 30 to 40% of cycle time, which is huge and complete game changers for organizations.
Doug
So now we’ve got a customer that’s gone through here to see, you know, we’ve got the board that’s there and they’re kind of seeing this. They walk out and they go home. What do we do next for them?
Jason
Doug, in terms of after visiting the facility?
Doug
Yeah, they’ve gone through the facility. They’re wowed. They love what we do. We’ve got concepts of digital thread, we’ve got, you know, additive manufacturing, we’ve got robotics, we’ve got some manual capabilities that are happening there, just like any manufacturing site. Now they go home and they’re kind of looking going, you guys were great. This was good. How do we operationalize that?
Jason
It’s a great question. Part of what we do with every visitor that we have through the facility is it’s more than just showcasing them a bunch of bespoke technologies. In fact, we spend quite a bit of time preparing each of those visitors to understand what their major pain points are within their manufacturing and supply chain organizations, such that when they come through the facility, we’re able to do more than just showcase tech, but rather speak to technology process and human capital and cyber concerns that are critical to their business issues.
And when you say they leave the factory, they don’t really ever leave. They physically leave. But we are continuing to work with them around those issues. And what they’ve seen as part of that visit then comes to life in post visit dialog where we apply some of the art of the possible that we shared to their business issues and help them to think differently.
And for each client, that could be a little bit different. From upfront strategy and visioning and road mapping to clients that have tried to do this but aren’t quite getting value out of their proof of concepts, to other clients that are pretty mature, but they can’t figure out how to scale.
Using some of this new tech in new applications are ways for us to drive those solutions across their networks. And so, there’s a natural connection point post visit. There’s actually as important, if not more important than the visit itself.
Caroline
That’s awesome. And I think, you know, it’s really good that you approach that on a customized basis, understand that customers are coming from different points of view with different challenges, and you have a solution for them. I think that’s really good.
And my next question is for Stephen here, kind of specifically on some of those challenges. I would imagine if I’m a customer coming into this amazing smart factory, I’d think, oh my gosh, this is like the utopia of industry 4.0, but I could never do this. I have problems. My infrastructure’s too old. I don’t have the talent for this.
Can you talk a little bit about like what are some of the common challenges or barriers that you hear customers bring up when they first come into this new factory environment? And how do they face those?
Stephen
It’s an incredibly relevant question, because the important thing to acknowledge is that most organizations and in our, you know, our experience, it’s actually less than 5% of organizations are actively investing in greenfield smart factory facilities. What that means is that for the remaining, you know 95%+, is that those organizations are seeking to adopt these types of capabilities which are going to result in the impacts that their shareholders are increasingly expecting from them as organizations.
They’re having to adopt them in the existing manufacturing footprint. And that’s really important, because now that means that organizations are faced with a number of challenges. Those challenges range from having the right talent in those existing locations and in the organization to be able to activate and truly gain the value and impact from these types of smart factory capabilities.
It’s inclusive of challenges with a very, very diverse set of manufacturing equipment, oftentimes from very different vintages. The underlying control systems, if they’re even present at all in in the equipment, may, may present challenges in terms of connectivity, the ability to have a common data model, and also thinking about connectivity, manufacturing facilities tend to not always be in the areas of the world with the most robust connectivity.
And how you think through those challenges really, really, really has a very significant impact on the time to value in these smart factory solutions.
Doug
Stephen, lots of great examples of their customers problem are seeing problems that the statements that we have you know too many machines, too many disparate things, maybe not even software in place. What’s different now, if we look back even two years ago to what we’re able to bring to these customers to help solve those problems?
Stephen
Yeah, and Jason mentioned it, that the real ability for organizations to operate in a truly digitized fashion, while many of the underlying components of that technology has been building up over time, it’s only really in the last few years that it’s come together in a way that that allows the full enablement. Many organizations taking a very explicit focus on the type of talent that is necessary to operate these types of solutions, not only operate, but to build those solutions.
One of the things that is fundamentally different with the workforce today is, is an expectation from the workforce of being able to operate in a more digital fashion. You know, through our consumer lives, we have all become very, very accustomed to our smartphones and having certain types of data available to us really at our fingertips, in a very convenient, organized way.
This is one of the hallmarks of these digital solutions that are being deployed as part of industry 4.0 and as part of Smart Factory. It’s what we like to call a persona-based approach or a persona-based solution. And in that particular case, this is really focused on getting the right type of data to the right individual at the right time in a way that they are best able to consume that information to make the next best decision in their role or in their job.
That’s a really, really, really important place for a lot of organizations to start with, is a distinct acknowledgment of who’s using the information and how is that going to help better make decisions. And then designing the tools around that skillset and capability. Underpinning that then, comes some pretty significant advancements in the ability to create common data models, to be able to apply algorithms to very diverse sets of inputs and signals to create a layer in which takes a lot of that complexity out.
And that’s a really important second feature of these types of solutions, is that even though the underlying machinery, the underlying control systems, the PLCs may be of all different vintages, and of different capability, that complexity layer is in place to really make that a more homogeneous type of type of solution for people to utilize.
And then the other thing that I would point to as distinctly different is our ability to interface with machines and equipment in ways that we previously were not able to. And this is a great example of where computer vision has come into play. I mentioned a little bit tongue in cheek earlier around the idea that, you know, maybe even some of these assets don’t even have control systems.
And in about 30% of the smart factory solutions that we deploy, that’s exactly the case. That either there is no previously existing control system or it’s too rudimentary to be able to interface with.
You know the smart factory there in Wichita, we joke with our guests a little bit, and we have a pretty common injection molding machine that many of our clients still use today. That has a, you know, early eighties vintage, and we kind of joke with them a little bit and ask them to try to find the Ethernet port on the injection molding machine. Of course, there is none, right?
So how we interface with that machine is really, really important. In this particular example, utilizing computer vision, via a video stream that’s being processed in a combination of both at the edge and in the cloud, to be able to ascertain the status of that machine is a very natural solution. But that’s something that’s really only become available to us, at scale, in the last the last two, three years.
Caroline
Awesome. It really sounds like technology is unlocking a lot of new opportunities for us.
Caroline
Thank you for tuning in to AWS Industrial Insights. If you want to learn more about today’s episode, head over to the blog for a list of featured resources on this topic. You can also find today’s blog in the episode description and also on our website aws.amazon.com/industrial/podcast.
Episode 2 Transcript:
Jason
To improve every day and then every week, then every month. And that is a much more engaging and fruitful career for a lot of people.
Stephen
That doesn’t mean that they’re devoid of people. In fact, in certain cases, it increases your reliance on an even more scarce skill set.
Caroline
Growing skills gap, increasing cyber threats, supply chain disruption. Do these sound familiar?
Doug
It’s a tough industry to be in and we’re here to help.
Caroline
I’m your host, Caroline.
Doug
And I’m your host, Doug.
Caroline
And you’re listening to AWS Industrial Insights, the podcast for manufacturing and industrial business leaders who aren’t afraid to think big.
Doug
We interview executives from well-known companies to share the disruptive ideas and topics like leadership, technology, and innovation.
Caroline
So let’s get started.
Caroline
All right. Well, welcome back to part two of speaking with Deloitte here. You know, last part, we kind of really dove into the smart factory, understanding what does that mean, The Wichita sample that you have for customers to visit, and understanding what are the challenges to building that and how customers are facing that now. If you were listening to the last episode, you probably heard the word “talent” or “people” come up quite often.
So today we’re really going to focus, dive in a little bit deeper to the future of work.
To get started, Jason, I want to ask you about the factory of the future. Things have changed a lot, especially during this pandemic. Can you talk a little bit about what is required now, today in the future or in the factory, and working at the factory, that’s different from 5 to 10 years ago?
Jason
Yeah, sure, and maybe what I’ll do is start at the macro level and drill down. We have seen a lot of shifts, some of which we spoke about in the first episode, but one that we didn’t speak about that’s of critical importance to manufacturers, is this notion of improving resilience. For those of you that are out there that haven’t been able to get something shipped to your house, or you’ve been to the grocery store and there’s not enough paper towels, this is a function of resilience.
We’ve seen the shift from organizations that were looking at their manufacturing operations purely from a cost basis, to now actually adding variables around their ability to supply in the face of challenges. And that’s this notion of resilience.
So with that, you’ve probably read numerous articles, organizations are starting to nearshore and onshore a lot more of their production that historically had really been shipped off to low-cost manufacturing countries. As a result of that, we’re starting to see more focus on “how can we be more efficient in higher cost environments?” Hence, the smart factory and smart manufacturing comes into the dialog.
At the same time, given all the technical advances that Stephen was alluding to, we’re starting to see another shift in focus associated with the individual organizations that we work with, which is, it used to be about how you optimize a single manufacturing site, and that’s how plant managers and operators were incentivized.
But now what organizations are starting to learn is that they’re actually able to make better choices if they look at their entire network, their entire global manufacturing network. And that’ll drive you to make different choices, different decisions as you’re manufacturing, what is a growing and more complex product base that the market is demanding.
And so having visibility into those networks, being able to see when things shift or demand centroids move or there’s changes in supply, you’re now able to look and optimize that network rather, than a single site. And that makes a huge difference in terms of how you operate.
And I think similarly, just connecting it back to the first episode, as we see more and more demand for agility, higher product counts, more complexity, requests to bring products to market faster, an organization’s ability to think about their manufacturing footprint as a network, becomes more and more critical.
And then within each one of those facilities, how you think about agility across your assets and your lines has a very similar impact. Moving from a lot of fixed manufacturing capacity to variable capacity is really critical and is completely different today than it was even five years ago.
Doug
Jason, to follow up on that, so lots of changes that need to happen from that standpoint, but we’ve been talking about for years that things like the silver tsunami, people retiring, people leaving this industry and not enough people coming into it. So how – think about what you were just talking about with all these changes and how is that going to impact the next people that we need to bring in?
Jason
If you think about the history of manufacturing, Doug, what you come to conclude pretty quickly is going back 20 or 30 years ago, up till today, manufacturing jobs were not the most attractive ones.
It was typically filled with resources that were considered low skill, a lot of repetitive tasks, employees were not empowered to help improve and make marked change on how their operations run. They were simply given a procedure and asked to execute it.
And as a result of that, combined with the fact that over that time we were moving so much manufacturing overseas, the skills really just sort of fell off the map. And what we are seeing, and what I think is very exciting about Smart Factory, is that as we shift more of that manufacturing back to the U.S. or near shore, what you’re starting to see is a shift in terms of the kinds of talent, the kinds of people, that manufacturers are looking to bring into their organizations.
And it is a different world. We talk about the idea of moving from manual tasks to automated tasks. Well, that’s great. However, you’re always going to have, for most organizations, a component of manual tasks within your processes. Very few organizations can truly run lights out.
And so, hiring and bringing in talent that allows you to showcase sort of the human machine interface, how humans work with automation, and the kinds of capabilities that you need to enable that are just fundamentally different. They’re resource sources that are higher skilled, that have more technical aptitude. You want to empower those resources.
In other words, as Stephen mentioned earlier, bring them the right information at the right time to help them make better decisions that allow them to improve. To improve every day and then every week, then every month. And that is a much more engaging and fruitful career for a lot of people than being told what to do and coming in and just executing your manual tasks for 8 hours.
You’re starting to see the shift in terms of the kind of resources that are coming into the operations, but then also the kinds of things they do, and the amount of autonomy that they have, to help make better decisions for the organization. What we are seeing is when you do that correctly, you’re actually now seeing more of net adds to talent within manufacturing rather than more of a scampering away from manufacturing, which historically has been what we’ve seen, especially in the US.
Doug
In episode one we talked about THE Smart Factory using Smart Factory concepts and everything. Here we’re talking about talent and capabilities from there. How is the Smart Factory going to help this area?
Jason
When we built the facility, we intentionally chose to not do a lights out facility, Doug. And as part of that, what we did is a combination of what we call fully automated, semi-automated, and manual steps in order to be able to sort of highlight that human machine interface I alluded to earlier. Right, how do you work in an environment where you are going to have operators being critical to the output, throughput, and pace that you can run that line?
And we’ve taken advantage of that at WSU, using students from the university that are participating on our line and working with us every day, to do a lot of the things I just spoke about, which is: “How do we improve our line? How do we think differently?”
How do you take that operator experience and make it more fruitful and impactful, such that it becomes a place where, students and seasoned professionals alike really want to they want to participate in it. It’s something that is just fundamentally new and different. And like I said, it’s an empowering role, and we’re seeing that within our factory, both with our undergraduate and graduate students working with us.
Caroline
Definitely. And Jason, one of the points you brought up about making sure that people have the data they need to make decisions, reminds me a lot of what Stephen said in part one, where technology can kind of unlock a lot of those opportunities. I can imagine that it would be really hard to find a balance between the amount of automation and the amount of human interface that’s necessary to achieve those goals.
So Stephen, can you talk a little bit about like how do you reach that balance and what are maybe some things to avoid?
Stephen
Yeah, it’s a phenomenal question. And it’s one that we often get asked, as some of our guests are kind of concluding their experience out there with us at the Smart Factory in Wichita. And the question usually will come about in a way where they’ll say, “Hey, does this mean that a lot of organizations are moving towards like lights out facilities? Or fully automated facilities?
And we’re able to really showcase, as Jason mentioned, a lot of the different types of capabilities that are necessary across that entire spectrum. And for most organizations, what’s going to be the right answer for them is usually somewhere in the middle, and that’s being driven by a couple of really distinct things.
On one hand, you’ve got this concept of resiliency which Jason brought up, and that’s really tied to getting the right data to people that are helping make better decisions. Now, that can be done with perhaps some augmentation, some tools, right? So, one might call that a form of automation of our data.
But in this particular case, you know, the resiliency could be driven by a machine learning model that is powering a recommendation engine, that is alerting that particular operator to a condition, and suggesting, “Hey, actually, there’s three choices you can make. And you let me know which one of these choices you’d like to put into place,” and the system will enact that change. That’s an example that occurs at the operator level.
There may be another example that occurs as a supply chain planner is looking at demand and production outputs from manufacturing plants and perhaps some unforeseen weather event. And they’re being given recommendations by a machine learning model that presents that person with, “Hey, here’s three options. Choose the one that you think best suits this need.”
So we like to call that person in the loop control, and for any automation engineers out there listening today, they’ll recognize the importance of kind of what has historically been referred to as “man in the loop control” where you’ve got an automated process but it’s in service to a…
Caroline
Person!
Stephen
That’s exactly, yeah, exactly right and…
Caroline
Exactly.
Stephen
We’re going to increasingly see those types of augmentation type solutions that allow people to better make sense of what is a massively proliferating set of data that exists in their manufacturing environment. When you start combining that with the broader supply chain, we start combining that with the broader enterprise, it really starts to exceed an individual or even a team’s ability to rapidly distill all that information. And that’s where you need some of these advanced technologies.
There’s another really important aspect, though, to automation, and that is the skill sets that are required in this automation, because even in facilities that are truly lights out, that doesn’t mean that they’re devoid of people.
In fact, in certain cases, it increases your reliance on an even more scarce skill set. And that’s around a lot of your engineering staff, around a lot of your automation specialists of skilled trades in a way that isn’t always the right answer. So that brings us back to this important, you know, notion of balance really being somewhere in the middle for most organizations.
Doug
Hey, Stephen, you mention one thing, and of course, it’s pretty dear to our hearts at AWS, and for you guys as a consulting company as well, is: the technology, advanced technology, what are two areas that technology has unlocked that wasn’t available five years ago?
Stephen
Yeah, I would say that those, those really fall in into 2 areas that would go something like, our ability to execute solutions at scale, and then also our ability to implement the solutions in a time scale that is more appropriate for creating that impact in the organization. So let me give you an example of what that might be.
A lot of the actual math, a lot of the actual algorithms that are in use in many kinds of machine learning or recommendation engines, those aren’t necessary net new algorithms. In fact, many of them were developed in the forties, fifties, and sixties. What we now have is the ability to run those types of algorithms in, not only a more expeditious way, but in a more accessible way.
So scale is really enabled by the raw horsepower. That’s an underlying piece of technology that’s really important. But then also pretty rapid advancements in the technology that sits on top of on that core.
And that is, pieces of technology that don’t necessarily require a mathematician or a data scientist, but now through a variety of kind of low code and no code solutions, allow people to access that raw horsepower in a way that is a little bit more suited towards individuals that you’re going to find in the manufacturing environment. So really scale and accessibility, Doug, is the two core areas that have helped unlock this power.
Doug
Awesome. Very good. Thanks for that.
Caroline
And as we wrap up today’s conversation, Jason, I want to give you an opportunity to have a last word as well here, talking about kind of people who are listening to this podcast. What should they take away from this episode about the future of work? And what do you think that they should do differently?
Jason
Yeah, Caroline I think that there’s just there’s no doubt, based on our research, that organizations are going to go on this journey to digitize their operations. My advice would be there are so many different ways to approach doing that, don’t find yourself getting paralyzed. And we do see clients, given there are so many avenues, there’s so much tech, there’s so much opportunity in this space, that they chase the shiny object and they find themselves spinning their wheels.
We have a saying here at Deloitte, this notion of think big, start small, and scale fast. What we mean by that is start your journey by setting really lofty ambitions. Even if those ambitions may take ten or 15 years to get there, it’s still important to have a true north such that every decision you make as an organization really pivots around your ability to over time to get to that point.
And then the notion of starting small is, it’s all about ROI, right? You can’t afford to go spend five, six, seven years proving out solutions. You need to be able to do it much faster using agile techniques, using sprints. So, pick some spots where you can make a difference, where you can show an impact, where you can show return on investment and place some bets there. Then use those victories and those learnings as an opportunity to really springboard yourself into scale.
And once you’ve proven something within part of your network, being able to then take that and expand it out to your broader network gives you obviously a multiplier effect in terms of return. Doing that quickly is important once you’ve proven something out.
It really is that sort of mantra of think big start, small, scale fast. We would encourage organizations to think about that because the last thing you want to do is find yourself doing a lot of what we call: random acts of digital, around the organization that aren’t really connected.
You can’t drive towards a common vision. You find those projects over time fizzle out and they become poor investments, and so don’t get bogged down in that. Really look for opportunities to get going, get started, and show some victories that will move the needle.
Caroline
Thank you for tuning in to AWS Industrial Insights. If you want to learn more about today’s episode, head over to the blog for a list of featured resources on this topic. You can also find today’s blog in the episode description, and also on our website at aws.amazon.com/industrial/podcast.